This Program Is Brought to You By...: Distributing Television News Online 9780300216240

Journalism, television, cable, and online media are all evolving rapidly. At the nexus of these volatile industries is a

210 30 3MB

English Pages 336 [329] Year 2015

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

This Program Is Brought to You By...: Distributing Television News Online
 9780300216240

Citation preview

THIS PROGRAM IS BROUGHT TO YOU BY...

This page intentionally left blank

THIS PROGRAM IS BROUGHT TO YOU BY... distributing television news online

joshua a. braun

Published with assistance from the foundation established in memory of Philip Hamilton McMillan of the Class of 1894, Yale College. Copyright © 2015 by Joshua A. Braun. All rights reserved. This book may not be reproduced, in whole or in part, including illustrations, in any form (beyond that copying permitted by Sections 107 and 108 of the U.S. Copyright Law and except by reviewers for the public press), without written permission from the publishers. Yale University Press books may be purchased in quantity for educational, business, or promotional use. For information, please e-mail [email protected] (U.S. office) or [email protected] (U.K. office). Designed by Mary Valencia. Set in Minion and ITC Franklin Gothic types by Integrated Publishing Solutions. Printed in the United States of America. ISBN: 978-0-300-19750-1 (paperback : alk. paper) Library of Congress Control Number: 2015935292 A catalogue record for this book is available from the British Library. This paper meets the requirements of ANSI/NISO Z39.48–1992 (Permanence of Paper). 10 9 8 7 6 5 4 3 2 1

For Alex Braun May the news in your life always be good— however it reaches you. And for Sarah Braun With love and appreciation for all the wonderful things that have found their way into my life through you.

This page intentionally left blank

CONTENTS

ACKNOWLEDGMENTS ix

CREDITS xii

CHAPTER 1 Introduction 1

CHAPTER 2 Why Media Distribution Matters/ A Critical Interlude

19

part one

TRACING THE PATH: HOW NEWS VIDEO REACHES ONLINE VIEWERS



CHAPTER 3 Mapping Sociotechnical Systems That Deliver Media

39



CHAPTER 4 To Turn It On and Have It Play/ The MSNBC.com Video Player

55



CHAPTER 5 Clip & Share: Engineering for the Conversation Economy

76

part two

OUT OF ONE, MANY CHAPTER 6 Breaking Apart the Monolith/Considering Contemporary Media Organizations 105

contents  viii



CHAPTER 7 Friends of Mine, Friends of Theirs/ The Origins of Newsvine 122



CHAPTER 8 Different Things to Different People/ Newsvine and MSNBC.com 146



CHAPTER 9 If We’re Not Online, We Don’t Exist/The Rachel Maddow Show Builds a Web Presence 170



CHAPTER 10 The Maddow Blog Is Born 188

part three

MASS REACH AND MASS MEDIA



CHAPTER 11 One Size Does Not Fit All/Heterogeneous Audiences and Proliferating Brands 211



CHAPTER 12 One Size Does Fit All/Engineering Flexible Architectures 222



CHAPTER 13 Mass Reach with and without Mass Media 231

CHAPTER 14 Conclusion 247

Epilogue 258

NOTES 263



BIBLIOGRAPHY 279



INDEX 303

ACKNOWLEDGMENTS

As many before me have attested, a book is the product of myriad people’s work, inspiration, and generosity behind the scenes. For their help with this book, I wish first to acknowledge my family. My parents, June and David Braun, my aunts, Susan Ponville and Cathy Colloff, and my uncles, D. J. Ponville and Lauren Colloff, provided unwavering love and support that allowed me to pursue and attain my Ph.D. My in-laws, Greg and Mary Arnold, gave selflessly, providing tremendous help on the home front on a daily basis, making it possible to see this work through to publication. And most especially, I wish to thank my wife, Sarah Braun. Her encouragement, assistance, and understanding from the primordial beginnings of this project through to its release as a book made all of this possible. This project began life as a doctoral dissertation, and I wish to thank my adviser, Tarleton Gillespie, and my committee members, Stephen Hilgart­ner, Bruce Lewenstein, and Dan Cosley, for their incredible generosity, their invaluable knowledge and guidance, and their untiring advocacy on my behalf throughout the graduate process. Tarleton, in particular, was— and is—an unfailingly excellent mentor, patron, and friend. Both the prose and the ideas in this manuscript also owe a great deal to the herculean efforts of, and refinements made by, two wonderful reading groups in which I’ve participated over the years, the first being the Cornell New Media & Society reading group, spearheaded by Tarleton and by Lee Humphreys, and the second being the Media and Digital Culture reading group: Mike Ananny, Andrea Hickerson, Aynne Kokas, Seth Lewis, Matt Powers, and Jen Schradie. Other colleagues who have been indispensable in their comments on this work over the years have included C. W. Anderson, Juliette De Maeyer, Amy Gonzales, Megan Halpern, Tony Liao, Elizabeth Newbury, Oya Rieger, and most especially Dmitry Epstein and Daniel Kreiss. I greatly appreciate the help of Yale editor Joseph Calamia, as well, in steering this book to press. My former colleagues at Quinnipiac University, where much of this manuscript was written, also provided me with considerable assistance in ix

Acknowledgments  x

the form of summer funding for research and writing. My dean there, Lee Kamlet, my department chair, Liam O’Brien, and the director of the Interactive Media graduate program, Phillip Simon, were incredibly cooperative in helping me to juggle my course load and service responsibilities at critical points in the preparation of this manuscript. The discretionary conference funding provided me by Lee Kamlet also created important opportunities to attend events and have conversations with individuals whose insights are reflected throughout this book. The ideas in this manuscript were also shaped by my participation in, and the support of, a number of other terrific groups of scholars, including the Culture Digitally workshop series on cultural production in the digital age convened by Hector Postigo and Tarleton Gillespie and funded by the National Science Foundation. I was incredibly lucky to have had the opportunity to participate in this collective as part of my graduate and junior faculty experience. My conversations with members of this group extend back to the beginnings of this project as a dissertation and were formative in many ways, both to this book and to my sense of identity as a scholar. I am especially grateful to the organizers for involving me, and to Nick Couldry and Fred Turner for their input on the early concept for this manuscript. I am also indebted to the organizers and participants of the Connected Viewing Initiative (CVI) summit series, sponsored by UC Santa Barbara’s Media Industries Project in conjunction with Warner Brothers. I am eternally grateful to Michael Curtin, Jennifer Holt, Richard Hutton, Elissa Nelson, and Kevin Sanson, for unconditionally supporting the work of an obscure, recently minted Ph.D. and inviting me into conversation with such a bright and venerable group of scholars. This manuscript was dramatically broadened by my discussions with the CVI organizers, as well as other participants in the summit series, including Rachel Allen, Max Dawson, Elizabeth Evans, Jarice Hanson, Amanda Lotz, Paul McDonald, William Moner, Matthew Thomas Payne, Gregory Steirer, Sharon Strover, Ethan Tussey, Chuck Tryon, John Vanderhoef, and Patrick Vonderau. Reaching further back, my undergraduate mentors—Lisa Parks, Omer Blaes, Regina Fletcher, Janet Walker, and David Kohl at UC Santa Barbara —as well as my first thesis adviser, Charles Bosk at the University of Pennsylvania, all deserve a great measure of credit for teaching me both the possibility and the value of studying topics with a tendency to elude the standard curriculum. I also appreciate the wonderful moral support of

Acknowledgments  xi

Matt Carlson, Aymar Jean Christian, Gina Eosco, Gina Neff, Rasmus Kleis Nielsen, Stephen Reese, Laura Rickard, Tom Streeter, and Nikki Usher. Additionally, my students, in my time at both Cornell and Quinnipiac, have given me enormous moral support and food for thought, for which I am immensely thankful. I also wish to thank Cornell’s Graduate School; much of the research in this book was originally made possible by a research grant they provided. Finally, I know it is somewhat unconventional to thank one’s sources in an academic book, but truly this volume would never have been possible without the tremendous assistance and leap of faith provided by all the people I spoke with at Newsvine, MSNBC.com, MSNBC TV, NBC News, and other online and on-air news organizations. The individuals I talked with vary widely in their desire to be identified, so I won’t list their names here. But to each of you who invited me into your organization; who made arrangements for me to come; who made introductions to colleagues; who shared not just your valuable time, but also your ideas, opinions, and physical workspace; who read and commented on my work; and who volunteered your help in securing the myriad permissions necessary to bring this book to press: thank you. This volume, and indeed my entire line of work, would not be possible without you, and I am forever grateful for what you have done.

CREDITS

“Hosting the Public Discourse, Hosting the Public: When Online News and Social Media Converge” by Joshua Braun and Tarleton Gillespie. Copyright 2011. From the journal Journalism Practice. Reproduced by permission of the co-author and of Taylor and Francis Group, LLC, a division of Informa plc. Available at http://www.tandfonline.com/doi/full/10.1080/17512786.2011 .557560. “News Programs: Designing MSNBC.com’s Online Interfaces.” Copyright 2014. From the journal Journalism. Reproduced by permission of SAGE Publications Ltd. Available at http://jou .sagepub.com/content/early/2014/08/28/1464884914545730. “Transparent Intermediaries: Building the Infrastructures of Connected Viewing” by Joshua Braun. Copyright 2013. From Connected Viewing: Selling, Streaming, & Sharing Media in the Digital Age, edited by Jennifer Holt and Kevin Sanson. Reproduced by permission of Taylor and Francis Group, LLC, a division of Informa plc. “Going Over the Top: Online Television Distribution as Sociotechnical System” by Joshua Braun. Copyright 2013 by the International Communication Association. From Communication, Culture & Critique, Volume 6, Issue 3, pages 432–58, September 2013. Reproduced by permission of Wiley, a division of John Wiley & Sons, Inc. Available at http://onlinelibrary .wiley.com/doi/10.1111/cccr.12015/abstract. “Facebook’s Algorithm —Why Our Assumptions Are Wrong, and Our Concerns Are Right” by Tarleton Gillespie. Copyright 2014. From Culture Digitally. Reproduced by permission of the author. Available at http://culturedigitally.org/2014/07/facebooks-algorithm-why-our -assumptions-are-wrong-and-our-concerns-are-right/.

1//introduction

This book is about how infrastructures and strategies for distributing television news online are forged. The idea for a book about distribution of TV news emerged a few days before Thanksgiving in 2009, when I paid my first visit as a researcher to 30 Rockefeller Center, Manhattan’s towering General Electric building that is home to the headquarters of NBC News and its cable cousin MSNBC. I was greeted by Will Femia, who had been an online producer there for as long as many people could remember. During our conversation, he reminisced: “One of the things that was most shocking to me getting into the media business was the realization that regular people were making it,” he said. “Television to me—prior to working in television—was just like sunlight. You push the button and it just comes off the TV screen.” Femia’s quote nicely joined two concerns of media sociology, one well rehearsed and another—I soon realized—far less so. The first, well-­ trodden concern has been to challenge our often-simplistic assumptions about how media is produced. Indeed, one of the great contributions of media sociology since the 1950s has been to unearth a sense of contingency in the production of news. This is most readily apparent in discussions of the gatekeeping function of the news media. Both scholars and practitioners examining gatekeeping have sought to explain why some issues and events become newsworthy while others remain obscure. Answers have been offered up in the form of classic newsroom ethnographies like “The ‘Gate-Keeper’” (White 1950), Making News (Tuchman 1978) and Deciding What’s News (Gans 1980); critical studies of news content such as Stuart 1

Introduction 2

Hall’s (1973) “The Determination of News Photographs”; and innumerable lists of news values in the tradition of Johan Galtung and Mari Ruge’s (1965) “The Structure of Foreign News.” This body of research ultimately dispensed with what Herbert Gans (1980) called “mirror theory”—the naïve assumption, if it ever existed, that news products represent complete, veridical accounts of reality. Instead, these scholars argued that news is a constructed cultural artifact—a picture of reality resulting from situated values and arrangements of resources. No sooner had it been established that the content of the news media is neither unequivocally “the way it is,” nor “all the news that’s fit to print,” than the attention of sociologists began quickly to encompass the implications of these findings for social movements and societal change. At first this project largely demonstrated the manner in which social movements had been marginalized. Gaye Tuchman (1978), for instance, documented the various ways in which the women’s movement was ignored, then maligned and ridiculed by the press, before it ultimately managed to establish itself as a legitimate voice in the mainstream media. In his own take on the news media’s framing practices, Todd Gitlin (2003) famously implicated the mass media as a factor in the eventual dissolution of the 1960s student movement, detailing the ways in which Students for a Democratic Society ultimately lost control over their image to the news media. But as Tuchman and Gitlin both pointed out, despite the potential pitfalls that come with coverage, favorable attention from the news media greatly benefits social movements, and many sociological accounts have focused on how media-savvy interest groups adopt “dramaturgical styles of activism” (Epstein 1996, p. 220), carefully packaging their claims in ways that are likely to receive (positive) media attention (Best 1990; Gamson and Modigliani 1989; Hilgartner and Bosk 1988; Ryan 1991). In short, in the sixty years since Kurt Lewin (1951) and David Manning White (1950) first deemed the news media to be gatekeepers, media sociologists and other scholars of communication have developed a sophisticated language for discussing the various ways in which news organizations form a bottleneck in the public discourse, selectively limiting audiences’ access to information, while interest groups tailor their messages to the mass media’s whims and attempt to game its various selection mechanisms to bring their concerns to the public. Without questioning the validity or extraordinary depth of scholarship that has gone into developing this lens on the news media, I do wish to argue that some of the most

Introduction 3

interesting questions surrounding today’s news media lie outside its center of balance. This brings me to the second concern raised in Femia’s description of television content coming “off the screen like sunlight”—the one that will be the primary purview of this book. Online media, like television and other electronic media before them, feel immediate and are often defined by their potential for “liveness” (Couldry 2008; Gans 1980; Bivens 2014). This sense of liveness, or “immediacy,” to use Gans’s term, is carefully cultivated and an important part of news media’s presentation of itself as a form of direct access to world events. Unlike physical newspapers, and their attendant icons of the paper boy and delivery truck, however, we have little intuitive sense of the route that online news takes to get to us. When we can push a button and watch it come off the screen, it is all too easy to forget that online news is distributed—that systems of labor, infrastructures, institutions, economics, and numerous stakeholders are all involved in the route it takes to that screen (Downey 2001; Perren 2010). Indeed, just what sort of screen news is delivered to—whether a television, computer, mobile phone, or tablet—is a matter of huge concern, and increasingly contention, among media producers and distributors (Chamberlain 2010; Perren 2010). Both James Carey (1989) and Richard John (1995) remind us that we as a culture once associated the spread of information with the physical movement of messengers, before widespread adoption of electronic communication technologies, and the rhetoric of “eclipsing time and transcending space” that accompanied them, ultimately allowed us to largely divorce our notions of “transportation” and “communication” (Carey 1989, p. 14). Even as information now follows complex and shifting communication networks, these systems are regularly rendered invisible to the news consumer. In other words, the transport of news and information has been considerably “black-boxed” (Latour 1987). But even in the so-called information age, when communication seems so immediate, information must travel a route to reach us—and tracing that path brings a host of sociologically interesting relationships into focus. THE CONVERSATION ECONOMY

In his 2005 book The Search, John Battelle—a founding editor of Wired and former CEO of the defunct Silicon Valley trade journal The Industry Standard—offered up a concept that has since become known among

Introduction 4

the digerati as the “conversation economy.” More and more, he noted, web users seemed to be accessing news stories not by clicking through the homepages of news outlets, but by way of search engines and their associated news portals, or by following links shared by friends and acquaintances in various forms of online conversation. He argued that such trends put publications like the Wall Street Journal and the Economist at a disadvantage. These sites, which were built on a subscription model, put their stories behind paywalls, which in turn restricted the ability of search engines to index them, and of friends to share content with nonsubscribers. Moreover, the two practices—search and sharing—tend to work syner­ gistically to direct traffic online. The more often the link to a story is shared in blogs, discussion forums, and on social networks, the higher it will rise in the results of search engines that rank pages partly by counting backlinks (Halavais 2009; Introna and Nissenbaum 2000), and the more easily it will be found and shared in the future. Battelle predicted that as search and sharing continued to become more prominent methods of access to news, the influence—and subsequently the readership and r­evenue— of paywalled publications would begin declining apace, while their relatively open counterparts reaped the benefits. While the paywall debate isn’t over yet for many news sites, and much of the language surrounding the issue has shifted to other metaphors, like “metering,” some aspects of Battelle’s predictions have proved prescient. In particular, a great deal of traffic to news sites today is directed by sharing and search (Dwyer 2010; Riley and Usher 2010). In the first quarter of 2011, MSNBC.com reported that around 10 percent of its video traffic came from clips embedded on other sites, while ABCNews.com divulged that 70 percent of its video views were directed through links and search (Plesser 2011b, 2011c). Around the same time, the Pew Internet and American Life Project (Purcell et al. 2010) confirmed that link sharing had become an exceedingly common way of spreading and encountering news, reporting that “75% of online news consumers say they get news forwarded through email or posts on social networking sites and 52% say they share links to news with others via those means. 51% of social networking site (e.g. Facebook) users who are also online news consumers say that on a typical day they get news items from people they follow. Another 23% of this cohort follow news organizations or individual journalists on social networking sites” (p. 4). Similarly, by 2009 Facebook had already surpassed Google News in the number of users it directed to news media websites, and was soon generating over twice as many clickthroughs to news sites (Hopkins 2010).

Introduction 5

These substantial shifts have not gone unnoticed or unaddressed by journalists and media organizations. There is both anecdotal (Foremski 2010) and ethnographic (Riley and Usher 2010) evidence to the effect that online journalists have begun to select story topics partially on the basis of whether an article is likely to generate page views through sharing and search. And companies like Demand Media, at one point predicted by some to become the most lucrative technology startup since Google (Kerner 2010), have begun commissioning journalists to write stories on subjects suggested by algorithms that comb through search engine queries in pursuit of trending topics likely to generate the most page views (C. Anderson 2011). Indeed, metrics like the page view have fast become a major force in the editorial cultures of many online publications (C. Anderson 2011, 2013; MacGregor 2007; Outing 2005; Riley and Usher 2010). The idea that sharing drives traffic has become such conventional wisdom that it has led media economists like Tim Dwyer (2010) to remark that “on the web, news organizations are focusing somewhat less on bringing audiences in and more on pushing content out” (p. 50). Simply put, if the problem media sociologists have traditionally considered has been something like “How do nonjournalists attempt to get their messages distributed by journalists?” we appear to have entered a phase in which much of news work revolves around a somewhat different concern: How do journalists get nonjournalists to distribute their ­messages? Indeed, it’s tempting to look at all the Facebook “Like” buttons, embeddable video players, social media accounts, and RSS feeds cranked out by television providers and imagine that media organizations are doing everything they can to spread their content. “Spreadability” (Jenkins et al. 2013) is indeed important, but such affordances are also only one side of a wider ledger, with the other side being about constraints. Media companies are also using digital rights management (DRM) software to put limits on how far their content can spread; geofencing (which blocks international IPs from accessing content) to enforce national boundaries on the web; authentication tools to make sure that online viewers are offline cable subscribers; domain-blocking measures to ensure that their content isn’t embedded on pornographic sites or used to build unauthorized mirrors; IP-blocking tools to keep particular users or institutions from accessing or sharing content, and user-agent banning techniques to control what devices people use to access content. And all these technological means are intertwined with legal, economic, and social engineering (Gillespie 2007; Tryon 2013).

Introduction 6

In other words, the questions surrounding online media distribution —concerning who should have access to content, for what purpose, and on whose terms—serve to underscore Raymond Williams’s (1975) classic point about television: that “the effect of a technology is in fact a social complex of a . . . central kind” (p. 31). But this observation, which underlies the increasing scholarly interest in “the politics of platforms” (Ananny 2013; Gillespie 2010), has been made far more often about media production and consumption than about distribution; distribution itself has remained a surprisingly understudied topic. THE DISTRIBUTION GAP

Sitting on my desk right now is a volume with the imposing title News: A Reader. Edited by venerable journalism studies scholar Howard Tumber, it represented the state of the art in journalism studies upon its publication in 1999. It is divided into the field’s major topics of study, “Definitions of News,” “Production of News,” “Economics of News,” “Sources of News,” and “Objectivity and Ideology of News.” The 2011 edition of Michael Schudson’s book The Sociology of News carves up the terrain similarly, though it also includes a sizable section on audience effects. Generally absent from such compendia and the literature they represent is any discussion of news distribution practices. Not, anyway, the sort of practices we think of when we consider the canonical ethnographies written over the years on news organizations, in which we learn the names—or at least the pseudonyms—of reporters, copy chiefs, and managing editors, and the effects of their decisions on media products are mapped in rich detail. Rather, media sociology has tended to skip from production practices to audience effects, leaving a gulf when it comes to the actual process of getting news products out into the world. The perigee here may come in the form of research on the economics and regulation of media ­industries—but to study these areas is not the same as to examine the mechanics of distribution decisions made in these industries or the lived experience of distributing the news. In many ways, it’s understandable that such issues would lie at the periphery of media sociologists’ focus. After all, the print distribution routes and networks of broadcast affiliates that underpinned the activities of leading news organizations appeared relatively stable for much of the twentieth century. Occasionally a pallet of newspapers would fall off a ferry and sink into the ocean (Hetherington 1985), or a group of broadcast affiliates

Introduction 7

might refuse to air a controversial television news special (Carter 2004), but on balance the process of getting news from the printing press and the broadcast studio to the consuming audience was stable and standardized enough to go unnoticed. At the very least, it was sufficiently divorced from the day-to-day activities of journalists to warrant its exclusion from a sociology aimed at understanding their professional activities. That’s not true anymore. With the massive shift to online news that has occurred over the past two decades (Pew Project 2011) the process of delivering content to audiences has become an integral part of news work, from search engine optimization of headlines and stories (Riley and Usher 2010) to the maintenance of social media accounts for sharing links and headlines. Media researchers, however, while acknowledging these shifts, have mainly continued to ponder them in terms of traditional categories. Scholars have asked questions about how news organizations’ entry into an online information ecosystem might affect news production (Boczkowski 2004, 2010; Lowrey 2006; Mitchelstein and Boczkowski 2009; Paterson and Domingo 2008; Ryfe 2012; Singer 2010), as well as about whether the customization provided by Internet technologies might change news consumption habits and audience effects (Chaffee and Metzger 2001; Mitchelstein and Boczkowski 2010; Negroponte 1995; Sunstein 2007). Insofar as it’s been tethered to these traditional foci, research on news has somewhat awkwardly recognized the dramatic influence of new distribution channels on traditional industries without focusing on distribution itself.1 Something similar is going on in television studies, where critical theorists have developed a sophisticated language for discussing media texts, as well as audiences and active spectatorship. A healthy literature on production studies has also emerged among television scholars (Mayer, Banks, and Caldwell 2009), as have discussions of media law and economics (Streeter 1996). A small but growing handful of television researchers, however, including Amanda Lotz (2007), Brian Larkin (2008), Michael Newman (2012), Lisa Parks (2012a, 2012b), Alisa Perren (2013), and Chuck Tryon (2013), have begun to examine the world of media distribution in its own right.2 While these authors remain in the minority, they understand something uniquely important about distribution that I hope to bring to the fore in this book: decisions about distribution, whether made by media executives or file sharers, are—in the barest terms—attempts to control who has access to information and culture, and under what conditions.

Introduction 8

DISTRIBUTION ROUTES AS SOCIOTECHNICAL SYSTEMS

While television and journalism studies have so often remained production-focused, scholars in the field of science and technology studies have developed sophisticated lenses for examining how products are made to move. Historians and sociologists of sociotechnical systems have looked at, among other things, the construction of power distribution networks (Hughes 1979, 1983), shipping routes (Law 1986, 1987), railway systems (Law and Mol 2002), public transit lines (Bugos 2000), freeway systems, and digital packet routing (Hughes 1998).3 If, as I have posited, a major reason that media distribution systems have remained understudied is that during much of the twentieth century they were stable enough to fade into the background, another way of putting this is that these systems had become infrastructure. And as Star and Bowker (2002) point out, a primary feature of infrastructure is that it tends to remain invisible until it is disrupted, making it stubbornly difficult to study. This is another reason for my turn to science and technology studies, where infrastructure and its potential to influence human activity have long been a subject of fascination. A particularly useful framework put forward by systems scholar John Law is that of heterogeneous engineering. Law (1987) initially uses as an example the volta, a portion of the Portuguese trade route to India that required centuries’ worth of economic, legal, social, scientific, and technical engineering. The construction of the volta required the enrollment of numerous recalcitrant agents, both human and technical, that simultaneously served as resources and shaping forces for the system builders—in this case, merchants seeking to create a working trade route. The winds and currents that helped to carry ships to their destination immensely complicated their return. The celestial charts and other tools created in response by astronomers enabled creative navigational solutions to these problems, but were illegible to semi-literate sailors, the training of whom put further stresses on the system. When finally all the pieces were in place, their influence, writes Law, was quite literally inscribed on the map: The volta can thus be seen as a geographical expression of a struggle between heterogeneous bits and pieces assembled by the Portuguese system builders and their adversaries, that is, the winds, the currents, and the capes. It traces on a map the solution available to the Portuguese. It depicts what the Portuguese were able to impose on the dis-

Introduction 9

sociating forces of the ocean with the forces they had available. (Law 1987, p. 120) The Portuguese were interested in acquiring and distributing trade goods, not television news, but there is a universal point here. Again, it’s easy to forget that online media products are distributed. But while there are differences between information and physical objects (Benkler 2006; Child and McGrath 2001; Hess and Ostrom 2003)—between the Portuguese merchants’ desire to accumulate goods and the need of contemporary media workers to push information across services and devices—both trade goods and television programs travel a route to get from one place to another. In both cases the path they take reflects the interests and limitations of myriad heterogeneous systems and actors. Portuguese ships traveling the volta threaded a needle between an inhospitable desert coast on one side and unfavorable winds and currents on the other. And their route was equally influenced and displaced by the interests and limited malleability of sailors and royalty, navigational astronomers and shipwrights, so that at each turn they took advantage of opportunities these agents provided while simultaneously slipping skillfully between the limits imposed by all of them. I want to think of online distribution in much the same way, as an “inscription on the map of the solutions available” to content providers. In this way I hope to illuminate how the route that video and other information takes to our screens is at once the result of heterogeneous resources (at times precariously) lashed together, and the threading of needles—weaving between the limits imposed by all these resources and those enrolled in competing systems. In short, what Law suggests about the volta I find utterly engrossing and intriguing for the study of media: that a sort of archeological interest in the various kinks, epicycles, and roundabouts found in a distribution route can uncover the leverage of powerful actors and the mark of material agency, that it can expose sociotechnical systems at work and lay bare the influence of infrastructure. In a day and age when broadband access to the Internet has reached a majority of Americans, and when cloud services and peer-to-peer technologies have the potential to create instantaneous and remarkably efficient flows of information, such roundabouts turn out to be myriad. Newscasts that air live over cable channels and traditional broadcast networks are often time delayed, paywalled, or recut when they

Introduction 10

appear online; at times they crop up in unexpected places and sometimes they don’t appear at all. As video content wends its way to us online, it also goes through intermediaries that most viewers have never heard of. Tremor Video, for example, is a company that—among many other things—converts streaming video from numerous providers, ranging from Disney to CBS News, into a plethora of special formats tailored to our ever-growing menagerie of mobile devices, and then packages advertising with it on its way to the consumer. YuMe is another company with major industry clients. It scans the blogs, homepages, and other sites on which users place embeddable videos and determines whether a page is “brand-safe” (that is, features no objectionable content) before displaying paid ads with a clip. Yet another firm, Nexidia, produces software that makes audio recordings searchable by keyword. Its clients include intelligence-gathering operations interested in ­data-mining phone conversations (some sources suggest it even provided the software that the National Security Agency (NSA) used to scrutinize phone calls during its widely publicized warrantless wiretapping scandal), as well as television providers like MSNBC looking to make their videos easier to find in online search results. These white-label or “business-to-business” companies are players in the media ecosystem that remain largely, and often entirely, invisible to end consumers, and have until now remained equally unstudied by researchers and media critics. But they become visible when we begin to pay attention to the path that content takes to consumers. They play an essential role in determining where and when television content is available—and to whom—one that aligns neatly with more traditional concerns of media scholars about the terms on which we access and participate in our culture. Moreover, social and technical dynamics that influence the path of content, making it variously easier or harder to move, access, and manipulate on particular terms, not only exist in the distribution pipeline to end users, but also are at work as content moves within and between big media organizations, which today are more often than not assembled out of many smaller companies, all with their own legacy publishing platforms and organizational subcultures. Getting all these to work together harmoniously within a large media corporation is as much a feat of heterogeneous engineering as is navigating the commercial, legal, and technical hazards of digital distribution to end users. Thus, the dynamics and challenges of production and distribution are not quite so distinct as they are often taken to

Introduction 11

be, and in fact may be more intimately related than at any time in recent memory (Ananny 2013). This fragmented state of affairs within media organizations, and the process of navigating it, has variously been described to me by television news workers and executives as “innovative,” “messed up,” “crazy,” and “chaotic.” Organizational scholars prefer terms like “heterarchic” (Stark 1999, 2009) or “postbureaucratic” (Kellogg et al. 2006). Still, even as organizational researchers have begun to describe trends toward fragmentation and liminality within media firms, their findings remain in stark contrast to the picture of these institutions painted by classic media sociology, which in foregrounding the norms of journalism as a pervasive and remarkable achievement of professional socialization has tended to privilege structure over the agency of both individuals and distinct groups within large media institutions (Cottle 2007; Bivens 2014). It is also a jarring departure from accounts of vertical integration by some media economists, which depict the concentration of media firms as a shoehorning of smaller media outlets into organizational hierarchies that squelch agency and diversity in favor of command and control. But if media scholars are just beginning to grapple with the controlled chaos that exists within large contemporary media organizations, the same presence of myriad, loosely coupled firms and teams existing within and across organizations (alongside more traditional managerial hierarchies) is a hallmark of the “postmodern” style of systems building and management described by scholars of science and technology, like Thomas Hughes (1998), in which “discontinuous change is the expectation of project professionals” (p. 302). For this reason, and more broadly owing to their facility for explicating emerging, technologically mediated social structures, the history and sociology of sociotechnical systems have become of increasing interest to organizational scholars (Chia 1995; Girard and Stark 2002; Grabher 2002; Jackson et al. 2002; Kellogg et al. 2006; Neff and Stark 2004), and the tools of systems scholars can be neatly applied back to media to create a better understanding of contemporary firms and how they innovate. This brings me to a final reason for shifting my analytical lens to the realm of the sociology of sociotechnical systems: a great deal of contemporary literature about online television and news has criticized traditional media organizations for failing to change with the times. There is now an established body of work on the reluctance of “legacy” media organizations to adopt new technologies and business models over the past two decades.4

Introduction 12

But as Dwyer (2010) has noted, in the second decade of the twenty-first century, “traditional media have evolved to the point where online platforms are now integrated and necessary components of their businesses” (p. 52). A focus on systems sociology appropriately reframes the issue of the “reluctance to adopt” Internet technologies and practices as “the challenges of embracing” these new developments. This is particularly true of the framework of heterogeneous engineering, which opens our eyes to the incredible diversity of resources that are necessary to create a new, reliable sociotechnical system like a distribution route or a social publishing platform. Ultimately, then, while this book at times draws from—and hopefully contributes to—the fields of journalism and television studies, it is ultimately about networked distribution systems and how they are forged. My hope is that, taken this way, this book will have the potential to contribute to our understanding of how it is that a great diversity of media products make their way to the screens sitting atop our desks, gracing our living rooms, and tucked in our pockets. THE CASE OF MSNBC

In Digitizing the News, Pablo Boczkowski (2004) wrote about the efforts of print news organizations to develop their initial online presence. Just over ten years later, the U.S. print news industry has been digitized—or rather, the part that hasn’t is rapidly disappearing. The Pew Project for Excellence in Journalism (2010) estimates that over the previous decade, American newspapers lost roughly 30 percent or $1.6 billion in annual reporting and editing capacity. National news magazines, like Time and Newsweek, have also lost substantial resources. While the economic recession greatly accelerated the decline in the organizational resources of news outlets, the larger story has been a shift by audiences to online news consumption. The resulting pressure to monetize online distribution has led to what Nikki Usher and Patricia Riley (2010) have referred to as a “transform or die” mantra at many news organizations with regard to digitization and the adoption of new media technologies. Television has not been exempt from American journalism’s revenue crisis. Network television news has lost over half of its reporting resources since the 1980s (Gold 2010a, 2010b; Stelter 2010a), while local news divisions have suffered as well (Pew Project 2010). Among the major American legacy news media, only cable channels have avoided this major dropoff in revenue and resources in the wake of the recession, though even this

Introduction 13

may be changing (Pew Project 2013). Moreover, as broadband penetration in the United States has increased and more distribution channels for television programming have opened up online, providing potential avenues for consumers to bypass traditional cable and broadcast delivery mechanisms, the economic future of television as a whole and cable in particular has become subject to uncertainty and speculation, much as the world of print journalism was a decade ago (Glaser 2010; Lotz 2007; Tryon 2013). Meanwhile, the economic future of journalism more broadly remains in such flux that journalism professor Jay Rosen (2011) has declared, “A common mistake in seeking out the next business model for news is to begin by positing that such a model exists within the known universe.” In short, journalism, television, cable, and—to add a wrinkle—the Internet itself are all evolving in rapid and often uncertain ways. And at the nexus of these volatile industries is a small but rapidly growing group of individuals and firms whose job it is to develop and maintain online distribution channels for what is still understood to be television news programming. Their work, and the tensions surrounding it, provide a fulcrum from which to pry analytically in new ways at some of the largest shifts within our media landscape. This project began in 2009 with a series of interviews with online producers, software developers, and executives at the news divisions of each of the three major broadcast networks, and my continuing discussions with individuals at all these companies are reflected in this book. But the distribution networks by which media and information travel are as varied as the providers are numerous, and as I highlighted earlier, large media companies are sufficiently full of staffs, project teams, partnerships, subsidiaries, and vendor relationships that to study any one organization at scale is in fact to take on responsibility for documenting a host of interlocking case studies. As such, while keeping an eye on the broader landscape, I ultimately chose to focus on one company, MSNBC, as an exemplar. MSNBC is a rich case: as a hybrid company built to link television and the Internet, it struggled with how to innovate through new forms of distribution while sustaining its reputation in a traditional medium. The protean network of alliances and tensions that ultimately resulted in MSNBC’s acquisition and reimagining by Comcast centered around distribution concerns as well, and provide a fascinating example of how media conglomerates shift and evolve in response to these issues. The distribution network established by MSNBC is a hybrid, too, full

Introduction 14

of competing internal efforts, third-party information platforms, networks bought and incorporated, and kludged solutions to complex technical demands. To learn more about the challenges of embracing new media within legacy media organizations, as opposed to the reticence toward them chronicled in earlier scholarship, I searched for groups that were pursuing online distribution aggressively. Consequently my primary field sites within MSNBC included the popular Rachel Maddow Show, which vigorously pursued changes to MSNBC’s online strategy; MSNBC.com’s East Coast offices, where producers adapt television content for the web; and Newsvine, a subsidiary of MSNBC.com that built a great deal of the technological infrastructure through which content from MSNBC and NBC News programs was put online.5 While I focused on these groups, over the course of fifty interviews I also spoke with individuals responsible for many other properties and projects at NBC News, MSNBC TV, and the MSNBC Digital Network (which in 2012 became part of NBC News Digital), as well as people in related positions at other firms and networks. The book also examines how MSNBC’s online video player technology was assembled, as well as its next-generation content management system, SkyPad. It ultimately narrates how online distribution strategies at large media organizations are forged not necessarily from the top down, but just as often “horizontally” (Girard and Stark 2002; Kellogg et al. 2006; Neff and Stark 2004; Stark 1999)—that is, through the actions of small teams and individuals throughout the company (and sometimes outside of it), each working toward their own provincial objectives.6 STRUCTURE OF THE BOOK

Prior to launching into the book’s main arguments and cases I set the stage in Chapter 2, where I lay out the broad importance of studying media distribution, delineating how research and critique in this area has the potential to inform our view of a wide array of traditional concerns in media scholarship. Part 1 of the book then follows the path taken by MSNBC television programs as they made their way to online audiences, highlighting the role of different actors and artifacts along the way and introducing conceptual tools for analyzing them. In Chapter 3, I suggest that the way to understand media distribution is through the notion of the “sociotechnical system,” which comes from the field of science and technology studies. Like the power grids, public transit systems, and shipping routes studied

Introduction 15

by scholars of systems, media distribution involves a complex network of technologies, economic forces, social norms, and institutional actors. By way of example, I introduce MSNBC and its inner workings in the period between 2010 and 2011, and outline the path that MSNBC’s television content took between the time it was recorded and the moment when it was ready to appear on online viewers’ screens. In laying out this “chain of custody” of various actors (human and technological) handling television content on the way to the consumer, I preview many of the examples and concerns that will be explored in greater detail in later chapters. This tracing of the path of content continues throughout Part 1, highlighting how at each turn particular technologies, market concerns, and institutional actors made themselves felt in ways that influenced which audiences got to see what. Chapter 4 examines the construction of MSNBC.com’s web video player, the primary means by which NBC’s cable and broadcast news programs were seen online. Each of the video player’s features and idiosyncrasies was the result of a complex negotiation between different parties within Microsoft and NBC Universal, as well as a response to market pressures and legal concerns. In Chapter 5 I explore “off-site” and mobile distribution of MSNBC video, including the development of functionality that allowed the video player to be embedded on users’ personal websites and various technical solutions developed or purchased by MSNBC .com for piping its video to phones and tablets. This discussion introduces and examines the role of several “business-to-business” companies that I dub “transparent intermediaries” (Braun 2014)—that is, companies and technical infrastructures whose existence and operation go generally undetected by end users, but who play a major role in sculpting the business models that media companies use in the online video marketplace. While scholars of technology have often highlighted the way that technologies are “black-boxed” (Latour 1987)—stabilized and closed down into usable tools—the video player and mobile solutions described in Chapters 4 and 5 were under rapid development at the same time that they appeared stable to end users. Deconstructing their various parts and telling the story behind them serves as a model for how we might begin to think about online distribution in a more complex and nuanced manner. Part 2 of the book explores MSNBC and its online distribution platforms from the vantage points of different system builders within the larger organization, examining how small teams pursuing provincial objectives

Introduction 16

can generate changes within a much larger postbureaucratic organization. This exploration begins in Chapter 6 with the development of conceptual tools for considering large contemporary media organizations, which I then take to two of the book’s major exemplars. This chapter contains some of the main conceptual arguments of the book, many of which apply broadly to the study of news and media work. In Chapters 7 and 8, I examine the case of Newsvine, a Seattle-based, news-themed social network acquired by the MSNBC Digital Network in 2007 that became the basis for a new content management system used to put MSNBC TV and NBC News content online. While the original Newsvine site still exists, under MSNBC ownership the startup behind it served largely as an internal software development firm, responsible for much of the technological infrastructure that underpinned the online presence of MSNBC’s primetime television programs by the end of 2011. In this context Newsvine had become a “white-label” intermediary, in the sense that its brand and its role in delivering content were largely invisible to end users. The infrastructure that Newsvine built for MSNBC, however, was constructed on top of the original software framework that powered the startup’s existing social news site. These two chapters tell the story of Newsvine, from its origins until the time it became an MSNBC property, and how its attempt to design a dual-purpose platform that simultaneously powered both MSNBC’s mainstream media sites and its own social news venture ultimately resulted in unexpected changes to both how MSNBC content was delivered and how the original social news site functioned. While Newsvine’s origins as a company, a software platform, and an online community initially had nothing to do with television, they left indelible marks on MSNBC, and offer a prime example of how system builders whose interests appear at first glance to be far removed from TV are increasingly becoming central to the process of television distribution and important to understand. Chapters 9 and 10 explore the case of the Rachel Maddow Show, and highlight how MSNBC evolved not just in response to external pressures, but also via internal disruption. Hired by MSNBC as a primetime host in late 2008, Rachel Maddow had gained much of her prior celebrity through the Internet. She and the staff she hired at MSNBC were determined to pursue an online strategy that was far more aggressive than those of previous programs (“We realized pretty early on that MSNBC.com was not our website. And so therefore if we wanted to have anything useful online, that

Introduction 17

we were going to have to do it ourselves.”). These two chapters tell the story of how the Rachel Maddow Show forged a new and different online strategy that ultimately became the model for other MSNBC television programs. Having established the provincial and heterarchic manner in which online distribution systems were forged at MSNBC, Part 3 explores how the online company’s distribution channels were likewise built to serve a diverse array of online audiences. Chapter 11 explores the notion of the “conversation economy” in more depth, deepening it via a variety of conceptual tools while examining MSNBC.com’s strategy of cultivating numerous brands to capture the attention and sharing activity of different niche audiences. Chapter 12 looks at the technical underpinnings of this organizational trend, which came in the form of one of the company’s most ambitious infrastructure projects ever—a next-generation content management system called SkyPad. Built to take in and spit out almost any type of multimedia content and allow for the construction of entirely new websites and brands on the fly, SkyPad—and a counterpart called M3, produced by Newsvine—represented an emerging vision for enlisting provincial editorial groups and diverse audiences in the distribution of mainstream media products online. Chapter 13 brings this strategy into focus by examining its ultimate effects on the original Newsvine architecture and closes by considering the nature of the relationships between distribution, audience engagement, and more traditional editorial concerns. Over the course of the book, the reader will encounter MSNBC from the perspective of numerous different groups both inside and outside the organization. It is, in short, a highly complicated, heterarchic collection of entities with many moving parts, both big and small—exactly what we might expect of a contemporary media organization. What turns out to be surprising is not that MSNBC is complicated or messy, but that the tangle of different actors and cross-purposes that made it up, and somehow allowed it to function successfully, can be rendered in conceptual clarity using the theoretical tools introduced and developed in previous chapters for understanding sociotechnical systems. Following content as it winds its way through the organization to end users, and interrogating the arrangements we encounter along the way, helps uncover the actors, relationships, and important interests that are commonly left out of—and are sometimes even invisible to—scholarly analyses that focus solely on the production of television or its effects on audiences. In other words, the focus on distribution gives us an altogether different (and hopefully

Introduction 18

more comprehensive) view of a large media organization and its evolution that not only is valuable in itself, but also speaks in important ways to the more traditional concerns of media scholars. Chapter 14 ties together the conceptual interventions from across the book into a broader approach for examining contemporary, multifaceted media organizations and their efforts at online distribution—a task I argue becomes increasingly important as we concern ourselves with the health of our media institutions and the terms of our access to culture in the new century. The Epilogue closes the book by providing an update on MSNBC and NBC News’s online footprint since their acquisition by Comcast, exploring the restructuring of MSNBC.com into an expanded NBC News Digital and the creation of an entirely new website for the MSNBC cable news channel. While these developments occurred after the initial interview research conducted for the book, I show how they square with and were in many cases precipitated by developments documented therein, and how the trends I have laid out are equally relevant going forward.

2//why media distribution matters a critical interlude

Media scholars discuss distribution somewhat rarely, tending instead to frame their questions in terms of the ins and outs of media production, analysis of media content, or the impact of media on audiences. Moreover, Alisa Perren (2013) astutely observes that existing scholarship on distribution is rarely framed as such, but is more commonly approached as the study of media infrastructures, political economy, or any of a number of other topics. Yet it is precisely because distribution touches on so many other areas of social importance and scholarship that it deserves to be viewed as a subject of study in its own right. To the extent that media researchers have obsessed over the changes to media production, consumption, and audience effects wrought by cheap, instantaneous, and customizable transmission of information via the Internet, they’ve in many ways made the point for me that distribution matters deeply to traditional areas of scholarship and the normative concerns these address. Inquiring systematically into how this transmission of content is actually accomplished, then, serves a quintessential goal of social inquiry. Indeed, media distribution, and the balance of structure and agency enacted in different approaches to it, are very important to an array of major social and scholarly concerns, including the conditions under which people congregate and imagine their communities; the terms under which they labor and the manner in which industries are structured in an increasingly knowledge-based economy; the way in which social groups become variously visible and/or scrutinized within our larger society; the limits of freedom of speech and assembly; the means by which people converse and 19

Why Media Distribution Matters  20

engage in political participation; and the norms and possibilities of our literary and artistic culture. WHY, WHEN, AND HOW PEOPLE CONGREGATE

In his own work on distribution, Chuck Tryon (2013) quotes Charles Acland (2003), writing: “The organization of how, when, and under what conditions people congregate is a fundamental dimension of social life.” Here the technologies and practices of media distribution play a key role. Decisions about distribution determine, for example, whether people must assemble at the theater to watch a newsreel or film, whether they can do so as a family at home, or whether they may view it individually on a subway ride to work. Tryon (2013) argues that “contemporary media platforms actively solicit an individualized, fragmented, and empowered media consumer, one who has greater control over when, where, and how she watches” (p. 14). This is significant in that scholars from Louis Althusser (2011) to contemporary theorists like Michael Warner (2002) and communication researchers like Yong-Chan Kim and Sandra Ball-Rokeach (2006) have asserted that media infrastructures likely play a major role in defining our sense of which communities we belong to and to whom we have civic responsibilities. Distribution platforms provide affordances and constraints that help to determine how we as potential viewers can be segmented into target audiences. They influence which publics we share experiences with and imagine ourselves a part of. Communication infrastructure theory (Kim & Ball-Rokeach 2006) suggests that audiences served by the same local media—who share a broadcast radius or a print circulation footprint, for example—are more likely to conceive of themselves as part of a common public with shared concerns. Richard John (1995) illustrates this compellingly with his history of the U.S. Postal Service and the manner in which the physical presence of a distribution infrastructure of post offices and mail carriers made the young republic tangible for even the most far-flung of its citizens. He quotes John Barton Derby saying, in 1829, that “while the general government of the United States poises and holds together the whole, no man in the country ever feels its direct action (when it is peacefully and constitutionally administered), excepting in the appointment of a postmaster of his village” (p. 5). Similarly, John McMillian (2011), in his work on the 1960s underground press, quotes Abbie Hoffman saying, “The most valuable thing

Why Media Distribution Matters  21

about the alternative press is that it’s . . . a visible manifestation of an alter­ native culture. It helps to create a national identity” (p. 189). There is a sense of “imagined community” here in the style of Benedict Anderson (1991), or what Kristy Hess (2013) has more recently referred to as the “bonding” function of media—its role in “constructing a sense of ‘connectedness,’ ‘community,’ or ‘sense of place’” (p. 118). The presence of papers on newsstands and programs on the air serves in its own way to make various subcultures and segments of the public visible to the larger populace, and therefore plays an important role in identity politics. This same presence also “interpellates” publics in Althusser’s (2011) sense of the term, by calling them into being. Youth encountering the underground papers of the 1960s, for example, weren’t just made aware that the counterculture existed “out there.” They learned that it was something immediate and local being picked up by members of the community, which in turn made it possible to imagine oneself as a participant in the counterculture. Likewise, Mitchell Duneier (1999), in his ethnography of predominantly African American book and magazine vendors in New York City, explores how these vendors were among those who helped to create a secondary market for “black books”—African American Studies textbooks and other books on black history and culture—by selling and conversing about them on the street: “Though African-American Studies reached ­African-American college students through academic channels, the emerging discipline also had—and continues to have—a secondary impact among African Americans outside the universities, through the influence of alternative distribution networks among the greater African-American population” (p. 25). Duneier gives rich examples of young African Americans reading and conversing about these texts, as well as discussing the manner in which their experience of them feeds their larger sense of cultural identity and self worth. As he puts it, “The very presence of books on the street tends to prompt discussions about moral and intellectual issues” (p. 25). But if distribution infrastructures have the ability to create or reify social imaginaries and public spheres, they can also traverse between them to great effect. This ability of distribution networks to spur social change by cutting across different public spheres—both socially and g­ eographically— and helping to reconfigure them has long been noted in civil rights circles. In his collected writings, Martin Luther King, Jr., for example, eloquently describes the importance for regional protesters of gaining access to the widespread distribution afforded by national broadcast networks and

Why Media Distribution Matters  22

prominent newspapers (Clayborne 1998). He attributed many aspects of the movement’s success to the ability to circulate images and descriptions of racism and brutality beyond the geographic regions where these acts were deemed normative or socially acceptable. By bringing “the conscience of the nation” to bear on the norms of particular regions, large-scale distribution contributed to the movement’s ability to reconfigure the social and political alignments of the United States. But if large distribution networks can bring positive change by cutting across social and regional boundaries, critics fear they can also hinder regional and cultural diversity in some contexts. Tryon (2013) and other media commentators worry, for example, whether our increasingly international media distribution infrastructures will provide support for local media, or whether new distribution systems such as Hulu, Netflix, or other “subscription video on-demand services” will end up as merely another tool for the mass export of American culture. Of course, such concerns are neither new nor uniquely international. As a historical study of the U.S. Postal Service by John (1995) reveals, similar debates were occurring in the United States in the eighteenth century regarding how best to organize postal subsidies for newspapers in a way that would not lead to the trampling of small, local papers by large metropolitan publications. Brian Larkin (2008), too, illustrates the manner in which such cultural imports have in some historical contexts reinforced divisions along racial and class lines. He discusses, for example, how the distribution of the BBC and other foreign radio broadcasts in Northern Nigeria by the British introduced new markers of class (for instance, those who could afford a radio versus those who could listen only by congregating at public listening stations), education (such as a British-educated middle class that spoke English and could understand the new broadcasts versus those who could not), and culture (for instance, groups who understood the station’s Arabic-language programming versus the majority who spoke the regional Hausa dialect). The dynamics of distribution—especially digital distribution—may be different today, but many of the concerns and issues discussed by Larkin are still valid in various contexts. DISTRIBUTION IN MEDIA WORK AND THE MEDIA INDUSTRIES

Mark Deuze (2007) has argued that as more and more sectors of the economy fashion themselves as producers of information and intellectual property, it becomes more important than ever to understand the conditions

Why Media Distribution Matters  23

of production in the media industries, which have increasingly become bellwethers for the way people live and work across a wide variety of economic sectors. And indeed, from Deuze’s own 2007 volume Media Work to Thomas Malaby’s Making Virtual Worlds (2009) to Gina Neff ’s Venture Labor (2012), many contemporary scholars have taken an interest in the conditions of work within the media industries. Unsurprisingly, distribution has had a major influence on these conditions. For example, Pablo Boczkowski (2010) and Rena Bivens (2014) have illustrated that as the means have become available to move news and information faster, long-held professional values within journalism surrounding “immediacy”—the need to have the latest story, to get it out quickly, to have up-to-date information, and, where possible, to “go live”—have risen in importance to the point of at times crowding out other journalistic norms. In their accounts, immediacy now often trumps concerns like accuracy, comprehensiveness, and originality of content. Bivens (2014), for example, documents the manner in which delivering a live feed from the scene has often proven more important in television news than actual information gathering, while Boczkowski (2010) compellingly argues that in an environment where reporters can monitor one another’s output in real time, news sites are inclined to offer the same information as their competitors, rather than generate original stories. Attention to distribution practices and their negotiated features also uncovers tensions and fault lines between different cultural industries. Consider, for example, the divergent interests of Hollywood on the one hand and consumer electronics companies on the other in deciding whether to limit the ability of devices to record, copy, and play back digital video in accordance with various forms of “digital rights management” (DRM) advocated for by entertainment firms (Gillespie 2007). Debates have similarly cropped up between new media distributors like Apple and Amazon on the one hand and publishers on the other over how to price products available in digital storefronts. And as we shall see, tensions between television channels and cable providers over who has the right to distribute content online have been a major concern in recent years. Additionally, insofar as media ownership models affect media distribution strategies, Jennifer Holt (2011) has argued that the simultaneous concentration of media ownership and convergence of media technologies has created a crisis for regulators, rendering the traditional language and procedures of anti-trust enforcement deeply inadequate.

Why Media Distribution Matters  24

VIEWING AND BEING VIEWED: AUDIENCE VISIBILITY AND SURVEILLANCE

If, as Anne Balsamo (2012) puts it, audiences use the increasingly personalized media available to them in piecing together “a sense of self,” media companies find themselves quite busy creating a sense of each of us, too, in the form of consumer profiles. Within the media industries, audiences constitute an essential but complex unknown whose wants, needs, opinions, and interests must be selectively measured and simplified to produce an actionable market. And this is precisely the business in which many of the most prominent companies providing contemporary distribution infrastructures see their future. Simply put, a major effect of online distribution is that distributors have become ratings and analytics firms. In enabling millions upon millions of video views, the companies that provide tools like video players and ad networks are simultaneously collecting enormous volumes of information about audiences and their use of content. As Tryon (2013) puts it in discussing the rise of on-demand distribution, the “offer of liberation from the viewing schedule is often accompanied by increased surveillance, giving studios, streaming video services, and social media companies more precise information for their efforts to market directly to . . . individualized viewers” (p. 14). Joseph Turow (2004, 2011) and other scholars like Philip Napoli (2010), C. W. Anderson (2011), and Josh Heuman (2006), have dutifully underscored the notion that “the manner in which [media producers] imagine their audience has public consequences” (C. Anderson 2011, p. 530). Since not everything that we value or find consequential about our cultural identities or public discourse can be easily quantified, measured, binned, graphed, or regressed, and because those things that are measured may be condensed, weighted, and acted on in different ways by content providers with varied commercial interests, an important discussion has arisen among media scholars about how media providers conceptualize their a­ udience—what researchers have variously called the “configuration” (Woolgar 1991), “institutionalization” (Napoli 2011), or “industrial construction” (Turow 2004, 2011) of audiences. Turow (2004, 2011), in particular, notes how marketers and advertisers use analytics to reduce audiences into categories of “target” and “waste” in ways that have influenced the content that publishers put forward. Similar concerns—about the nature of audience measurement and its influence on the type and availability of content produced—have long swirled around the strategic choices and ambiguities surrounding Nielsen’s modeling and

Why Media Distribution Matters  25

representation of television audiences, and have only increased as Nielsen itself has sought to take advantage of localized and digital analytics in ways that may or may not accurately represent the habits and interests of minority audiences (Heuman 2006; Napoli 2011). From a civic perspective, the sorts of traditional mass media measured by Nielsen are, at their best, what Cass Sunstein (2007) calls “general-­ interest intermediaries,” exposing us to views we might not have chosen of our own accord and making the interests of minority and opposition groups visible to the majority. For concerned citizens, what has always mattered about Nielsen statistics is whether they succeed in making these groups visible and represent their activity fairly. The politics involved are the politics of representation and the risk of misrepresentation or, worse, omission and invisibility. The worry here is that groups whose consumption activities are not accurately recorded will not be not sought after as audiences. Their interests and views may consequently be less readily represented in the content of media, and thereby omitted from the public agenda. Such concerns have historically invited challenges—warranted or not—to Nielsen’s sampling methods, and made changes to them the subject of much political debate. The new media analytics provided by online distributors, by contrast, have the ability to provide “universe data”—a complete record of viewing activity, rather than a random sample. Because of this, they may, at their best, succeed in making visible the existence of interests and publics that had been lost in the noise of previous systems of representation. At the same time, in a world where providers have all the data about their audiences, identifying trends is no longer a matter of which data are statistically significant—everything is—but rather which data are interesting. And as social critics point out, this begs the further questions of “interesting to whom?” and “for what reason?” The obvious answer in a commercially driven media environment is that audiences are interesting when they are profitable. Thus the worry is that rather than resulting in a better reflection of our myriad organic social interests and concerns, the development of new media analytics provided by distributors has tended, and will continue to tend, toward treating us as consumers rather than citizens, niche markets rather than publics. As Amanda Lotz (2007) observes, television “increasingly exists as an electronic newsstand through which a diverse and segmented society pursues deliberately targeted interests. The U.S. television audience now can rarely

Why Media Distribution Matters  26

be categorized as a mass audience; instead, it is more accurately understood as a collection of niche audiences” (p. 5). Like traditional audiences before, online viewers will be defined, first and foremost, as consumers in many contexts. But while this may be a continuation of previous trends, civic-minded critics have long seen this tendency as a necessary evil of commercial media. Moreover, observers like Turow, Napoli, and C. W. Anderson worry that, as audiences are interpellated into ever more convenient, instrumental, and commercially viable social identities, the drawbacks of commercialization may eventually outweigh its benefits. Additionally, these trends are taking place in an increasingly fragmented media environment. Napoli (2011) and others have pointed out that as media companies demand, and distributors provide, ever more extensive analytics in their attempt to monetize smaller audiences and niche markets more efficiently, the sorts of data on consumers that they attempt to collect, correlate, and sell will increasingly run up against users’ expectations of and legal rights to privacy. DISTRIBUTION AND STATE OPPRESSION

As we’ve seen, distribution renders societal groups and subcultures visible on a number of levels. First, the very presence of a newspaper, broadcast, website, online forum, or other distribution channel becomes “a visible manifestation” of that culture, in the sense of both giving a platform to the authors and inviting prospective audiences who share the authors’ views, affinities, or simply geographic region, to identify as a group. Second, distribution platforms increasingly have built-in means of surveillance in the form of detailed analytics and user information that make audiences visible in a far more concrete fashion.1 The unfortunate corollary to this heightened visibility is that those who mean to suppress, or prevent the formation of, particular social groups and identities often target the distribution of their media, as when law enforcement agencies and prosecutors in the 1960s took aim at the New Left by selectively enforcing obscenity laws, charging the proprietors of underground papers with distributing pornography and other objectionable materials (McMillian 2011). Similarly Mitchell Duneier (1999), in his ethnography of street vendors in New York City, explores how communities of largely poor and unhoused individuals in Greenwich Village who make

Why Media Distribution Matters  27

their living selling books and magazines have often been the selective targets of “broken windows” regulations and law enforcement policies aimed at evicting the poor by destroying their livelihoods. Their books and magazines were regularly tossed into garbage trucks and their activities labeled threats to the social order. In the area of new media distribution, the Internet censorship practices of countries like Iran and China may spring to mind when discussing the ability of states to impinge on distribution as a form of political oppression. But even in liberal democracies we’ve seen the rise of regulatory frameworks aimed at cracking down on illegal file sharing. It’s true that even strong advocates of intellectual property reform have condemned some aspects of file sharing. James Boyle (2008), for example, has stated representatively, “I see no high-minded principle vindicated by middleclass kids getting access to music they do not want to pay for. It is difficult to take seriously the sanctimonious preening of those who cast each junior downloader of corporate rock as a Ché Guevara, fighting heroically to bring about a new creative landscape in music” (p. 77). At the same time, scholars including Brian Larkin (2008), Ramon Lobato (2012), and Michael Newman (2012), as well as Sharon Strover and William Moner (2013), have highlighted the manner in which “piracy” often serves to fill gaps in formal distribution networks, extending the reach of content to underserved nations and communities. The creation and (at times selective) enforcement of national and international copyright agreements regulating distribution can thus have political ramifications. This is especially true when, as scholars like Boyle (2008) and Tarleton Gillespie (2007) have highlighted, even legitimate remix practices, as well as unrelated speech occurring on platforms deemed infringing, can be potentially targeted through the blunt technical and legal instruments of contemporary copyright enforcement. There are thus highly legitimate worries, reminiscent of those espoused by the publishers of the 1960s underground press, about the potential to squelch “undesirable” speech by targeting distribution channels. This issue has been foregrounded recently in debates over regulatory frameworks like the PROTECT IP Act (PIPA) and Stop Online Piracy Act (SOPA) that even legitimate high-profile sites from Wikipedia to reddit argued would hypothetically give copyright holders unprecedented license to bring down their content, as well as to freeze their traffic and assets.2

Why Media Distribution Matters  28

RESPONSIVE DISTRIBUTION

As the distribution of media increasingly occurs on digital platforms, a number of scholars, most notably Gillespie (2014a, 2014b), have remarked on the rising importance of “public relevance algorithms” designed to programmatically make decisions about how best to place content in front of audiences. In particular, Gillespie (2014a) notes how these algorithms— which, among other things, curate our social media feeds—increasingly blur traditional distinctions and expectations surrounding the difference between the distribution of personal messages and the curation of entertainment media, as well as between editorial and distribution concerns more broadly. His discussion here is of sufficient import that it’s worth quoting at length: Social media . . . truly violates a century-old distinction we know very well, between what were two, distinct kinds of information services. On the one hand, we had “trusted interpersonal information conduits” —the telephone companies, the post office. Users gave them information aimed for others and the service was entrusted to deliver that information. We expected them not to curate or even monitor that content, in fact we made it illegal to do otherwise; we expected that our communication would be delivered, for a fee, and we understood the service as the commodity, not the information it conveyed. On the other hand, we had “media content producers”—radio, film, magazines, newspapers, television, video games—where the entertainment they made for us felt like the commodity we paid for (sometimes with money, sometimes with our attention to ads), and it was designed to be as gripping as possible. We knew that producers made careful selections based on appealing to us as audiences, and deliberately played on our emotions as part of their design. We were not surprised that a sitcom was designed to be funny, even that the network might conduct focus group research to decide which ending was funnier (A/B testing?). But we would be surprised, outraged, to find out that the post office delivered only some of the letters addressed to us, in order to give us the most emotionally engaging mail experience. Now we find ourselves dealing with a third category. Facebook promises to connect person to person, entrusted with our messages to be delivered to a proscribed audience (now it’s sometimes one person, sometimes a friend list, sometimes all Facebook users who might want

Why Media Distribution Matters  29

to find it). But then, as a part of its service, it provides the News Feed, which appears to be a running list of those posts but is increasingly a constructed subset, carefully crafted to be an engaging flow of material. The information coming in is entrusted interpersonal communication, but it then becomes the raw material for an emotionally engaging commodity, the News Feed. All comes in, but only some comes out. In other words, we are increasingly encountering online tools—“public relevance algorithms” (Gillespie 2014b) and the distribution platforms of which they are a part—that selectively deliver messages based on their surveillance and interpretation of our past behavior. This mode of responsive distribution results in audiences, variously dubbed “networked publics” (boyd 2007) or “calculated publics” (Gillespie 2014b), whose membership is constantly shifting based on how the software involved variously prioritizes, classifies, associates, and filters the larger population of users (Diako­poulos 2014). Responsive distribution, in other words, is increasingly a key aspect of the conditions under which we congregate, socialize, and exchange information. But even as we’ve begun to develop the keen sense that responsive distribution matters in important ways, as a society we are still wrapping our heads around its consequences. DISTRIBUTION AND PUBLIC PARTICIPATION

Although social media sites and other online distribution platforms that deliver messages to and from private citizens also increasingly distribute news and entertainment, violating traditional distinctions between these realms, scholars including John McMillian (2011) and Fred Turner (2006) point out that distribution networks have long contained important parti­ cipatory affordances—means by which audiences and users can circulate their own voices. On a mundane but nonetheless important level, letters to the editor and public access cable channels provide distribution at scale to everyday citizens via the circulation of print and television. But more dramatic examples also exist. McMillian, for example, shows how the Underground Press Syndicate and Liberation News Service syndication networks forged by the 1960s alternative press gave significant exposure to accounts by ordinary members of the counterculture on the ground at events across the country. Fred Turner likewise argues that Stuart Brand’s Whole Earth Catalog and subsequent Whole Earth ’Lectronic Link online bulletin board service created

Why Media Distribution Matters  30

many-to-many distribution networks that put people from very different backgrounds in communication with one another. He illustrates how these distribution platforms ultimately served as network forums, wherein ideas from back-to-the-land counterculturalists intermingled with those of artists and members of the academic-military-industrial complex in ways that generated—and invited participation in—important new cultural categories, identities, and understandings of digital technologies. Many of our contemporary ideas and framings of the Internet and digital culture first gained traction as a result of these distribution spaces. Despite these compelling examples, the participatory promises of distribution platforms are often underrealized. As Tryon (2013) points out, for instance, the early expansion of cable television in the United States involved ample—and often unrealized—promises of increasing the reach of independent television channels and the diversity of voices available to television consumers. Still, online distribution platforms like social media sites are today often built around participatory affordances. Even digital platforms designed for distributing online more traditionally “mass” forms of news and information often have user publishing, commenting, and feedback mechanisms built into them in one form or another, which present a variety of opportunities and challenges to traditional media production routines. Moreover, Mike Ananny (2013) has shown how the way news organizations structure the application programming interfaces (APIs) and other participatory affordances built into their distribution platforms is significant not only for the manner in which they interpellate an “institutionalized audience” in Napoli’s (2011) sense and publicly position the news provider, but also for the way that responsibility for these mechanisms creates and maintains professional boundaries within organizations themselves. For example, NPR [National Public Radio] emphasizes the educational nature of its member network and envisions its API as delivering professionally produced content that supports its public mission and network members. The [London] Guardian stresses its API’s ability to be a central repository of vetted data around which developers and strategic partners might organize that will, in turn, support its advertising goals and financial health. The NYT [New York Times] describes its API as a trusted data source, fulfilling its journalistic mission to inform the public and creating a developer community to help drive traffic to NYT.com

Why Media Distribution Matters  31

and earn revenue. The NYT’s publicly stated motivations also point to less visible aspects of APIs: their structuring of intraorganizational relationships among software engineers, traditional newsroom workers, and news management staff. (Ananny 2013, p. 636) IMPLICATIONS FOR POLITICAL ACTIVISM

Noortje Marres (2012) argues that the material environment and built infrastructures we inhabit are key to understanding democratic participation. She takes her cue from pragmatists like John Dewey and Walter Lipp­man, who suggest that a public is best thought of as a community of individuals affected by the consequences of others’ actions, such that the community develops an interest in regulating those actions. In this scheme, the public is not a singular or constant entity; rather, many publics exist simultaneously and are continually springing up in response to shared concerns. Marres outlines what she calls the “dual positioning” of such publics, wherein individuals are by definition simultaneously affected by an issue, but not as yet recognized as stakeholders with a legitimate voice in the matter. As such, the challenges faced by democratic publics have to do with the ability of their prospective members to work out the nature of the problem affecting them as well as discern and articulate a relationship to it that legitimates their concerns in the eyes of others, ultimately coming to understand themselves as members in a “community of the affected.” All of these challenges are intimately intertwined with the material conditions in which individuals find themselves. Not only are the problems that such communities confront material in nature (for example, pollution, or a lack of transportation), as Roger Silverstone (1994) highlights, so too are the communication infrastructures and distribution channels by which individuals learn about their environment and come to identify themselves as publics in the first place. Here the aspects of distribution highlighted earlier, such as its role in interpellating and making visible different social groups and subcultures, become exceedingly important. Moreover, as Tryon (2013) illustrates, contemporary distribution platforms present us with a fragmented, high-choice media environment— what Lotz (2007) refers to as the “post-network era”—filled with hundreds of cable channels and hundreds of millions of websites. Much has been made of this environment and its prospective effects on political partici-

Why Media Distribution Matters  32

pation. Commentators like Sunstein (2007), for example, worry that in an information universe where people have unprecedented choice over the facts, sources, ideas, conversations, and views to which they are exposed, society will become more and more politically divided and polarized. Meanwhile, Markus Prior (2007) highlights that the same backdrop of vast numbers of media choices made possible by proliferating and increasingly efficient distribution channels has all but eliminated the captive audience for news. Now that consumers have more choices for what to watch or read aside from news, a large portion of the electorate—who, according to his data, also happen to be the more politically moderate segment—are choosing to read and view less news and more entertainment. He further argues that the result has been lower political literacy and voter turnout among a large portion of the population. At the same time, Prior (2007) illustrates the manner in which those who do monitor the news and share civically important items via blogs and social media—who engage in what industry professionals refer to as “social distribution” of news content—increasingly play the role of “monitorial citizens,” as originally outlined by Schudson (1998, 2000). The idea here is that as the practice of deciding public issues in contemporary democracies has shifted over time from the ballot box to the courtroom and other arenas, citizenship has become correspondingly less about informed voting in election years and more about constant “environmental surveillance” in the sense of identifying personal and civic harms and bringing them to the attention of the public. OUR MEDIA CULTURE

Beyond worries about what fragmentation does to our politics, those who value elevated media literacy for its own sake have variously expressed hopes and concerns about what contemporary trends in distribution will do to film and literary culture. For better or for worse, and probably for both in various respects, as new distribution platforms make it possible to view more forms of media in more places, the social protocols surrounding when, where, and how it is acceptable to use media are clearly in flux—as are the forms our media use takes. For example, regarding the expanding number of distribution channels for media of many different formats, Tryon (2013) neatly summarizes, New distribution practices . . . have the potential to shape our very understanding of the film or television text. As Schatz argues, franchise

Why Media Distribution Matters  33

filmmaking encourages the development of an “expanding textual system,” in which the movie is just a small part, a phenomenon vividly described by Henry Jenkins in his discussion of convergence culture, in which “the art of storytelling has become the art of world-building, as artists create compelling environments that cannot be fully explored or exhausted within a single work or even a single medium.” (p. 9) The argument here is similar to that made by Steven Johnson (2005) to the effect that the availability of new distribution channels for bringing viewers up to speed, from DVD box sets and online streaming services for so-called “catch-up viewing” of television shows to the appearance of television wikis, fan forums, and online episode summaries, has fundamentally altered the freedom of storytellers to create narrative arcs that progress over the course of many films or programs, as well as to concoct plot lines and jokes that improve with repeat viewing. This is similar to the argument made by some of my sources in the present volume concerning how audiences’ on-demand access to online breaking news has altered the editorial tone of evening newscasts, leading them to focus more on recap, analysis, and “meta” stories about news coverage, as opposed to play-byplay ­coverage. But not all distribution-inspired changes to media production involve an increase in sophistication. Tryon (2013), for example, also notes the manner in which media writers and producers feel increased pressure to hook the audience early into a clip, lest they click away to some other distraction. And both he and Aynne Kokas (2014) have explored how Hollywood studios are increasingly writing and editing films with an eye toward what will be enticing to China’s massive media market. Tryon (2013), while pointing to a number of ways in which digital distribution has arguably improved the experience of consuming film and television, such as the ability to access programming with far greater convenience and to socialize about it in more ways than ever before, admits to a great deal of “ambivalence about the ways in which digital delivery has altered the media industries” and media culture (p. 17). In particular, he points to the manner in which viewing has shifted in many contexts from being a shared, in-person experience in theaters and living rooms to a largely individual one, enjoyed by each family member on their own personal device—while conversations now take place more on social media and, perhaps, less in person. In this same vein, Christine Rosen (2004) has expressed worries about

Why Media Distribution Matters  34

our media culture very similar to Sunstein’s comments on politics, arguing that the power, ushered in by distributors like Netflix and Amazon, to select only those media products consonant with our existing preferences will lead to us as individuals possessing unfortunately narrow tastes in culture and a corresponding inability to appreciate or grapple with challenging works of art. Tryon’s (2013) concerns are consonant with these worries. He observes that contemporary film communities may have begun to center somewhat less on the local multiplex or arthouse theater and more on online fan forums organized around particular movies, genres, or filmmakers. He also highlights various discontents surrounding digital distribution methods introduced by media companies, such as windowing and licensing practices that make it difficult or expensive to obtain content on one’s desired service or device; the disappearance of collectible media, as DVD and CD libraries have given way to streaming services like Netflix, Hulu, and Spotify; the increased transitivity of our access to media, as titles now pop in and out of Redbox kiosks and the aforementioned streaming libraries at the whim of publishers; as well as digital rights management (DRM) technologies that can make viewing unnecessarily cumbersome and even prohibit some uses of media, such as sharing personal copies, that had been permissible under previous systems of distribution. Tarleton Gillespie (2007), Jonathan Zittrain (2008), Hector Postigo (2012), Gabriella Coleman (2013), and many other scholars similarly take aim at the manner in which technologies like DRM and legal frameworks for controlling distribution like the Digital Millennium Copyright Act have the potential to fundamentally alter people’s relationship with both the media and the devices they own, prohibiting creative, self-­cultivating, and occasionally politically engaged activities like the remixing and mashing up of media or tinkering with software and electronics. As Tryon (2013) neatly puts it, paraphrasing Vincent Mosco, many players within the media industries increasingly employ “myths of freedom and mobility to promote delivery methods that in fact give them greater control” over the content and devices accessed by users (p. 50). While the meta-level issues described in this chapter are meant to set the stage for greater sustained attention to media distribution as an object of study, many of the sociological and normative concerns that get “invested and contested” (Kompare 2010) in distribution systems become

Why Media Distribution Matters  35

apparent only as we get closer to the ground. That is because they emerge from the highly situated values and actions of the people who build and use these systems. The distribution systems described in this book represent an amalgam of different value frameworks adhered to by a surprising variety of different architects whose provincial goals affect their design and use. They contain within them these actors’ various pro-social commitments to supporting citizen journalism or engaging more directly with audiences, along with much more commercially driven concerns like the protection of intellectual property and the selective pursuit of audiences with greater expendable income. As Lisa Parks (2012a) has pointed out, we (perhaps increasingly) live in a world in which both news and entertainment share resources with a broad variety of corporate interests—where oil conglomerates share satellite bandwidth with regional television networks, and where the array of stakeholders whose interests shape our media infrastructures is often dizzy­ing. Her comments echo Silverstone’s (1994) observation about television, which could apply equally well to many other media: “It is no longer possible to consider television as a cultural apparatus or as a cultural industry without considering it alongside the supporting and interweaving technologies and political and economic structures that integrate it, both in production and consumption, into a more complex cultural and industrial whole” (p. 79).

This page intentionally left blank

3//mapping sociotechnical systems that deliver media

In John Law’s scholarship on the Portuguese trade route to India, he examined how merchants seeking a working avenue for their goods simultaneously took advantage of and contended with a wide array of forces, some of them natural, others economic, scientific, and social. In the same way that the influences of these heterogeneous actors became inscribed on the map of the Portuguese trade route—the volta do mar—I wish to carefully unpack the similarly diverse array of influences that make themselves felt on the path that television news takes to online audiences. Doing so in a nuanced way, however, will require us first to develop a few conceptual tools for dealing intelligently with the unwieldy multitude of actors and artifacts that affect the route of content on its way to us. RECALCITRANCE: CELEBRATING THE KLUDGE

In Ron Howard’s film Apollo 13, NASA engineers at mission control are charged with figuring out a way to repair the imperiled astronauts’ life-­ support system using only objects available aboard the spacecraft. More specifically, they famously need to fit a square peg into a round hole to connect two pieces of equipment that otherwise wouldn’t work together. Each one of the items aboard the vessel, with which the engineers intend to conduct the repair, is what you might call recalcitrant (Burke 1965): it helps to solve one problem, but often creates another, which has to be solved by adding yet another item, which in turn creates another issue, and so on. Kenneth Burke (1965) paints working with recalcitrant objects like these as a dialogical process, whereby the world pushes back on our raw inten39

Tracing the Path  40

tions, forcing us to iteratively temper them into a viable strategy. As he puts it, “The recalcitrance of your material discovered en route may eventually compel you to revise” your plans (p. 258). Andrew Pickering (1995) later encapsulated this notion of continual reappraisal in his memorable phrase, the “dance of human and material agency” (p. 251). More than the simple notion of “solving one problem while creating another,” recalcitrance, then, is a way of looking at agency, material and otherwise—the stubborn unwillingness of parts to bend to fit the need at hand. It’s the idea that things never entirely give up their shape. In the film, the end result is an unholy looking assemblage of bungee cords, duct tape, socks, and plastic bags—resources that are individually troublesome, but together solve more problems than they create. On a larger scale, John Law (1987) called this process of fitting together recalcitrant actors into a working assemblage heterogeneous engineering. But in Law’s view, to focus strictly on the gadgetry involved in such enterprises is inevitably misleading. People and corporations are regularly in the position of having to engineer solutions to their problems with recalcitrant tools, but when the introduction of a gadget to the emerging solution creates a new problem, it’s frequently patched not with another gadget, but with a lawyer, a merger, a marketing campaign, an employee training program, a new pricing scheme, and so on (Hughes 1987, 1998; Law 1987). Like technological implements, these bits and pieces are also recalcitrant and the problems they create may be solved with another gadget or another lawyer, and so on. This is the “heterogeneous” aspect of heterogeneous engineering—technology is only one sort of resource amid the various legal, social, scientific, and economic implements that get lashed together into a working system like, say, an online television distribution strategy. That the result looks complicated when we open it up to examine it is no surprise—there are no solutions in the world save for those full of socks and duct tape. This is an important insight in that many of the most interesting scholars on the subject of the politics of technology focus on its obduracy. In driving home the point that artifacts have politics and design has implications, they look at how artifacts often stick around in such a way that we must live with the consequences of their design (Bijker 2006; Flanagan, Howe, Nissenbaum 2008; Introna and Nissenbaum 2000; Latour 2002; Lessig 2006). This is true also of many media-studies works on infrastructure and, to a point, even of Hughes (1979), from whom Law draws

Mapping Sociotechnical Systems  41

much of his inspiration in thinking about sociotechnical systems. The conceptualization of systems given by Hughes (1979, 1987, 1998, 2004) is compelling as a basis for a concept of heterogeneous engineering for a couple of reasons.1 To wit, he demonstrates both the diversity of the resources involved in technological design and deployment, as well as the manner in which contemporary system builders came to appreciate and strategize around this complexity, particularly over the course of the twentieth century: “[Engineers] and other professionals influenced by them increasingly conceptualized the world around them in terms of systems. Formerly, they might have seen an airplane in isolation. Now they saw it as part of a system involving airfields, air controllers, fuel depots, and maintenance facilities. Formerly they might have conceived of a highway in isolation; now they placed it in a network of facilities, including automobiles, service stations, and traffic controls” (Hughes 2004, pp. 82–83). But Hughes, Law contends, imagines that while there may be thorny problems—so-called reverse salients (Hughes 1987)—that require the focus and attention of system builders from many different angles, once these problems are solved a system will largely be built according to plan. Moreover, once constructed, many of the systems Hughes studied were assumed to have a great deal of obduracy—that tendency to stay built.2 Turow’s (1992) power-role theory of media also takes a clever systems approach—painting the various organizations involved in media production and distribution as system builders competing to assemble resources and leverage in a hostile environment—but similarly the theory ultimately conveys a sense of obduracy, illustrating how the roles, resources, and practices of different media industries become firmly structured over time. For Law (1987), assumptions of obduracy can be misleading. Rather, he emphasizes the conflict and entropy that are inherent in the networks of alliances necessary to build a social and technological system that does what a builder wants. Instead of viewing the world as inherently stable, heterogeneous engineering describes sociotechnical systems as continually coming apart at the seams and emphasizes the great labor involved in merely keeping things the same. Law’s perspective can be read as a celebration of the stopgap and of the kludge, and so is particularly well suited to an examination of our fluid digital culture. Like the Red Queen in Alice in Wonderland, the heterogeneous engineer is running as fast as she can just to stay in the same place. Moreover, when it comes to distributing television online, what looks

Tracing the Path  42

like a solution to Apple or Cablevision may be anathema to Google, Warner Brothers, or CBS. Solutions are difficult, and the heterogeneous tools themselves are recalcitrant, in part because many parties are working simultaneously around and against one another to construct a system that advantages them—where necessary, at the expense of others. A given engineer seeks to make other engineers subject to her own interests, and to summon a particular configuration of the available bits and pieces, be they human, technological, legal, commercial, or otherwise, optimized for her own purposes—a process Michel Callon (1986) refers to as enrollment. It is precisely because numerous actors are at once playing this game at cross purposes that the malleability of the resources available to system builders is often highly limited. A WORLD FULL OF SYSTEMS

Heterogeneous engineering conceives of the world as full of systems, each competing to draw various actors into stable networks that accomplish the objectives of their respective system builders. In Law’s original conception, this is frequently a zero-sum game, in which the successful association of one network of resources results in the dissociation (breakdown or dissolution) of others. In other words, Law foregrounds the challenges of competition and of entropy in assembling stable sociotechnical systems. Thinking in terms of heterogeneous engineering, a system is never a done deal, never complete—at any given moment it is only more or less stable. I accept these insights, but rather than thinking in terms of association and dissociation, I prefer the concept of recalcitrance, described by Burke (1965)—rather than simply dismantling one another, networks that rub up against each other more often cause competing or enrolled system builders to alter their strategies and the resources with which they pursue them. In either case, a major advantage of using as a framework the idea of heterogeneous engineering, with its emphasis on both entropy and heterogeneity, is it foregrounds the fact that the construction and maintenance of particular social and technological arrangements—of distribution routes, for example—is not a one-time enterprise, but a continual, dialogical process in which the challenges to be examined are not stable or exclusively social phenomena, as sociologists have at times assumed. Rather, the materials, practices, and institutional arrangements surrounding contemporary distribution paths—the “voltas” of media circulation—bear more resem-

Mapping Sociotechnical Systems  43

blance to the “complex adaptive systems” described by David Stark (1999) and other organizational scholars, whose ideas we will visit later on. TECHNOLOGIES AS HETEROGENEOUS ARTIFACTS

We have so far been examining technologies as parts of heterogeneous systems. But it should also be pointed out that technology is itself a manifestation of heterogeneous forces. Children with active imaginations will watch some miracle of automation, like a mechanized pinsetter and ball return at the bowling alley, and substitute creative images of what’s going on behind the scenes—an army of gnomes springing into action to right the pins and run each ball back to the carousel. In a less fantastical sense, this is not far from a conception of automation by someone like Edwin Hutchins (1995, 2003). But rather than discussing fanciful causalities that we might assign to technical artifacts, Hutchins instead considers the heterogeneous skills, knowledge, and actions that become so embedded into our technologies that they appear to function magically. In examining how information is summoned, transferred, and stored in technologically mediated work settings, from naval bridges (Hutchins 1995) to airline cockpits (Hutchins 2003; Hutchins and Klausen 1996), he documents how people seamlessly offload information to the built environment around them and outsource mental labor like sorting, monitoring, or numerical calculation to machines. But where such automation is concerned, Hutchins looks beyond the gadgetry to its architects. For him, asking a computer to run an algorithm is not abdicating human decision making, but is in fact a matter of summoning all the human labor, knowledge practices, and decision making that went into the creation of that algorithm—what Latour (1988) refers to as delegation.3 When, like Hutchins, you think of automated artifacts as calling forth the human institutions that gave rise to them, each time you add a column of numbers in a spreadsheet application it’s almost as though dozens of software developers, product managers, and interface designers spring into action behind the scenes, like so many gnomes arranging bowling pins, to come up with a sum and present it to you in a beautiful Calibri font. This metaphor is perhaps fanciful, but it is also a helpful sort of lens for calling into focus both the agency of artifacts and the agency behind them. In the same way that reading a book might be thought of as an odd form of conversation with the author, living or long dead, in Hutchins’s

Tracing the Path  44

perspective our use of technologies connects us across time and space to their human architects. Similarly, Hughes (2004) emphasizes the manner in which our technologies are no longer simply responses to or means of manipulating the natural environment, but are increasingly responses to and means of manipulating a world that is already “human-built,” shaped in part by nature, but equally by existing technological infrastructures and layers of legal, economic, and organizational practices. Where such shaping of action through infrastructure is intentional, it is what Joseph Turow (1992) refers to as an exercise of structural power. Scholars like Bryan Pfaffenberger (1992) or Lucas Introna and Helen Nissenbaum (2000) suggest that such a way of thinking highlights the manner in which the various political, scientific, legal, technical, and economic arrangements that went into the initial construction of a tool are continually replayed, compounded, and reified each time it is used, or conversely challenged each time it is circumvented or appropriated for an unanticipated use. Consider users who found a way to watch TV shows from Hulu using Boxee’s home entertainment software while Boxee was banned from Hulu (Braun 2013), who “jumped” the New York Times paywall using one of the various workarounds published by angry readers (Garber 2011), or who watched pirated streams of live sports online rather than paying for cable (Ernesto 2009; Kravets 2011; Sandoval 2011). This sort of continual realignment and tug of war between those who would regulate through a technological system and those who would use or configure it differently (or abandon it altogether) is what Pfaffenberger (1992) calls a technological drama, emphasizing both the political and dialogical aspects of technological use and development. Importantly, the notion of the technological drama, as opposed to ‘‘text,’’ is intended to capture the ongoing tensions and stakes presented by these technologically mediated relationships and the dialogical nature of the systems being enacted and contested. It also denotes the performative nature of our interactions around and through technologies—how they become props in dialogues that involve an array of legal, commercial, and discursive channels. In other words, as Pfaffenberger (1992) asserts, the way technologies are designed and framed, then subsequently reconstituted, reframed, and worked around by their users, matters deeply. Such a characterization helps us to understand the deployment of technologies on the one hand, and their circumvention or appropriation by users on the other, as being deeply enmeshed with broader forms of public discourse—acts that are

Mapping Sociotechnical Systems  45

equal parts technical and symbolic, and that frequently encompass claims about who should have access to information and on what terms. Heterogeneous engineering and tugs-of-war are all around us in the world of online television news distribution, and over the course of this book, as we examine the route that content takes on the way to us, we will see the influence of competing, or more accurately “interacting,” systems that are both inscribed in the tools of the trade—content management systems, video players, analytics engines, and so on—and traced out on the map of that route. THE EXAMPLE OF TIME DELAY

Time delay is a powerful example of how online video’s route is influenced by the various forces and architects in play in online and social television news distribution. On the one hand, audiences will rarely see a live stream of a national TV news broadcast on the web. This is because the owners of the corresponding news divisions still make considerably more money selling their programming to cable providers, who often pay a retransmission fee to carry even those news shows that are “freely” broadcast over the air. As Charles Tillinghast, former president of MSNBC.com, put it in a 2010 interview, “The MSOs [multiple system operators, colloquially known as cable companies] aren’t too keen on having channels that they pay fees to run appear on the website for free.”4 Cable and network news operations, then, are beholden to their corporate parents, who in turn don’t want to anger cable providers by giving away online, for free, content for which cable subscribers would otherwise have to pay.5 Even so, you’ll often see the entirety of these same shows posted for free to the web as little as ten minutes after they air. This is because when they’re not available from the source, popular news clips quickly get bootlegged to YouTube and other video sharing sites; users have become highly adept at getting hold of the content they want irrespective of the availability of an authorized source (Newman 2012). During my time at MSNBC in New York, web producer Will Femia noted the importance of “getting it out first.” If a television outfit doesn’t upload its own embeddable clip to the web as quickly as possible, much of the online traffic will go to unofficial copies of a segment, rather than to the news provider. This explains in large part why MSNBC.com developed and released its own embeddable player (Plesser 2008a). When embeddable clips were available to users in a timely fashion, embeds made up a significant portion

Tracing the Path  46

of MSNBC.com’s video traffic—as much as 10 percent in 2011 (Plesser 2011b). There are thus competing legal and economic forces that determine when video of television news goes online—it cannot be posted live, or it will ruin lucrative cable distribution contracts, but it should be as close to live as possible or it will conversely ruin the online revenue stream. And so online distribution of television news clips neatly threads a needle between the recalcitrant interests of the cable companies on the one hand and the equally recalcitrant interests of web users on the other. Both parties can be valuable resources lashed together by the news providers to create a profitable operation, but neither is so agreeable that it lets the provider do whatever it wants. In the remainder of this chapter, I expand on the case of placing MSNBC television content online to provide a more comprehensive example of how distribution can be understood as heterogeneous engineering, as well as lay the groundwork for further discussion. To convey an idea of the distribution system that MSNBC constructed for online television news in the years leading up to 2010, I’ll trace the path taken by a typical cable news program on its way to an online audience, introducing the various actors, technologies, and organizational components enrolled in—and pushing back on—this system along the way. Importantly, I am defining “distribution” as the movement of content from the time it is produced to the time it reaches the consumer, an analytical definition that opens up both content’s movement inside and outside an organization to scrutiny. In short, I will map out MSNBC’s online video volta, unpacking the heterogeneous elements—whatever form they take—that impact the route its television news products take to us online. While I begin this process in the remainder of this chapter, following this path where it ultimately leads and unpacking the sociologically rich relationships we encounter along the way is a task that will extend throughout the book. FLIPPING VIDEO

While news programs may place a variety of content on the Internet beyond just televised clips, any discussion of online television distribution understandably begins with the manner in which “traditional” TV content is made available online. The online distribution process at MSNBC would begin when a web producer, based in New York and affiliated with MSNBC .com, watched an MSNBC television program as it aired. During this time,

Mapping Sociotechnical Systems  47

she or he would make notes concerning which portions of the show should constitute an online clip. Generally the entirety of a show would be made available online in the form of clips, but just where the division points between these clips should be was a decision made by the MSNBC.com web producer in tandem with a web video editor in Redmond, Washington. Commercial breaks provide “natural” break points in a program for television producers, and most MSNBC cable shows are produced in individual “blocks,” which the producers sequentially letter A through F, each corresponding to a span of the show between commercial breaks.6 Show blocks produced for television sometimes work as individual online segments as well, but web producers deciding how to split a whole show into clips would generally choose to divide the program up even further if there were logical transition points that came between commercial breaks. As a web producer decided how a program should be divided into clips, she or he would make a running list of these clips, noting when each should begin and end. The same producer would then decide on a headline and a caption for each one, where necessary referencing the script and chyron graphics from the show to ensure that names were properly spelled and that captions offered appropriate context. This list would then be emailed, along with any necessary additional instructions or information, to the video editor on the Microsoft campus in Redmond, Washington, who would also have simultaneously recorded the show from a live satellite feed.7 In principle, the editor in Redmond would then cut the show up into the clips specified by the web producer in New York. In practice, the editor would usually cut the show into clips as it aired as a time-saving measure, so all that remained to do after the show was to add the headlines and captions, along with other metadata. Discrepancies between the clips decided on by the web producer and those cut in advance by the editor would then be reconciled over email. Finally the editor would convert the clips to file formats appropriate for streaming using a piece of software from Telestream, Inc. called FlipFactory. The utilization of FlipFactory to transcode televised video for use online led to the parlance across MSNBC of “flipping a video” to refer to the process of putting a video clip on the Internet. Videos that had been transcoded were said to have been “flipped.” The video editor in Redmond would watch the resulting clips for quality control purposes, lest any corrupted audio or video make it through the transcoding process. When the clips were deemed satisfactory, a database entry called a “tease” would then

Tracing the Path  48

be created for each of the transcoded videos. The video editor would select and edit a thumbnail image from each clip that would appear when a video was static and wherever else a still image would be needed online to represent the clip. These images were added to the teases, along with the headlines and captions provided by the web producer in New York. The metadata added to the tease would also include the clip’s source (generally the show it came from), and any relevant topical categories to which the clip should be assigned (“politics,” “business,” “entertainment,” and so on). Finally, the video was published to the web, at which point the clips became available online, though not yet in the sense you might expect. Rather, at this point the streaming video files were in place on MSNBC .com’s web servers, ready for use, but not yet published in the sense of being placed on the site’s webpages, playlists, and other public-facing channels. PRESSURES ON THE SYSTEM

How has the path thus far been sculpted by heterogeneous engineering? What needles have been threaded by the movement of video along this part of the distribution chain, from a televised show to a collection of clips ready to be put in front of online audiences? At first glance, creating copies of a television program for streaming and download would seem to be a simple, contained affair. After all, even before the FCC-mandated switch to a digital high-definition standard in 2009, both television production and distribution were increasingly digital enterprises. Moreover, there are much smaller entities, from local TV stations to lone entrepreneurs who make podcasts and web video, that navigate this process in a comparatively uncomplicated fashion; in fact, by 2010 NBC Universal itself was running numerous websites, including the video-intensive NBC .com, and several NBC News–affiliated current affairs sites like Education Nation and the Grio, whose video operations depended little or not at all on Microsoft or its Redmond offices. How, then, did putting MSNBC cable shows online come to involve coordination between employees and computer networks in New York City and the Seattle suburbs? MSNBC began in December 1995 as a joint venture between NBC (later NBC Universal, and currently NBCUniversal) and Microsoft to create a co-branded website and cable news channel. The deal would lessen the cost pressures on NBC’s news division by making the network’s investment in NBC News bankable across more hours of television (Collins 2004; Furman 2002; Harper 1998). It would also allow NBC to compete with CNN, which had stolen away ratings and viewers during the first Gulf

Mapping Sociotechnical Systems  49

War (Collins 2004; Taylor 1996), and give the company an aggressive foothold on the web, which at the time was clearly of growing commercial importance—and where CNN was also already establishing a highly competitive presence (Taylor 1996). Microsoft, which was enjoying meteoric profits in the wake of its Windows 95 release, was looking for ways to sustain its rapid rate of growth, which meant continuing to expand into industries beyond software (Harper 1998). The online media business seemed potentially lucrative as an area for such expansion, and Microsoft invested millions of dollars into the launch of the Microsoft Network (MSN) dial-up service as part of Windows 95 (Lewis 1996)—though as Bob Sullivan, an MSNBC.com staffer recalled, Microsoft initially didn’t foresee the Internet as being a part of this strategy: I was actually hired to work on something called MSN News . . . out in Redmond, . . . the precursor to MSNBC. At the time [of] the Windows 95 launch, the MSN network was akin to the AOL network, which was its own totally separate dial-up, closed community. You couldn’t get to the Internet from AOL and you couldn’t get to the Internet from MSN. My first few months there I was dialing up computers at the University of Missouri to get on the Internet. At Microsoft. To show people the Internet was pretty cool and they should think about it. And there was a time where the guy who was in charge of the whole thing . . . [at] a big company meeting, said, “MSN will never be on the Internet.” And in fact, “We’re better than the Internet. So the Internet’s coming to us.” That’s been a profound lesson to me over and over again, because basically closed systems just don’t work in our world. By the end of 1995, however, “the exploding popularity . . . of the web [had] compelled Microsoft chairman Bill Gates to rethink the company’s on-line strategy” (Taylor 1996). To put the situation in perspective, while the MSN dial-up network accumulated over a million subscribers in its first eight months of operation, the number of Americans with access to the web grew to 15 million during this same period (Lewis 1996). This put Microsoft in the position of playing a costly game of catch-up. As New York Times contributor Peter Lewis noted in 1996, “Microsoft spent millions to build the Microsoft Network before abandoning it for the Internet model late last year. The transition forced more than 200 companies [inside and outside Microsoft] that had been developing products and services to in essence start from scratch to deliver web-based products.” As part of its transition to web media, the company struck its deal with

Tracing the Path  50

NBC in December of 1995. Over the course of the following year Microsoft also launched the online magazine Slate.com and made MSN News freely available on the open web in preparation for its transformation into MSNBC.com (Taylor 1996). The remainder of MSN’s media content subsequently migrated to the Internet as well (Lewis 1996), forming a family of MSN websites—including MSNBC.com—that still exists today.8 What’s more, Bill Gates and his executives at the time saw some form of interactive television as likely to be the next major media market following the web, and they viewed a partnership with a major television company as essential to competing in this arena (Collins 2004).9 Microsoft made abortive inquiries with the BBC to this effect, then aggressively pursued a partnership with the Turner Broadcasting System (TBS), which could have resulted in the two companies teaming to buy CBS (Collins 2004; Weiner and Stalter 1995). Prospects of a Turner-Microsoft partnership fell through, however, upon the sale of TBS to Time Warner, after which Microsoft was courted by, and ultimately signed a deal with, NBC to create MSNBC (Collins 2004). “Convergence” was becoming an industry buzzword by the mid1990s, to the extent that, as Seattle Times correspondent Chuck Taylor (1996) put it, “Conventional wisdom foresees a single medium combining television and computers. Microsoft and NBC say they want to be ready for this.” While in the intervening years online firms have indeed taken on roles once associated with television providers (Lotz 2007; Tryon 2013), talk of convergence should not be overstated, particularly in the present context. From their inception, the website MSNBC.com and the cable channel MSNBC remained largely distinct enterprises in many ways. Christopher Harper (1998), who briefly chronicled the launch of MSNBC in his book And That’s the Way It Will Be, noted that despite sharing a brand identity, the television and web companies were run largely independently of one another—an observation shared by a number of other journalists who also looked at the origins of the joint venture (Collins 2004; Taylor 1996). The cable channel grew out of NBC News and one of NBC Universal’s existing cable channels, America’s Talking (which MSNBC replaced), locating its studios in Secaucus, New Jersey, while MSN News became MSNBC.com and remained headquartered on the Microsoft campus in Redmond, Washington (Taylor 1996). Though the website maintained basic information about the cable shows, including transcripts and selected audio clips, the bulk of its content remained largely independent of what viewers saw on television. According to Scott Collins (2004), who chronicled the early history of

Mapping Sociotechnical Systems  51

the joint venture, while the startup costs of both the website and the cable channel were shared across the two parent companies, Microsoft largely viewed the task of launching and developing a working formula for the TV channel as NBC’s responsibility and, similarly, NBC saw the website as a job for Microsoft.10 By many accounts, this became the persistent position of the two companies. In 2005, for instance, New York Times correspondent Bill Carter reported that “NBC executives have said for years that Microsoft [has] never had any editorial control of the [cable] channel.” The idea that the website would be a separate entity with access to, but not necessarily a focus on, MSNBC television content was reflected in the financial terms of the original partnership agreement, in which Microsoft assumed a client role with respect to television, agreeing to pay $20 million annually, over and above its initial investment, for the rights to use MSNBC and NBC News television content online (Collins 2004). What’s more, by some accounts the terms of the joint venture were more favorable financially to NBC than to Microsoft—a topic that became a source of tension between the two corporations (Carter 2005; Collins 2004). In 2001, Microsoft’s chief executive, Steve Ballmer, went as far as to state publicly that if his company had known in 1995 what it did six years later, its executives would not have agreed to start the cable channel at all (Carter 2005; Collins 2004; Lieberman 2001). Part of this expressed malcontent may have stemmed from a perception of inequalities in the terms of the partnership, but there were also technical and market elements to it. Remember that Microsoft anticipated being able to make money from MSNBC.com partly by turning it into a portal for interactive video. As USA Today’s David Lieberman (2001) put it, Ballmer “envisioned offering news video to a country hooked on broadband,” but when MSNBC.com launched in 1996, broadband penetration was very low, and industry observers at the time likened attempts to stream online video to pumping peanut butter through a straw (Harper 1998). Unfortunately for Redmond, the U.S. market and infrastructure both proved recalcitrant. In 2001 broadband access was still relatively sparse and progressing far more slowly than Microsoft had projected, making the terms of the joint venture less favorable to the Redmond company than they might otherwise have been. These various discontents came to a head in December of 2005, when Microsoft quit the television side of the joint venture, selling its stake in the cable channel to NBC Universal (Carter 2005).11 The MSNBC joint

Tracing the Path  52

venture remained in the form of the website (which continued to pay a license fee to NBC Universal), but the television and web operations were now formally separate, driving home the dot-com’s role as the online distribution channel for NBC News and MSNBC cable. In the interim, U.S. broadband penetration—which was 9 percent at the time of Ballmer’s 2001 statements, and still less than 50 percent when Microsoft quit the television venture in 2005—has finally reached a majority of Americans. Department of Commerce figures showed that 68 percent of U.S. households had a broadband connection by October 2010, and that this group made up nearly all of the 71 percent of American households with any form of Internet connection (National Telecommunications Information Administration 2011). By that point the market for online video had become enormous and was growing. According to the market analysis firm comScore, during March 2011 alone, 83.5 percent of the U.S. Internet audience watched online video at an average of 14.8 hours per viewer (comScore 2011). Pew figures from 2014 are somewhat more modest but also highly impressive, suggesting that “more than six in ten U.S. adults now watch videos ­online—and roughly half of those, 36% of all U.S. adults, watch news videos” (Olmstead et al. 2014). As online viewing rose, advertisers increased the amount of money they spent on online advertising (BrightRoll Research 2011). While Microsoft had begun losing money on the media business generally, this growing market made the once-onerous license fees it paid to NBC for video news seem, once again, like a rational investment. Moreover, Microsoft’s contractual rights to distribute NBC News and MSNBC TV video online extended through the year 2094 (Carter 2005). In 2009, when the veteran distribution company Comcast acquired NBC Universal, it no doubt found Microsoft’s lock on online distribution of NBC News and MSNBC TV content to be inconvenient if not aggravating. It would eventually arrange to buy out Microsoft’s remaining shares in the MSNBC .com and dissolve the joint venture altogether. But long before Comcast began measuring the drapes, MSNBC TV and MSNBC.com had already grown into highly distinct enterprises and even ran on very different computer infrastructures. Show segments at the cable channel and NBC News were cut together using nonlinear digital video editing systems from Avid, which provided not only basic video editing hardware and applications, but also software for script-writing and internal communications among editorial staff, as well as a full line of spe-

Mapping Sociotechnical Systems  53

cialized post-production programs—all linked together via the NBC News intranet. From the time that the first words and graphics for a script were created, to the time that the resulting segment was cut for air, to the time the anchor read the introduction off the teleprompter, content lived within this ecosystem of Avid software. Afterward, it was archived in NBC’s digital archive system, known as Ardome. These software tools and computer networks used by staff at NBC News and MSNBC TV were different from, and developed largely independently of, the extensive collection of ones employed by MSNBC.com, resulting in what one web producer referred to jokingly as a host of “bridges, demilitarized zones, and virtual desktops” necessary for getting the two systems to talk to one another. The route that MSNBC cable shows took to the Internet was influenced by the interaction of a number of powerful heterogeneous engineers. Executives at NBC and its erstwhile owner, General Electric, wanted a twentyfour-hour cable news network in order to reduce the cost pressure on NBC’s news division, keep viewers from moving to an insurgent CNN, and develop an Internet presence in time to be competitive in the growing online news market that its rivals were also entering. Microsoft executives saw a partnership with a major American television company as a necessary step in what it perceived to be a potentially sizable market for interactive television, and welcomed an outside investment that could help the company to retool its online media strategy. Each lashed together and made use of its existing, geographically situated resources—Microsoft built and staffed an MSNBC website in large part with the resources of MSN News in Redmond, while MSNBC TV built out the resources of NBC News in New York, and took over the studios, satellite transponder, cable providers, and cable subscribers of NBC’s defunct America’s Talking channel in Secaucus, New Jersey.12 By 2010 the movement of video from television to the web necessarily traversed a route that reflected the original interests, financial incentives, competencies, and contractual obligations of the involved parties—NBC would make television, and Microsoft would put it online. The distinct content management systems and technical infrastructures of NBC News and Microsoft also reflected these different organizational foci and influenced the route that video content took to viewers. Rather than being pushed to the Internet directly from NBC’s digital video archive, MSNBC cable content was generally broadcast via satellite, re-recorded in Redmond, then divided into clips and assigned metadata

Tracing the Path  54

negotiated between MSNBC.com’s staffers in New York and Washington, and finally entered into the separate content management system at Micro­soft. This path, again, bears the impression of myriad heterogeneous actors lashed together by the groups involved and traces out on the map the solution available to MSNBC for putting video online. Moreover, we’ve come to see that MSNBC and its digital distribution strategy are not the product of a single system builder or heterogeneous engineer, but of multiple system builders.13 The video clips we’ve been following have still yet to be linked to, embedded, or otherwise presented to the world, but already MSNBC’s “video volta” has covered three thousand miles from New York to Redmond, Washington—a route that indeed “traces on the map” the solution available to the parties involved with MSNBC. Once the segments for a show had been flipped, the video editor in Redmond would send an email to the web producer in New York advising that the clips were available and providing the final metadata from MSNBC.com’s content management system (CMS, its online publishing platform) for each of the newly published videos, including the URL, ID number, story slug, and GUID.14 The web producer would use these to help place and promote the segments, and might also decide to request additional versions of a clip from the video editor if, say, he or she wanted to make a highlight from a show segment available as its own stand-alone clip alongside the full version. But once the clips were available online, what forces and concerns shaped the paths they took from there? The brief, but not terribly useful answer is: many, many places. For a more valuable discussion of where videos went next, it will help to organize the discussion a bit by examining distinct distribution paths for video in turn. First, video reached audiences via MSNBC.com’s web video player, a highly flexible and modular player that appeared in numerous contexts throughout MSNBC.com’s website. The construction, the character, and the significance of this conduit to audiences will be traced out in Chapter 4. But online distribution of MSNBC television clips didn’t stop at its own website. As we proceed through this volume, we will explore more fully the array of intermediaries involved in embedding and syndicating MSNBC content to other sites across the web, as well as pushing it across our continually expanding array of Internet-connected devices.

4//to turn it on and have it play the msnbc.com video player

There is an undeniable irreverence and playfulness in much scholarship on technology. It’s an impish impulse that sociologists and historians of technology share with some contemporary sculptors whose sense of fun involves provocation and thumbing one’s nose at widely held definitions of art and design. Like many, I once found it hard to understand how Marcel Duchamp’s signed urinal or Jeff Koons’s neatly organized ensemble of upright vacuum cleaners were worthwhile pursuits, let alone sculptures or works of art (or “anti-art” as such provocateurs thought of them). But if one point of art is to get people to see the world around them differently, then surely these things qualify. Nailing a found object like a Pepsi can to the wall in a gallery may require laughable effort as compared to Monet’s Water Lilies, but this simple act transforms something utterly banal into an object of interrogation. Passersby who routinely toss out hundreds of identical cans each year with the other remnants of their lunches will pause—if only briefly—to inspect, consider, and reencounter the object they otherwise take for granted. Surrealist André Breton described found sculpture as involving “manufactured objects promoted to the dignity of art through the choice of the ­artist” (quoted in Arnason 2004, p. 248). It is precisely the question, “Is this art?”—a forced reckoning over whether the object in itself has politics or is making a statement—that causes us to re-evaluate the mundane artifacts that wallpaper our existence. In the world of technology studies, Bruno Latour’s (1996) use of “quotes” from an inanimate metro train, Law’s (2002) scholarly exegesis on 55

Tracing the Path  56

the design of an aircraft wing, and Trevor Pinch’s (2010b) ruminations on invisible dog fences are similarly attempts to open to interrogation mundane or taken-for-granted artifacts, along with the conventions surrounding them, by putting them on display. Some of the best accounts of technology and culture are those imbued with this spirit of anti-art. It’s with this sensibility that I wish to examine the MSNBC.com video player, which for years formed the primary conduit through which NBC and MSNBC news video reached users and audiences online. The player, which delivered millions upon millions of video clips to audiences in the United States and across the globe, is an object that many readers will have personally encountered. Recognizing the player as a portion of our developing “video volta” and dissecting its seemingly mundane features will help to clarify the complex tangle of relationships, interests, and motivations “invested and contested” (Kompare 2010) in the design of the everyday objects used in online distribution. When we left off tracing the path of video in Chapter 3, video files of MSNBC television segments had been placed on Microsoft’s web servers, ready to be streamed to audiences, but as yet not visible to them. The MSNBC.com video player would help to finish this job, but to understand it, we must first understand its role in the larger array of players and tools through which MSNBC video reached users. Video syndication was a large portion of the strategy by which MSNBC.com became one of the largest video providers on the web. Recall that MSNBC.com was part of Microsoft’s MSN network, a family of websites that was only very recently disbanded. When a video was entered into MSNBC.com’s content management software, by default it was syndicated to all of these other websites, posting automatically in some places, and in others becoming available for the respective editorial staffs to add to their offerings.1 Videos were also syndicated automatically to Microsoft’s Bing video portal, where they were featured prominently as the site’s main source of news video (see Figure 4.1). Syndication of all these clips to Microsoft properties was automated, as was the distribution of some NBC News programs to the online video portal Hulu. When a video editor, or any other person entering a video into MSNBC.com’s content management software, input the metadata for a clip, the topical categories they chose for the video included ones that, when selected, would automatically feed the video out to these other outlets. When viewed on another site, like an MSN page or Bing, clips ap-

To Turn It On and Have It Play  57

Figure 4.1. The Bing Video news page was dominated by MSNBC TV and NBC News content, which make up the first two rows of thumbnails in this screenshot, as well as the majority of videos in the featured player. Used with permission from Microsoft.

peared in MSN’s video player, rather than MSNBC.com’s. Clips on Hulu likewise appeared in Hulu’s video player. Anyone who has watched streaming video on the web is familiar with video players—those rectangular widgets appearing on webpages that play back clips, while allowing the user simple affordances like pausing the performance or raising the volume. Unlike many video players that consist of software purchased off the shelf from white-label companies like Ooyala or Brightcove and branded with the logo of the content provider, MSNBC .com’s video player was developed by and for the company itself. This by itself is interesting. Given the availability to MSNBC.com of an established video player developed by Microsoft for all its MSN sites, to say nothing of a variety of off-the-shelf solutions available from thirdparty vendors, it is worth exploring why MSNBC.com developed its own software for playing clips, rather than making use of available technologies and conduits for video. It merits reiterating here the importance of syndication to MSNBC’s

Tracing the Path  58

web presence and how MSNBC.com was able to dramatically increase the reach of its online video by pushing clips to other web properties in Micro­soft’s online services division, like Bing and MSN. Bing, for instance, was the official web search provider for Facebook for some time, as well as for Blackberry smartphones (Paterno 2011; Segan 2011). MSN.com was the default homepage for the market-leading Internet Explorer browser (Nayyar 2011), and the site not only showcased MSNBC video itself, but also included a Bing search bar and links to other MSN sites, which in turn featured MSNBC video.2 Internet users would encounter clips from MSNBC cable and NBC News not just on MSNBC.com itself, but across a wide array of Microsoft-owned and partnered sites across the web. As measured by comScore, which grouped Microsoft’s various web properties into a single entity, this network of sites consistently placed the company among the top ten online video providers, competing nimbly with—and often surpassing—giants like AOL, Viacom Digital, Tremor Media, and Hulu (Trout 2011). What’s more, as we’ve seen, MSNBC cable and NBC News video were available through some of these competitors. Stokes Young, MSNBC.com’s director of multimedia, noted that some of these arrangements appeared more natural than others to the system builders involved. For example, Bing Video grew out of MSN Video, of which MSNBC.com was a founding partner. This, along with joint ownership by Microsoft, meant that Bing Video and MSNBC.com had something of a revolving door when it came to developers and other staff. As Young put it, “We’re half owned by Microsoft, and we’re the exclusive news provider to MSN. So we syndicate our stuff to Bing. . . . That’s a natural thing. I don’t think there’s any negotiation—that’s just what we do.” As for when MSNBC TV or NBC News video should be syndicated to other outlets, as Young put it, “that’s essentially a business decision,” made based on “whether we think syndication there will yield not only a good audience and some economic return, but also that it’s not overly threatening in any way to what our core business is. We’re always willing to syndicate certain parts of our content. But we’ve never wanted to treat NBC News video content as a commodity, because it’s not. It’s something that’s a very rare asset. And so we treat it that way.” In conceptualizing the path of online video as a volta, we can recognize that automated syndication represents a set of decisions at various points about an appropriate flow of content amid competing pressures, which is ultimately inscribed on the map. Young’s discussion puts on display the

To Turn It On and Have It Play  59

Figure 4.2. Doonesbury from February 27, 2011. Cartoonist Gary Trudeau parodied the extent of NBC’s cross-promotional activities around the news. ­DOONESBURY © 2011 G. B. Trudeau. Reprinted with permission of UNIVERSALUCLICK. All rights reserved.

numerous concerns that impress themselves on this path. Bing was a “natural” outlet owing to the shared ownership, staff, branding, and resources of MSN and MSNBC.com. Elsewhere online video syndication was weighed carefully on a caseby-case basis, charting a course between various competing considerations. Paths were forged based on the potential of a venue to capture valuable audiences and their associated advertising dollars, but the system builders also had to avoid giving too much material to sites that could potentially become competition. And while the temptation might have existed to syndicate to every revenue-generating platform, the company was cognizant that “commoditizing” NBC News and MSNBC products through rampant cross-promotion across platforms could damage the value of its brand if the news division were perceived as anything but financially disinterested. (For an example of the sort of reaction MSNBC.com was trying to avoid, see Figure 4.2). Over-promotion was a risk that MSNBC.com ran even within the MSN network, and its executives needed to balance this hazard against the corporate and human resource ties that otherwise made Bing

Tracing the Path  60

Video a natural fit. Syndication, in other words, involved drawing together and orchestrating many heterogeneous resources, as well as managing the pressures they exerted on the system. GOING ITS OWN WAY: MSNBC.COM BUILDS A VIDEO PLAYER

Understanding the comfortable fit between Bing and MSN on the one hand and MSNBC.com on the other is important, but it initially makes it even more puzzling as to why MSNBC.com developed its own video player independently of MSN. The specter of too much cross-promotion, for example, could have been avoided simply by branding the existing MSN player with a different logo and appearance wherever MSNBC content ­appeared—at one point, in fact, MSNBC.com did employ the MSN video player. The issue, as it turns out, was that previous versions of that player were, by some accounts, not user-friendly or reliable, and had come to be seen by MSNBC.com as a liability in a market for news video where user experience was increasingly viewed as a differentiator. This risk was exacerbated by the fact that MSNBC.com, as a joint venture with NBC Universal, was responsible to more parties than just other MSN properties for maintaining a healthy bottom line. Randy Stearns, who worked at ABCNews.com before becoming the East Coast deputy editor for MSNBC.com, was involved with the project to develop the initial version of MSNBC.com’s custom player and recalled, “Once we built our own player to our own specs to do the things we wanted, number one of which was just play video—seems pretty s­ traightforward—that changed everything.” To construct the video player, MSNBC.com assembled the first large-scale cross-functional team in the company’s history, involving staff from MSNBC.com’s editorial, marketing, sales, and product management staffs, along with developers from around the company. Stearns recounted that the team’s charter had three main points: “Our goals [were] to create a video player that (1) plays video, and that (2) plays video all the time, and (3) plays video that can be monetized. That’s all it was. Everything else was completely tertiary. . . . Then we built up a lot of other, fancier things around [the player]. But the charter was just something as basic as that.” The reason for the elementary, almost tautological nature of the team’s charge—build a video player that plays video—was that the formerly stateof-the-art MSN video player on which MSNBC.com then relied had begun

To Turn It On and Have It Play  61

to show its age. Compared to other contemporary and emerging products at the time, including a beautiful new player under development at ­ABCNews.com, MSN’s player was beginning to seem finicky and outdated, both in its appearance and its behavior. “There was a competitive factor as other players came to the market with reliable, feature-rich video experiences,” recalled Stearns. Some of these, like Brightcove’s player, were built by vendors and available for purchase, but as Stearns noted, “the business deals around that were tricky,” and from MSNBC.com’s point of view, not all such players offered a sufficiently competitive user experience to justify the investment. Ultimately, the MSNBC.com team’s seemingly elementary charge reflected the need to create a simple, foolproof delivery mechanism at a time when playing video over the Internet had only recently been a relatively complex task. Not long before, streaming video had been accessed using client software like RealPlayer that ran outside of the web browser, and even as browser-based video became more widely available, getting it to play often meant ensuring you’d chosen the right browser and had the right plugins installed. The thinking at MSNBC.com, said Stearns, was that “This needs to be like television. You need to turn it on and have it play, and not think about whether the rabbit ears are adjusted right or whether you’re using the right version of a browser for this particular [clip]. If the consumer has to think about that stuff, you’ve lost them. So [we need to] make sure that we can catalogue, search, and deliver the video that people want. That was really it.” The MSN player the company was abandoning did eventually evolve to offer a similar “turn it on and have it play” experience to users. But the legacy of this split was that development of the MSNBC.com and MSN/ Bing players remained independent of one another—a splintering that was reinforced as many teams within MSNBC.com began to enroll and depend on the new software. In Stearns’s description of the development of the MSNBC.com player, we can see the competitive environment into which the player was to be deployed and how this affected decision-making about the player’s functionality. We also learn how its evolution was partly a response to existing artifacts, like Brightcove’s video solution, and how marketplace and contractual legal concerns turned Brightcove from a potential partner to a potential competitor in the online video space. Finally, we see the socially constructed aspects of technological development (Kline and Pinch 1996;

Tracing the Path  62

Pinch and Bijker 1984)—the player was born of and ultimately bore the imprint of many different constituencies within MSNBC.com, which both indicates the heterogeneity of the actors involved in the development of the artifact and reminds us that large media organizations like MSNBC. com are not as monolithic as they may appear from the outside. At the same time, it’s a lesson in how technologies become more monolithic—the construction of the original MSNBC.com video player involved a largely unprecedented gathering of teams from across the company, a successful kludging together of talent and input from many corners. But if the player was born of a novel experiment within the organization, once it had been enrolled in the systems of these many teams, it became an artifact threading the needle between all their diverse interests—overdetermined, as it were, by the provincial needs of myriad system builders. THE VIDEO PLAYER INTERFACE

Besides “just play video,” what did the MSNBC.com video player ultimately do? If, as we’ve seen, it was a product of heterogeneous engineering, is there more to be said about the disparate social and technical concerns it drew together and the various constituencies’ needs that became “invested and contested” in its design (Kompare 2010)? To answer these questions, we’ll need to dig further into the actual interface of the player, as well as the other artifacts that made it up, such as the video playlists that populated it with clips and the various links from which users launched the player. The display of the MSNBC.com player was what software developers call “object-oriented,” meaning in this case that its presentation was highly flexible, allowing it to be displayed in a number of forms.3 On a “story page,” for instance—a webpage containing information related to a particular news item—the player and video might appear on the page alongside a photo essay and a printed story on the same subject. On a “section front”—a page displaying the headlines of various stories in a given category, such as politics or health—the link to the video might appear as a thumbnail or a headline that, when clicked, launched the video in a “standalone” version of the player, which appeared in a pop-up window. And on the homepage for a particular show—generally referred to at MSNBC as a “show front”—the video player appeared prominently as the centerpiece of the page (Figure 4.3).4 The player also had a variety of skins, appearing with a differently branded backdrop depending on whether it was playing

Figure 4.3. A “show front.” Note the dominant video player and the “grid” of links and headlines that appears at the bottom of the page. NBCUniversal.

Figure 4.4. The standalone video player. Note the template specific to the show from which the clip is drawn, as well as the various news categories. NBCUniversal.

Tracing the Path  64

a Today Show clip or a segment from The Last Word, for example (see Figure 4.4). PLAYLISTS AND LAUNCHER LINKS

The video player was accompanied by other artifacts that enhanced its flexibility. For example, wherever a player appeared, it would be assigned a playlist that would begin with the clip requested by the user and afterward keep rolling related videos. This list of upcoming clips was also displayed in the player as a series of thumbnails and headlines, allowing the user to skip ahead to those videos she was interested in. In some cases, this playlist was generated automatically. The standalone video player, for instance, had a navigation sidebar from which users could select any of a variety of general news topics, such as “U.S.,” “World,” “Sports,” “Tech & Science,” and so on. These topical playlists were automatically populated with any clips that had been assigned to these categories when they were initially entered into the system. Playlists dedicated to other, more specific topics or news items could also be automatically generated from metadata searches and other sorts of database queries. Many such topic-based playlists were also manually generated, however, at the discretion of the site’s web producers. One MSNBC.com staffer who worked with the Today Show explained that “generally if there’s more than four, five clips on a topic, we will go ahead and make a playlist for it.” The playlists for show fronts—the video-centric homepages of the various MSNBC TV and NBC News programs—also had to be generated manually. Immediately after the video editor in Redmond finished publishing the clips for a given show, he or she would enter them into the playlist populating the video player that appeared on the show’s homepage. Similarly, lists of thumbnail links to videos—which when clicked in turn launched the video player itself—appeared in many places across MSNBC.com. In some cases, these lists of video links were populated by automatically generated feeds, while in other locations the links were manually inserted by MSNBC.com web producers. And while it’s already apparent that the video player itself was engineered to be something of a chameleon, able to appear in many different and adaptable forms depending on its placement on the MSNBC.com website, even artifacts as seemingly simple as playlists and launcher links turn out to be surprising works of heterogeneous engineering.

To Turn It On and Have It Play  65

HETEROGENEITIES SURROUNDING THE VIDEO PLAYER

The seemingly “given” features of the collection of artifacts encountered by users of the MSNBC.com video player—the player itself, playlists, and launcher links—are in fact savvy and at times idiosyncratic responses to the assemblage of recalcitrant actors surrounding the player. As with other portions of the video volta we’ve encountered to this point, they bear the influence of a wide variety of relationships and pressures. For example, given that the MSNBC cable channel and website shared a brand name, it might seem natural for MSNBC video and other content produced by the cable channel to have appeared prominently on the front page—what employees called “the cover”—of MSNBC.com. Recall, though, that while MSNBC.com assigned producers and other staff to work with the cable programs, MSNBC TV and MSNBC.com had been run as largely distinct enterprises. Moreover, the distinctiveness of the two grew over time. After MSNBC.com was formally split from the cable channel, the web venture began expanding on its own. In the period between 2007 and 2011, MSNBC .com extended its portfolio of websites far beyond its original domain. The MSNBC.com domain came to be referred to as the “Blue Site”—a moniker derived from the dominant elements of its color scheme—to distinguish it from the company MSNBC.com, which over a period of about five years had refashioned itself into the “MSNBC Digital Network,” operating a family of websites owned by the joint venture. During this expansion, the company acquired Newsvine.com and EveryBlock.com; launched an internal startup called Breaking News; expanded its content partnerships with the rest of the MSN network, as well as other vendors; and spun off the Today Show section of the Blue Site into its own domain, Today.com. There had also been a widening split in editorial strategy between the Blue Site, still the web company’s flagship property, and MSNBC cable. The cable channel began as an extended platform for NBC News and had initially espoused the same nonpartisan editorial philosophy. But the channel continually struggled in the ratings against competition from its other cable news competitors (Collins 2004; Moraes 2002). In the wake of the marked success enjoyed by the increasingly conservative panel of commentators on the Fox News Channel through the Clinton impeachment and later the 2000 elections, the management of MSNBC cable ultimately decided in 2002 that it would be necessary to shift to an opinion and commentary format in order to remain competitive in the cable news market

Tracing the Path  66

(Carter 2002; Collins 2004). At the time, MSNBC president Erik Sorenson described the strategy as a move to “op-ed” television (Moraes 2002).5 Initially, however, op-ed programming did little for the cable channel’s ratings, as evidenced by what many viewed as a highly disappointing turnout for Phil Donahue’s program after MSNBC hired him as a liberal commentator to fill a marquee primetime slot (Carter 2002; Furman 2002; Moraes 2002). The show was canceled in less than a year, but the channel remained committed to an opinion format in primetime despite continually trailing in the ratings. Then, several years later in 2006, MSNBC’s primetime host Keith Olbermann began to draw strong ratings and industry accolades with his criticism of the Bush administration and its Iraq War policy, and subsequently became established as an outspoken liberal commentator (Bauder 2011). His continued ratings success made him an influential figure within the cable channel at that time, and many of the contributors who became popular on his program later received their own MSNBC shows (Bauder 2011; Carter 2011; Farhi 2011). Former NBC president Jeff Zucker later recounted that after viewers began tuning in for Olbermann, the remaining primetime lineup was designed to flow around his program (Stoll 2011). MSNBC’s primetime programming eventually became a string of outspoken left-of-center anchors and commentators, including Ed Schultz, Rachel Maddow, and Lawrence O’Donnell—a liberal rebranding culminating in the channel’s current marketing campaign, titled “Lean Forward,” which paints MSNBC TV as a progressive news and commentary network (Lafayette 2011). This shift to partisan programming was widely credited with helping to lift MSNBC out of last place in the cable news ratings contest (Farhi 2011; Schechner 2011). Meanwhile, the Blue Site, which still hosted the online presence of these cable programs, had itself risen to become extraordinarily popular— in 2005, at the time of its split from the cable channel, the New York Times dubbed it “the most-used news site on the Internet” (Carter 2005), and in subsequent years it consistently remained among the top three most trafficked news outlets in the United States, alongside CNN.com and Yahoo! News (Stelter 2010b). MSNBC.com had achieved this success while continuing to brand itself as a nonpartisan news provider (Kaplan 2010; Stelter 2010b). By 2010, this difference in editorial strategy between two properties that shared a brand name and cross-promoted one another’s content had become a source of tension between the television and web companies. Brand-conscious executives and editors at MSNBC.com tended to worry

To Turn It On and Have It Play  67

that consumers would confuse the two properties, polluting the objective image of site (Kaplan 2010; Stelter 2010b; Wallenstein 2011). In one internal memo obtained by the New York Times, MSNBC.com president Charles Tillinghast opined that “both [the partisan and objective] strategies are fine, but naming them the same thing is brand insanity,” and that crystallizing the cable channel’s partisan identity with the Lean Forward marketing campaign “only exacerbates the brand misalignment problem” (Stelter 2010b). Behind the scenes at MSNBC a variety of plans were formulated to alleviate this predicament. A more immediate outcome, however, was that the Blue Site, in order to preserve its nonpartisan image, tended to avoid promoting MSNBC cable programming on its cover. Instead, MSNBC TV was given its own “micro-site”—a landing page with a video player and space for links located at the subdomain tv.msnbc.com. Meanwhile, the various TV shows were given their own blogs and/or show fronts, smaller sites located at subdomains like rachel.msnbc.com (for the Rachel Maddow Show) or hardball.msnbc.com (for Hardball with Chris Matthews). For the most part, these sites were designed to stand alone, drawing users through on-air promotion, search traffic, and the MSNBC TV micro-site. Some of the subdomains were linked to from the MSNBC.com cover, but they were grouped in the header with other related web properties, like Hotmail and Bing, in a way that suggested limited connection to the primary editorial functions of MSNBC.com (Figure 4.5). The show fronts for other non-primetime MSNBC cable programs, such as the Dylan Ratigan Show, were not linked to from the cover, and still others like NewsNation with Tamron Hall or Andrea Mitchell Reports had no show fronts or blogs at all, but rather single pages within the MSNBC TV subdomain. Another issue that still limits the promotion of marquee television content online, not just at MSNBC.com or its NBCNews.com successor but at many television news sites, is that the material produced for primetime television newscasts is generally an after-hours recap of the day’s events— but the bulk of web traffic to news sites happens during the daytime, when people are logging on from work, looking for the freshest stories (Boczkowski 2010). As one dot-com staffer at Rockefeller Center put it, “The web world fills this crazy gap of realtime media consumption, because before it came along it was like, ‘I know I check my news then and I check it now.’ Now people just check it all day long. You know, the tagline of

Tracing the Path  68

Figure 4.5. The MSNBC.com cover. Note the inconspicuous size and positioning in the header of links to MSNBC cable programming. NBCUniversal.

this building, for online, is, ‘Daytime’s our primetime.’ From nine to five is when we get all of our traffic. As it should be, right? I mean, you’re at work, you take a break.”6 The web version of the previous evening’s broadcast, however, is no longer breaking news; instead it’s often quite literally yesterday’s news. That not only decreases its relevance for many online news consumers, but also makes it hard to promote even when the television and web brands are well aligned. For instance, NBC Nightly News has long taken a nonpartisan reporting stance that was in many ways highly compatible with the mission of MSNBC.com. Yet the news cycle for the NBC Nightly News was heavily out of sync with the rest of the content produced for the Blue Site, which could be a problem when it came to placing the broadcast’s content. After all, the front page of MSNBC.com, like those of ABCNews .com and CBSNews.com, wasn’t a mere promotional vehicle for broadcast properties, but instead was a breaking news site with a mandate to compete with Yahoo! News, the Huffington Post, and NYTimes.com. Tossing up a fifteen-hour-old clip of the previous night’s broadcast as consumers filed into their desks in the morning generally wouldn’t help MSNBC.com attract or retain viewers looking for the freshest updates. Consequently, from the perspective of the Blue Site’s editors, only a small subset of what was produced for each evening’s newscast, such as “evergreen” feature stories or exclusive interviews, might be worth heavily promoting on the cover page.

To Turn It On and Have It Play  69

A middle strategy was to add links to television segments from the pages of stories posted on the site. MSNBC.com launched a redesign of its story pages in 2010 to be more multimedia-friendly, as well as to allow editors to be more flexible in their organization of content. In addition to the latest text of a story, editors could now quickly add and arrange embedded media objects, including, often, a video player and/or photo galleries related to a story. The newly flexible story page architecture was intended in part to allow stories to be continually updated with the latest reports, regardless of what format they might be in. Now that many media types could all be housed under the roof of a single story page, this also meant that a text story with the latest-breaking information could be supplemented with a slightly older clip. In this way, older videos could be reframed for the user as presenting additional background, rather than out-of-date information. Even so, MSNBC.com story pages still tended to feature NBC News broadcast video far more frequently than clips from MSNBC TV, most likely owing to the division in editorial strategy mentioned earlier. Neither broadcast nor cable clips were ever guaranteed a spot on a story page, much less on the cover of the Blue Site. Instead the web producers who worked with the various television programs had to pitch their clips to the cover editors if they wanted their various shows’ material to be featured on the more prominent, well-trafficked pages of MSNBC.com. A substantial part of the job of many television-attached web producers, then, was to maintain a peripheral awareness of the material that was being posted to the MSNBC.com front page, so they could be poised to take advantage of any opportunity to promote relevant clips from their shows to the cover editors. Cynthia Joyce, the MSNBC.com web producer assigned to the NBC Nightly News, described this state of affairs, “‘Ecosystem’ [isn’t the right word]. It’s more like a solar system. You have to see if there’s a star moving past and see if you can time it right. You’re just constantly trying to dovetail your content with something else that already has that m ­ omentum.” In the context of MSNBC’s larger organizational structure, Joyce’s comment foregrounds the advantage of having a player (and player launcher links) that could take on forms flexible and reliable enough to permit the placement of video in so many contexts across the site. These artifacts allowed producers to seize what were often fleeting windows of opportunity to cross-promote their video content on other portions of the site, which adhered to distinct and often divergent editorial logics.

Tracing the Path  70

This sort of architectural flexibility also extended to video playlists. Once an in-page video player or a launcher link had been placed and the user had loaded a clip, everyone at MSNBC.com would invariably like that user to consume additional video, since the more video that was watched, the more revenue would be drawn from the advertisers paying to insert video ads with each clip. This is where playlists came in, which were designed to show users more of what they were potentially interested in. As Pablo Boczkowski (2004) and other journalism scholars like C. W. Anderson (2013) have pointed out, a particular view of the user is inscribed in the design and layout of news sites and websites more generally. Moreover, designers and editors are often put in the position of considering multiple audiences and user groups simultaneously. For instance, John Battelle (2005) has noted that search engine designers must account both for users who are trying to find information for the first time (“discovery”), as well as for users who count on using the search interface as a navigation tool to easily locate and revisit webpages they’ve seen before (“retrieval”). Similarly, a video player and a playlist are constructed with multiple audiences in mind. Jesamyn “Sam” Go, who headed the team of MSNBC .com web producers in New York that put cable and broadcast content online, described three different typical user groups that the dot-com staff considered in creating a playlist, including (1) users who had seen the show on television and visited the site to re-watch or share a segment, (2) viewers who came to the site to watch the show online in its entirety because they wanted the content “on demand” or because they did not have access to live MSNBC programming, and finally (3) users who happened across clips serendipitously by browsing the site, by perusing playlists attached to other videos, or by stumbling across headlines on social media services. As Go explained, all three of these use cases were relevant and constantly salient to producers as they added metadata to videos and grouped them into playlists. To us, a lot of [the work of promoting and distributing videos] is tagging content by categories, and having those categories make sense. Because content has to make sense by itself, and it also has to live in this whole portal of video. And we need to write headlines that make sense—headlines that are attractive—because outside of the context of the original show, whatever buzz that individual segment generates [on social media] or wherever else it might be embedded, that’s the

To Turn It On and Have It Play  71

only way the video will get more attention. So we spend a lot of time doing that. Cathy and Will [MSNBC.com web producers attached to TV shows], every day they go back and forth about headlines and topics because they know . . . somebody could be watching a Nightly [News] clip on politics and see a Chris Matthews headline and consume that video. And the design of the portal is good enough that people consume more than one clip. So if a web user launched a related video from within the network, or from Twitter or the MSNBC .com cover, inevitably they would find something else they would like as well. That’s the brilliance of the design. At this point, we can begin to see why shows were broken up into clips for streaming in the first place—they allowed the flexibility necessary for targeting multiple user groups simultaneously. Users who wanted to see a complete show on-demand using the web could browse to a show front, where the player shown to them was queued to a playlist that would present the entire program in order. Meanwhile, individual clips could be grouped into playlists based on categories (such as politics) and topics (for example, a health-care bill) that could be placed with more stories across the Blue Site and would appeal to users approaching MSNBC.com with topical, rather than show-specific, interests. Sarika Dani was an MSNBC .com staffer who worked with Today.com around the time it was spun off from the Blue Site and redesigned as a portal dedicated to web content surrounding NBC News’s Today Show. At first glance, she pointed out, a news program made for television is intended to be viewed in its entirety. “It’s not meant to be clipped necessarily,” she said, “But at the same time it is, because here’s this three minute cooking segment and then we’re going to move over and change the subject, do the fashion segment. . . . Chopping up the show format for parceling out online . . . [allows us to] target the content to different audiences. You get more bang for your buck, because you’re not forcing someone to watch the whole show. . . . They can be targeted—they can get deep into one topic.” At the same time, beyond the basic automated categories to which a clip was assigned, there was a good deal of leeway in how to organize videos into playlists, which could lead to more contemplative debates among web staff as to how users look for content. Dani continued, One example is we have a “bullying” playlist: all these awful cases in the news of kids committing suicide over bullying in schools and what-

Tracing the Path  72

not. You watch one clip and you end up getting into this playlist. . . . You can see a psychologist give their take on it, or you can hear the latest news about the suicide of a teenager. But then you also lose what the show has—the serendipitous discovery of content, and you could argue some news users look for news that way. . . . That serendipity drives you deeper into exploring stuff on the site. [Alternatively, you have the example of] the fashionista who wants to keep watching ­fashion. . . . There’s a danger in over-categorizing. But there are benefits as well. You have to be aware. Harking back to the discussion in Chapter 2 concerning issues like the conditions under which publics congregate, participate in politics, and engage with a shared media culture, Dani’s observations in many ways recall arguments by scholars like Marshall Van Alstyne and Erik ­Brynjolfsson (1996), Robert Putnam (2000), Christine Rosen (2004), Markus Prior (2007), or Cass Sunstein (2007), who have debated the implications of the increasing ability of consumers to narrowly confine their exposure to information. The difference here is that playlisting was under the control of producers, rather than an “egocasting” user base in the sense outlined by Rosen (2004); there was no recommender system at work here, but rather human judgment about how to group stories in a way likely to be interesting to different subsets of users. Once again, we can see the final path taken by video is the product of heterogeneous forces—some bound together, others imposed. It traces out the solution available to MSNBC.com for presenting televised video online in the face of many complex and at times competing interests. Video was a great source of value for MSNBC.com, so it was featured prominently on micro-sites dedicated to specific shows, as well as to the cable channel more generally. Because the financial and editorial interests of the cable channel and the web venture had diverged over the years, however, with the cable channel capitalizing on opinion and commentary while the web venture fashioned itself into a nonpartisan news site, video content from the cable network also tended to conflict with the editorial and branding strategies of the website, which led to highly limited placement of cable video on the front page or on the story pages linked off the cover. Linking to video from the Blue Site’s story pages was further limited by the difference in news cycles between the primetime programming of NBC News and

To Turn It On and Have It Play  73

MSNBC TV and the daytime-focused web cover. Nonetheless, video players could appear in many places across the site and the player itself was thus designed to be as flexible as possible in terms of its appearance and manner of display, to maximize the opportunities for video placement when they arose. When and where video players did appear, video was kept rolling through the use of playlists, which were loaded and queued in accordance with the placement of the original video player. Like the organization of a news homepage, programming of playlists was itself a matter of juggling the various elements of automation, editorial control, and user preferences as these were perceived by the web staff (C. Anderson 2011). About the tensions between automation and editorial control: all of the programming for the twenty-four-hour MSNBC cable channel and NBC News was processed by a small handful of web producers. A single web producer, for instance, had primary responsibility for all the hourly shows between eleven in the morning and seven in the evening, while another handled the bulk of primetime. In this environment, automation could be an immense boon to productivity. Sam Go recounted that when MSNBC .com automated some of the processes that updated the cable microsite with links to fresh content, “it freed up three hours” in the daily schedule of the producer whose job it was to maintain that micro-site. “But it’s an ongoing debate in our newsroom. How much automation is too much? What becomes stupid? If you look at the bottom of our site [at another list of related stories appearing at the very bottom of the page], that’s all automated and you find most editors will say, ‘These topics don’t make sense,’ or ‘Some of them are not really what’s interesting.’ So there’s refining, I think, going on.” Similarly, Stokes Young, MSNBC.com’s director of multimedia, noted, “We and our newsroom don’t want to get in the way of any automation that takes away repetitive motion or work where you can replace manual labor with technology. But we want to make damn sure that none of that automation presents too much editorial risk in terms of what our audience sees or perceives around our content.” These concerns about automation permeated editorial work at MSNBC.com, as well as other national news websites, and they applied especially to playlisting.7 Web staffers used automation wherever possible, to generate playlists for general news categories, like “Politics” or “Science & Technology,” as well as data-driven clip queues like the “Most Viewed”

Tracing the Path  74

playlist. They also used automation to duplicate video navigation links across MSNBC.com, such as when the latest clips of Lawrence O’Donnell’s program, the Last Word, were presented as a list of automatically populated thumbnails and links on the show’s blog by way of an RSS feed, or when the latest clips of Hardball got linked to automatically from the cover of the MSNBC TV micro-site. This desire for show links to appear anywhere and everywhere an interested user might want to click on them was tempered, however, by a concern for editorial control—lest automation produce results that were “stupid” or, worse, “risky” in either the legal or PR sense—and by differences in editorial strategy across different parts of MSNBC.com. We can thus appreciate how the path of video is yet again a compromise, this time between a desire for automation and concerns about editorial decorum. Something else is going on here as well. Whereas Turow (1992) once defined the placement of content in front of audiences as the work of distributors, here we can see that distribution concerns—how to create a playlist that will get a Chris Matthews clip in front of a Nightly News user, how to craft metadata that will allow a clip to propagate well via social media, how to curb the spread of content to editorially “risky” contexts within a larger network of sites and “micro-sites”—have become firmly ensconced in editorial work. And while decisions about how prominent to make particular stories, or in which print sections they should be placed, have always been journalistic ones, as news outlets have become decoupled from physical deliverables like newspapers, or even standalone newscasts with predictable paths to the consumer, the movement and placement of content have become ever more prominent and unremitting editorial functions. We’ve followed the path that televised video took between the time it was first broadcast to the time it reached end users online. It’s important to note that even now, the route we have traced is not exhaustive. Live feeds of major events were also occasionally simulcast on television and the web. MSNBC and NBC News television properties produced other sorts of video—raw and behind-the-scenes footage, as well as “web-extra” (unaired), and “web-original” (made for the Internet) video—and procedures and tools had evolved for putting these online as well. Moreover, the actual use of blogs and social media accounts across different television programs and parts of MSNBC was highly varied.

To Turn It On and Have It Play  75

As I argued in the Introduction, the attention of media sociologists, communication researchers, and journalism studies scholars in recent years has been predominantly focused on the major changes to news production that have accompanied the rise of the Internet. But the actual distribution medium that has supposedly wrought these major changes has itself been left largely unproblematized. What we have done, in tracing out the path of online video distribution this far, is to get beyond the notion that online news video “just plays,” and to reconceptualize it as imbricated in a large and complex system. Just as Hughes (2004) took note of the twentieth-century transition, after which it became increasingly difficult for engineers (or historians or sociologists) to see an airplane as existing independently of “a system involving airfields, air controllers, fuel depots, and maintenance facilities” (p. 82), my aim is to ultimately make it equally difficult to see a virtual object like a video player as distinct from a complex network of relationships involving editorial staffs, advertisers, end users, device manufacturers, media conglomerates, software giants, application developers, white-label development firms, satellite links, corporate intranets, ownership shares, corporate contracts, and international copyright law, among other arenas.

5//clip & share engineering for the conversation economy

As we’ve seen, the MSNBC.com video player was designed above all else to be flexible—to appear in many forms across the large MSNBC.com website, to display with custom branding appropriate to the clip being loaded, and to pull in contextually relevant playlists in the hopes of involving users with a variety of different needs. But the player was even more flexible than has been described up to this point. Like YouTube, Vimeo, and many other web video providers, MSNBC.com engineered its video player to be embedded by users on their own websites. Arrangements were also made to allow video to play on mobile devices and other technological platforms that the original video player could not or would not reach. In this chapter I will explore “off-site” and mobile distribution of MSNBC video. In doing so, I will highlight the involvement of a range of “business-to-business” distribution firms that create software infrastructures of a very particular sort: technologies that increasingly enable, but also structure, the distribution of video online and, by extension, our experiences with and exposure to content. These firms produce many of the tools and services relied on by MSNBC and other content providers to place their videos in front of online audiences. Their products range from ready-made software infrastructures for mass distribution to particular algorithms that promise to increase the visibility of videos in search results. A common feature across these vendors and services is that, unlike branded video services and platforms like YouTube and Netflix, whose role as distribution intermediaries is readily apparent to users, the companies profiled in this chapter are most successful when they are invisible to audiences. Their goal is to develop and sell 76

Clip & Share  77

white-label products—video players, search functionality, and so on—that bear the brand not of the manufacturer, but of the content provider. My term for these distribution infrastructure providers—transparent intermediaries—is intended to evoke the manner in which we as users effectively “see through” them. Like a fish unheeding of the water in which it swims, our eyes stay focused on familiar brands and the desired content rather than on the interventions they provide. The term is also meant as a provocation and an invitation for further scholarship—transparency, as a descriptive term, also means “open to scrutiny.” At the same time, because these vendors produce products that are integrated into the strategies and engineering projects of major media companies, we must consider them in relation to these other enterprises. To study the role of transparent intermediaries and their relationship to large media firms is to plumb and pry at the balance of structure and agency enacted in the technical architectures of online distribution. While the MSNBC.com domain itself served approximately 200 million video streams per month by 2010, the site’s embeddable player served up an additional 10 to 20 million streams every four weeks on average from elsewhere around the web, making embeds one of the primary means by which MSNBC cable’s video content was circulated online (Plesser 2011b). If the owner of another website or blog wanted to embed an MSNBC video, he or she would first play that video on MSNBC.com, then click a “clip & share” button on the player. This would cause the video to pause and a lightbox to appear within the player with prominent buttons that allowed users to share a link to the clip by email, Facebook, or Twitter, or to embed the clip (Figure 5.1). Clicking the “Embed” button automatically copied to the clipboard of the user’s computer an embed code for the entire clip, which could then be pasted into the HTML source of the user’s blog or webpage. Alternatively, clicking another link labeled “share your own sub-clip from this video” would bring up a new interface in the space occupied by the original player that included a miniature-sized version of the video, as well as a transcript of the clip in a left sidebar and a “scrubber” bar (a timeline of the video). Users could highlight a portion of the transcript with their mouse or adjust the “in” and “out” points on the scrubber bar to select a portion of the original clip to share (Figure 5.2). Once the user had completed this selection, she could then click an email, Facebook, or Twitter icon at the bottom of the interface to share the “sub-clip” via one

Tracing the Path  78

Figure 5.1. The “clip & share” interface. NBCUniversal.

of these media, or copy the new embed code for the sub-clip appearing at the bottom of the screen. The dimensions of the video player could also be adjusted prior to selecting the embed code. While a few news organizations, including Fox Business News, the Associated Press, and the Washington Post, had created embeddable players by this point, MSNBC.com is widely credited with providing “the first embeddable video player among the major [professional] video producers in the United States” (Plesser 2009e). One goal of the player, as stated by MSNBC .com CEO Charles Tillinghast, was to make video from MSNBC.com more discoverable around the web, rather than relying on users to seek out MSNBC.com as a destination site. As indicated previously, this was in many ways a response to the frequency with which MSNBC cable segments were being bootlegged to YouTube and other online video hosts (Plesser 2008a). MSNBC.com’s deputy East Coast editor Randy Stearns recalled, YouTube was stealing a lot of our thunder. In fact, you still can find more Countdown content on YouTube than you can on our site. But until we had a really good video player and could allow people to share it and clip it and embed it, we really couldn’t go after YouTube. We really couldn’t go after people stealing our content effectively, because they wanted to do the right thing. They wanted to consume the content. We just weren’t giving them the tools. YouTube was, . . . even if it was piracy. So getting that video player was a key part of our trying to do what we could to control the rest of the distribution.

Users rapidly took to using MSNBC.com’s embeddable player, and

Clip & Share  79

Figure 5.2. The “share your own sub-clip” interface. NBCUniversal.

the development team quickly iterated on the original design, adding the aforementioned ability to create embeddable sub-clips from an original segment. MSNBC.com staffers hoped that uptake of videos would be greater if users could choose the specific portions of them that met their needs (Plesser 2009e). Bloggers, as well as advocacy groups—actors eager to pontificate on or via these bite-sized clips—were among the groups that led adoption of the player (Plesser 2009a, 2009d).1 One issue early on was that the embeddable player was originally 425 by 339 pixels and impossible to resize.2 Because of its fixed width, the player would either not appear correctly on—or worse, scramble the layouts of—some users’ webpages. Tillinghast described this lack of flexibility during an industry panel as “a bit of a problem for us,” remarking that “if our embeddable player dimensions didn’t fit well within the structure of a website, then they really didn’t use it.” For this reason resizing was one of the first new features to be added to the player (Plesser 2009a, 2009e). CHAINS OF CUSTODY

Many of the features and much of the work that went into populating the player, on and off the MSNBC.com domain, were tailored to the demands of the conversation economy—the legions of users whose linking and sharing activity is aggregated by search engines and social media sites in ways that selectively increase the visibility and reach of online content. In short, MSNBC.com developers and producers built and used the player strategically in ways intended to make video more findable and attractive to users who discovered and promoted content via sharing and search.

Tracing the Path  80

Stokes Young, MSNBC.com’s director of multimedia, explained that across both search and social media, good textual metadata—headlines and captions, for example—that accurately represented the content of video clips, photos, and other audiovisual materials were “necessary for [their] transmission and distribution.” Visual metadata, like thumbnails, were also quite important and needed to be chosen carefully, to ideally represent “the core ideas, visually” of a clip. In terms of distribution, textual and visual metadata were important, not only for cataloguing and arranging content internally within MSNBC.com, but also because they would be the first, and often the only, information seen by prospective audiences deciding whether to play or click through to a particular video. As Young explained, this encounter, whether it occurred in a list of search results, in a social media feed, or within MSNBC.com’s own pages, was known as the “engagement point”: “What we have tried to do as a newsroom is to apply largely to that engagement point . . . writing really good headlines and selecting really good thumbnail images that will best represent the video content—that will both engage people, and get them to click, but also not be a bait-and-switch. We don’t want people to feel they’re getting one thing and to then feel [misled or disappointed]. To honestly, but well, represent video content through engaging metadata selection is one important editorial function that we do.” Moreover, a great deal of work needed to be done to make video circulate in a search-centric information environment at a time when most search engines primarily indexed text, rather than images or motion pictures. Image and video search tools are increasingly ubiquitous, but even now nearly all of them operate on various textual descriptions of photos and motion pictures—textual metadata—rather than the multimedia objects themselves. That said, given the remarkable amount of video transcoded each day at an organization the size of MSNBC.com—which took in not only programming from a network news division and a t­ wenty-four-hour cable news channel, but also plenty of wire footage, “web-extra,” and “web-original” video—obtaining high-quality, standardized textual metadata for every clip was a tremendous challenge. Search-optimized key words were one form of metadata that would help a video become discoverable, both in MSNBC.com’s internal archives and on the web. But, as Tillinghast noted, “The big problem with metadata has been trying to get human beings to type in what those key words are at the point of production. And that’s very difficult because often times you have field producers that are

Clip & Share  81

not even aware that their video is going to be used on the web, who are expected to put in the right key words to make it searchable. That’s a very long chain of custody, if you will, for a piece of video to maintain integrity for metadata” (Plesser 2009c). Transcripts of video could be a tremendous help here. By themselves, however, searchable transcripts might help find a video, but not the specific section of it that users were searching for. This was not necessarily an issue when it came to short, three-minute television segments, but for longer clips, like say, videos of inauguration speeches, a standalone transcript was not an optimal navigational aid. In response to these challenges, MSNBC .com added some highly advanced features to its player—what Young fondly referred to at the time as “video spaceship technology.” In particular, they developed an automated solution for syncing the transcript of a clip with the exact moment in the video at which the transcribed words were spoken.3 Young explained, “At the level of the phoneme—the individual sound unit—each audio expression [in a video] is broken out into an entity [and stored in a relational database that allows it to be matched against a transcript]. . . . You can [take a clip of] a Meet the Press interview that lasts eight or nine minutes, and look at the transcript, quickly find the part you want, or search for it in the transcript, click on the word in the transcript, click play and the playhead in the video immediately jumps to that place.” Young’s quote begins to drive home the extent of the effort that has gone into negotiating the recalcitrant nature of online audiences. Capturing users who would resort to uploading, embedding, and/or watching bootleg YouTube clips wasn’t simply about matching YouTube’s functionality, but also about providing a technology to users that was in some ways more flexible—ensuring that official clips would be easier to locate, faster to cut, and more customizable than those users would find elsewhere. TRANSCRIPT SYNC: NEXIDIA

The first use of the transcript-sync technology on the MSNBC.com website was the creation of a searchable archive of presidential inauguration speeches, which were featured in the lead-up to President Obama’s inauguration in January of 2009. To create the searchable transcript feature, and later incorporate it into the new MSNBC.com player and its “create subclip” functionality, available across all video on the site, MSNBC.com partnered with Nexidia, a transparent intermediary whose underlying software

Tracing the Path  82

was the workhorse that broke recorded speech into phonemes, creating the searchable metadata that allowed transcripts to be synced to each clip. Nexidia claimed that its algorithms could make a sixty-minute video fully searchable in less than sixty seconds (Nexidia, n.d.). This technology was sufficiently advanced that search engines outside of MSNBC.com’s own video-specific search had yet to integrate it in 2010—at the time of my fieldwork even Microsoft’s own Bing, which syndicated MSNBC video and powered web searches originating on MSNBC.com, had not yet integrated the timeline-linked video transcripts. The search-optimization advantage of the technology, then, was not fully realized. Even so, soon after its introduction Tillinghast claimed that the availability of video-archive-specific transcript search had begun to facilitate a shift in user behavior on the site. What we’re finding is that demand has shifted significantly from the most popular videos, the top videos, to the tail, and that we’re seeing much more take up on tail viewing of video in a way that we have not seen before. . . . The niche stuff. And so it used to be that, in round numbers, that 80% of the videos watched were the top 100 most popular videos, or 80% of the streams were amongst the top 100 videos. Now we’re seeing that it’s almost shifting so that the majority of the . . . videos viewed, the streams that we’re aggregating are outside of that top 100. So, interest is being spread across a large set of topics and news stories and it’s not as concentrated on just a few stories that we happen to put on the cover. (Plesser 2009c) The functionality in the embedded player that allowed users to create their own sub-clips by highlighting a portion of the video transcript was also based on this technology and released later on. Like the original subclip feature, it was intended to increase distribution by allowing bloggers and other end users greater flexibility to grab and use MSNBC and NBC News content in ways that met their needs. But it was adopted and built out for more than this specific purpose. The same technology that allowed users to clip video based on the text of transcripts was also intended to help automate additional portions of the behind-the-scenes encoding process at MSNBC.com—to reflow, in fact, a large portion of the volta that’s been discussed so far. “We had what I used to euphemistically call the world’s largest TiVo,” said Tillinghast of MSNBC.com’s method of re-recording NBC News and MSNBC cable content in Washington from a satellite feed as programs aired on the opposite coast. “Now that we’re using the auto-

Clip & Share  83

mated metadata we’re able to drop [video] into a database inside of NBC, pull it from there directly into our systems, and then make it searchable for our users directly off our site. And then our editors are focused on . . . [other work while] the rest of it is being handled in an automated way” (Plesser 2009c). Tillinghast’s 2009 interview makes it sound as though the “old” way of transcoding video had already been supplanted by transcript syncing. In fact, while the process may ultimately have been automated in the way he described, the editor and web producers I talked with were still using the “TiVo” method of flipping video for the Internet when I arrived to conduct my fieldwork in 2010. In other words, Tillinghast blends expectation and achievement in interesting ways in his quotation, previewing the teleological endpoint toward which heterogeneous engineering was being deployed. The precise reason why this plan had not yet unfolded as expected was not explained—it may have had to do with the inertial quality of existing routines, issues with the reliability of the phoneme-matching software (which wasn’t infallible by any means), or it may simply have been faster at that moment to cut clips “the old fashioned way.” Most likely the explanation lies in some combination of the three. Whatever the case, the transcript-based sub-clipping enabled by the embeddable player can be understood as a public-facing part of a much more extensive and cross-cutting attempt to simultaneously optimize (from MSNBC.com’s perspective) the flow of content at several points along the developing distribution chain. EMBEDDABLE VIDEO AS A RECALCITRANT RESOURCE

Another question from the beginning was how to monetize the embeddable player. The company worried that a heavy or obtrusive ad load might hinder adoption of the player, which could cause users to continue favoring unauthorized alternatives. When the embed feature was first launched in 2008, then, clips appearing in the embedded player featured no advertising. Tillinghast explained that the company needed to carefully evaluate how it would introduce ads—while advertising would be important as a way to derive revenue from embedded videos, ads that were “excessively long” or that “took away from” the user experience could easily interfere with adoption of the player. “If basically it becomes a Trojan horse for a whole bunch of ads, then we’ve failed,” he said. “But I think there is a place to monetize it in a tasteful and appropriate way” (Plesser 2008a). Launching an embeddable player is a prime example of the sort of

Tracing the Path  84

move that loosens an organization’s control over where, and to whom, content is distributed. Unlike Hulu, the popular video portal that focuses on distribution of entertainment programming and has always employed numerous technological restrictions aimed at limiting the spread of embedded content to its target market, MSNBC.com employed relatively few technological countermeasures with its own embedded player.4 Most of the individuals I interviewed or talked with on the subject were unaware of any such restrictions. East Coast Deputy Editor Randy Stearns, however, noted that there were some important, if limited, restrictions in place. “We do some geofencing, for instance. There’s stuff that’s rights-restricted outside of North America, so we can geofence it. They’re pretty crude tools, but they work. They shut off the access where necessary—in Iran, for instance, with certain sensitive stories.5 We can restrict access to North America in those cases, which is where most of our market is anyway.” Blocking entire nations—even continents—from access to content with a few form fields and lines of code is, in some ways, an enormous amount of control to have over the flow of content and begs various questions like those in Chapter 2 about how distribution of news can alter the conditions of social and political congregation. At the same time, as Stearns points out, in other ways this is a very crude, high-level restriction. Outside of Iran and other countries on the wrong side of the geofence, video could still be embedded nearly anywhere. Those familiar with the broad variety of restrictions imposed by online distributors of entertainment programming may initially be a bit surprised that there weren’t more attempts at MSNBC.com to control the contexts in which embedded videos were available. Tillinghast explained one of the key reasons for this difference: Our content is highly perishable. And so it’s not like an entertainment production where somebody has spent weeks, months, years to produce something at great expense and then the only way they can make money is to get people to pay to see it. We produce content that diminishes in value very, very quickly. Like every hour. And so people aren’t in a very good position to make money off of our content by stealing it. They may dilute our ability to make money by putting it up in too many places other than our own website. So what we’ve done is made our video really easy to embed so that it’s actually more convenient for users just to steal it from us than record it and put it on YouTube. Then,

Clip & Share  85

if we do need to police it . . . we offer a free alternative. It’s not like Hulu where, “No, you cannot watch that clip, ever.” Rather, it’s, “Yes, you can watch the clip. But it must come off YouTube, and here’s the link to watch it on our site. It’s all free. Go for it.” Readers may recognize in Tillinghast’s statement just the sort of dynamics of perishable content that Yochai Benkler (2006) described in laying out his economics of information production. But just several years earlier, the major networks and cable channels had taken a significantly different view of embeddable video. Indeed, several national newspapers had offered embeddable video players quite some time before the networks did. Meanwhile, the networks and their news divisions attempted to fashion themselves as destination sites for video, displaying it only on their own sites and those of their business partners, where they had total editorial control over the context in which it appeared. The increasing popularity of embeddable content—including content that was being bootlegged from network and cable news—ultimately led all the national television news organizations to embrace embeddable video. At MSNBC.com, this was a decision largely made in managerial circles, but the editorial staff I encountered embraced it, too. While they were not thrilled by the prospect of their content being embedded and lambasted on the blogs of harsh critics, or appearing in sordid contexts around the web, they tended to welcome healthy debate online and see less hospitable uses of their videos as inevitable, if not always desirable, elements of free speech generally, and particularly on the web. These journalists, like others documented by media scholars over the years, fully expected—and even occasionally took as a point of pride—that their output would rile partisan viewers and organizations (Bivens 2014). Moreover, from an analytics perspective—meaning in this case the number of views received by MSNBC.com videos—in some senses all traffic was viewed as good traffic, regardless of the context or intent with which the videos were embedded. Television-affiliated web producers at MSNBC.com regularly took the time to contact those bloggers and sites using bootleg embeds of MSNBC content, asking them to kindly switch the source of their video to the official version of the clip. This is not to say that these producers were tasked with rooting out every bootleg clip on the web. Rather this was something that was taken on a case by case basis and generally done only when a highly visible site that had the potential to generate

Tracing the Path  86

a significant number of video streams (like the popular blog Mediaite) was using an unofficial copy of the clip. Web producer Will Femia noted that this practice could become awkward when a site or blogger was drawing a good deal of traffic by making caustic remarks about an embedded clip. On the one hand, asking such sites or bloggers to replace a bootleg clip with an official one still seemed important, both for the sake of having a consistent (if informal) policy and because MSNBC.com would still benefit traffic-wise from the resulting video streams. On the other hand, making such a request under the circumstances could be tricky: the web producer wanted to avoid inflaming the situation by giving the impression that MSNBC was out to silence its critics, or conversely seeming to ask, hat in hand, for more video traffic from those critics. But if MSNBC and MSNBC.com themselves were in many ways agnostic about the contexts in which their clips were viewed, the same could not be said for advertisers. EMBEDDED ADVERTISING: YUME

Placing ads on embeddable clips is always going to be a tricky proposition when users can easily default to an ad-free alternative like YouTube. As a result, MSNBC.com experimented with various forms of minimally intrusive advertising, including placing display ads in the player below the video, as well as interjecting brief “pre-roll” (pre-clip) ads clocking in at fifteen seconds or less. But getting users to accept these ads and continue watching was only one part of the problem faced by content providers in this situation. Getting the advertisers on board was its own challenge. Video advertisers are very concerned about the context in which their brands appear—think of all the times that traditional television advertisers have pulled their commercials from programs they found objectionable. This same concern has been magnified considerably in the world of embeddable video, wherein a clip can appear on virtually any sort of website. To help persuade advertisers to support embedded video, MSNBC.com turned to a transparent intermediary called YuMe, a company that ran its own ad network and specialized in “brand-safe advertising,” meaning it employed numerous technologies and forms of certification to guarantee that advertisers’ commercials would only be packaged with a video when the embed appeared in an unobjectionable context. YuMe’s online marketing material from the time of my fieldwork explained: “We understand

Clip & Share  87

that you want to buy effective online video advertising campaigns without worrying that your valuable brand will appear alongside objectionable content. By focusing on premium publishers, enforcing high content standards, and creating proactive technology to prevent ads from running in undesirable syndicated environments, we’re able to deliver the most comprehensive safeguards available to advertisers for online video campaigns” (YuMe Corporation n.d.). YuMe vetted publishers’ video content at regular intervals to assure its quality to advertisers. The service also algorithmically surveilled the sites on which embedded players served videos, automatically “collect[ing] detailed information about the in-page environment of a syndicated video player when it makes an ad request.” Pages and sites deemed unacceptable were added to the company’s “constantly growing blacklist of more than 1.6 million web domains that contain inappropriate content. When we receive an ad request from a video player, we use our domain-detection capabilities to confirm that the player is not embedded on a blacklisted domain before serving an ad.” Finally, the subjective quality of sites where a player was embedded was ranked by YuMe based on the information harvested by the ad service. This information was then used to score the performance of publishers in terms of the general worth to advertisers of the sites on which their players were commonly embedded. YuMe also claimed to “regularly encourage publishers to offer greater campaign performance to advertisers by removing their players from lower-performing domains” (YuMe Corporation n.d.). YuMe itself is a wonderful case study in actor-networks and systems building. It brings together the aims of advertisers and of content producers, translating the resources and intentions of both into the terms of the other. The technical tools it provided were important in this regard—as one manager at YuMe explained to me, “The development of our infrastructure is really important because without the technology component that we bring to the table, we’re just another video ad network.” But even as her comment emphasizes the value of YuMe’s technical components, it also highlights that the company is more than a provider of an inert technological infrastructure. Rather, it is a “video ad network,” a firm with resources beyond the strictly technical, resources that are deployed with an eye toward its own interests and strategies. YuMe, like other transparent intermediaries, is a system builder in its own right, employing a variety of

Tracing the Path  88

heterogeneous tools—it wielded not just a technological platform, but also commercial resources in the form of painstakingly developed networks of advertisers and publishers. Meanwhile, when YuMe’s network of resources were lashed into MSNBC.com’s online distribution system, they allowed MSNBC to navigate delicately between, on the one hand, the recalcitrant interests of advertisers who demanded control over the context in which their ads appeared, and on the other, the users of the embeddable player who wanted the flexibility to place MSNBC cable content anywhere on the web, just as they would with a YouTube clip. The solution? MSNBC.com’s embeddable player could be placed on nearly any site, but ads would be served only in contexts deemed appropriate by advertisers. TRANSPARENT INTERMEDIARIES AS SYSTEM BUILDERS

Some of the technical problems solved by the transparent intermediaries we’ve encountered, like Nexidia and YuMe, may eventually fade away, and in this sense they represent fascinating institutional kludges. For example, a number of producers and developers I talked to opined that users were more savvy than advertisers gave them credit for when it came to distinguishing the source of embedded ads from the site on which a video embed appeared. As embeddable video becomes ever more commonplace and consumers become more shrewd about its use, advertisers may ultimately become more comfortable about placing messages with embedded players without resorting to the kind of technical compromise described here. Even if advertisers remain skittish about the context in which their messages appear, as patents expire and the technologies behind companies like YuMe or Nexidia become more widely or cheaply available, publishers may more frequently elect to create their own solutions to the problems these services attempt to solve. Even so, these intermediaries should be recognized as heterogeneous engineers in their own right. All of them were thinking long term and strategizing aggressively against the prospect of obsolescence. For example, both YuMe and Transpera—another intermediary we’ll encounter shortly—used their positions to develop analytics services and large advertiser networks with high-profile clients ranging from ­Pepsi­Co to Unilever. With these in place, they aimed to have something to market even after the technical problems they negotiated vanished. Indeed, Transpera has since been acquired by competitor Tremor Video, a decision

Clip & Share  89

that was driven, in all likelihood, as much by its client portfolio as by its technological offerings. Nexidia, for its part, aimed for stability by pursuing a diverse portfolio of clients for its patented software. Since the early 2000s, when the company was called “Fast-Talk Communications,” its main clients have been covert intelligence agencies looking to use the software’s search capabilities to data mine audio and video recordings potentially containing sensitive information. While specifics are not always provided in news clippings and press releases pertaining to Nexidia, they suggest that the company’s products have been marketed to, and are likely in use by, the CIA, FBI, and armed forces (Fast-Talk Communications 2002). The software’s ability to “retrieve any word, name or phrase from voice data, regardless of speaker or dialect, with up to 98% accuracy and up to 72,000 times faster than in real time” (Verton 2002) also made the company a darling of the NSA. Indeed, Nexidia technologies are considered likely to have been among the tools used by the NSA for combing through recorded phone conversations during the wiretapping controversy of the early 2000s, wherein the phone calls of thousands of Americans were recorded and analyzed without judicial warrants (Bamford 2008). Nexidia also heavily marketed its software solutions to companies that operated call centers, which might wish to run keyword searches on incoming calls to look for patterns in customer feedback, or to monitor employees for compliance to approved scripts. The company carved out a niche in the legal world, too, by providing its audio search technology as a tool to aid law firms combing through audiovisual materials during legal discovery processes. The array of stakeholders relying on Nexidia’s software is thus extremely diverse and by no means limited to traditional media companies. Even while maintaining this diverse client portfolio, however, Nexidia has attempted to expand its reach in the media sector. For example, with the recent passage by Congress of the Twenty-First Century Communications and Video Accessibility Act (CVAA), content providers whose video is closed captioned on air are responsible for providing similar closed captioning information when the video is placed online. That captioning information, however, has heretofore often been omitted when video is encoded for the web and mobile devices, leaving major content providers with massive libraries of captionless web video that need to be paired and synced with equally large libraries of closed captioning information—a job

Tracing the Path  90

that would take an immense amount of time to perform manually. Nexidia’s software, meanwhile, can pair the audio tracks from hours’ worth of video with existing captioning information in seconds. Not surprisingly, Nexidia turned this technical ability into a marketing pitch, with the aim of making itself an “obligatory point of passage” (Callon 1986) for video providers who want to make their content available online. In short, just as the major distribution players that emerged in the past century, such as theater owners and broadcast affiliates, were not simply inert projectors and rows of seats or providers of towers and cables, these new transparent intermediaries and builders of infrastructure for online distribution are aggressive companies with their own interests, business models, and bottom lines. Their interests often stretch far beyond news and entertainment, though their choices will likely continue to influence these industries and their audiences. OTHER USES OF THE EMBEDDED PLAYER

The embeddable player not only enabled unaffiliated bloggers and other visitors to MSNBC.com to clip video to their sites. It was also one tool that enabled the company to enter into traffic-sharing agreements with other major news sites like Newsweek, the Washington Post, or the New York Times. In such arrangements, the Blue Site would host and/or link to articles from other online publications in exchange for similar treatment, with the hope that the mutual brand exposure and flow of users back and forth between sites would benefit both enterprises. Often the primary mechanism through which these other publishers pointed traffic back to MSNBC.com was by embedding MSNBC videos on their sites. As one MSNBC.com staffer explained, “We do these business deals all the time [in which we] say, ‘You’re really our competitor, but our audience and content are similar. So it makes sense from a distribution standpoint to partner up.’ And we’ll make sure it works for everybody. We’ll split revenue. We’ll have insight into how many people are looking at your article on our site. It’s kind of a weird thing to have someone you consider a competitor [and be] putting their links up. But the web world works so much differently than other mediums, so all of it’s kind of funky.” It’s worth noting that only a few years ago, journalism studies scholars found online publications to be extremely reluctant to link to other sites, let alone other news organizations (Dimitrova et al. 2003). This working of the “competition” into an online distribution strategy is both a reminder

Clip & Share  91

of how quickly online distribution is changing and an excellent example of the binding together of recalcitrant agents into a working system. The business managers at MSNBC.com came to realize that web users get their news and information not from one portal, but from multiple online destinations—and that one key to drawing in audiences and revenue was to be in many places.6 In this vein, web producers and PR staff at MSNBC.com also frequently emailed links and video embed codes to influential and high-traffic blogs and sites, like the Huffington Post, as well as to media columnists like Brian Stelter, then at the New York Times. Much like traditional press releases, these promotional activities often resulted in articles being written. When the embedded player accompanied such a piece, many additional streams— and their attendant ad revenues—accrued to MSNBC.com. What’s more, this sort of direct promotion of links and videos to other sites had already become so common that it had earned its own industry shorthand, “digital word of mouth,” or “DWOM” (pronounced “DEE-wahm”) for short. In 2010, these promotional activities were handled largely i­n-house at MSNBC.com, but in the past, the company had hired firms specializing in DWOM to promote its top stories across the web. The Huffington Post, which like MSNBC TV featured a left-leaning audience, had been particularly receptive to these efforts over the years, becoming one of the leading users of MSNBC.com’s embedded player (Plesser 2009d). By February of 2011, when the Huffington Post was acquired by AOL, it ranked “second only to MSNBC.com as the largest distributor of MSN[BC] TV video” (Wallenstein 2011). The embeddable player not only enabled distribution on other domains, but also allowed for further integration of different publishing systems within the company itself. MSNBC.com was one of the first sites affiliated with a broadcast news division to launch blogs, and because many of its early efforts were experimental, rather than centrally managed, until around 2011 there were multiple blogging platforms in use on the site. These platforms included SixApart’s TypePad and one from Telligent Systems called Community Server that were run in tandem with, but never fully integrated into, MSNBC.com’s primary publishing systems for articles and multimedia. The development of an embeddable player, however, meant that video could be used on MSNBC.com blogs just as easily—and in fact, in exactly the same way—as on sites that were totally unaffiliated with MSNBC.com, all without doing any heavy lifting to integrate the back

Tracing the Path  92

ends of the various blogging platforms with MSNBC.com’s primary content management system. This strategy was successful enough that even when in 2010 the company began converting all of its blogs, including those associated with MSNBC TV, to a standard blogging platform, they didn’t use the primary MSNBC.com CMS. Rather, they were able to build the new blogging software on top of a mature but independently developed publishing system from the social media startup Newsvine, which had been acquired by MSNBC.com in 2007 (and whose role I explain more later). While publishing of video through Newsvine’s blogging software was streamlined over time—acquiring and inserting an embed code came to take less pointing and clicking than it did for an ordinary blogger—it nonetheless still relied on the embeddable player. Finally, the embeddable player could be used on Twitter and Facebook, meaning that when users on either service linked to a clip—in such cases, the actual embed code wasn’t necessary—the embedded player automatically appeared on the service. Users could share such a link either by copying and pasting the URL from their browser when loading it on MSNBC.com, or by using the social media share buttons that appeared in the player itself. MSNBC.com and MSNBC TV staff also regularly used the same functionality to post clips to Twitter and Facebook. Once again a fair amount of heterogeneous engineering was necessary to enable this social syndication of video. Both Facebook and Twitter require that video players be whitelisted, meaning that both the video publisher and the player itself must be approved by the social media service to enable this functionality, and additional code must be written to allow that service to obtain the embed code for a clip based on its URL. While publishers are regularly approved by Facebook and Twitter, relatively few players are whitelisted by the services (Cronin 2009), meaning that most publishers resort to an already approved player such as those provided by Brightcove, YouTube, or Longtail Video. That MSNBC.com went through the trouble of getting its own player whitelisted on Facebook and Twitter— likely making technical tweaks to ensure it operated correctly with each service and complied with their respective terms of use—was a testament to the growing importance of social media and the conversation economy to online video distribution, as well as the influence that Facebook and Twitter in particular had begun to wield in the online distribution space.

Clip & Share  93

CROSS-PLATFORM DISTRIBUTION

At the 2011 Consumer Electronics Show, MSNBC.com president Charles Tillinghast told industry observer site Beet.TV that audiences on mobile devices were uniquely valuable to the company, both because they tended to be more “engaged” consumers of content, and because they tended to skew younger than traditional television audiences for NBC News and MSNBC TV. Ultimately, he concluded, “It’s not so much that mobile is better than TV or desktop is better than mobile. . . . It’s really being able to reach every possible user on the device of their choice” (Plesser 2011a). Mobile devices were viewed as a resource, offering MSNBC TV and MSNBC.com the opportunity to grow consumption and brand affinity by reaching prospective audiences—and especially additional audiences— wherever they might be, on whatever device they chose.7 Many news organizations have seen a similar promise in mobile technology, but there is as yet relatively little scholarship on how news organizations are creating distribution channels to mobile devices. What discussion does exist has primarily focused on mobile technology’s use in news gathering, mobile technology as a means of news consumption by audiences, or its use by citizen journalists to bypass the publishing bottleneck of traditional news organizations (Goggin 2011; Gordon 2007). Scholars like Gerard Goggin (2011) have also highlighted the increasing consumption of news through mobile devices. What I would like to add to this discussion is an examination of how these cross-platform distribution channels are forged. In the process, we’ll discover a number of strategically placed transparent intermediaries lashed together by MSNBC.com and other large media organizations in order to chart a course responsive to the diverse pressures imposed by a rapidly changing distribution environment. As early as 2001, MSNBC.com contracted with Roamable Corp., a company that generated email updates and text messages from database queries, to offer users the ability to receive headlines, weather, sports scores, and other free updates by sending a blank email or text message to any of a list of email addresses (like [email protected]) that corresponded to their information request (Marlowe 2001; MSNBC.com 2001). MSNBC.com continued to aggressively pursue potential customers on the go, subsequently signing deals with additional syndication companies including Skytel, i3, and others to bring the site’s headlines and stories to devices ranging from cell phones to bank ATMs (“Wireless Content Deals” 2000). The specifics

Tracing the Path  94

of these early agreements are beyond the scope of the present discussion, but it is worth noting that MSNBC.com took on many such ventures before they would actually generate ad revenue in hopes of becoming competitive through early entry into these markets (“Wireless Content Deals” 2000). MSNBC.com’s efforts at cross-platform distribution of television programming began around 2004 when access to the cable channel was sold as part of a package by MobiTV, a company that streamed live cable channels to subscribers’ cell phones.8 Available only on Sprint’s cellular network, the service reflected the limited technology and bandwidth available at the time, streaming video at a glacial rate of one to two frames per second and rapidly draining customers’ cell phone batteries (Langberg 2004a, 2004b). In 2005, Microsoft began offering a subscription service that allowed television clips, including video headlines from MSNBC TV, to be synced from a computer to devices running Microsoft’s Windows Mobile operating system. Soon after, MSNBC.com began distributing clips of popular shows like Hardball, Countdown, and NBC’s Today Show in the form of podcasts, and eventually offered full episodes through iTunes, Zune, and RSS (Gough 2005; Tofel 2007). MSNBC.com was able to lash together all these distributors, including their technological resources and access to audiences, into a working system for piping video content to consumers on the go. Still these forms of distribution were limited in their reach—by the terms of individual syndication deals, by distributors’ exclusive agreements with particular cellular carriers or device manufacturers, by limits on mobile bandwidth and the capabilities of the devices on the market at the time, and finally, by the proprietary compatibility restrictions imposed by hardware and software providers like Apple and Microsoft itself. All of these heterogeneous resources were recalcitrant, simultaneously enabling and shaping the path of video from MSNBC.com, but also restricting it. The same could be said of the expectations of customers, who might anticipate an audio podcast of their favorite show, but who were far less likely in 2004 to demand live television on their postage-stamp-sized cell phone screens—an idea one columnist at the time dubbed the “wireless equivalent of a stupid pet trick” (Langberg 2004a). THE MOBILE WEBSITE EVOLVES

Meanwhile, the web was MSNBC.com’s native medium, and it offered a potential cross-platform solution for reaching a broad variety of phones and mobile devices. By 2006, however, only around 1 percent of visitors ac-

Clip & Share  95

cessed MSNBC.com via phone browsers (Noguchi 2006). In response, the company decided to create a mobile interface for its website—one it hoped would ultimately allow visitors to not only read stories and look at photographs, but view video as well. Indeed it launched a mobile version of its regular website (Bournique 2006). Unfortunately, many mobile browsers were limited in their display capabilities, which restricted what web developers could accomplish and ruled out video as an option on a normal mobile site. To get around these problems, MSNBC.com also contracted with a company called Action Engine, which designed mobile phone applications using a custom rendering engine superior to that of most phone browsers of the time, to create a “portal application” for MSNBC.com that would better display its web content on mobile devices (“MSNBC Goes Mobile” 2006; Noguchi 2006). Cost to consumers was another obstacle for mobile web traffic. In 2006, most multimedia content was delivered to phones through media subscription services like MobiTV or VCast, for which customers paid hefty fees over and above what they were already paying for their basic cellular service and an accompanying data plan (Noguchi 2006). At the time, 28 percent of mobile phones on the market were capable of displaying video, but only 1 percent of customers actually utilized video services (“MSNBC Goes Mobile” 2006). To circumnavigate this obstacle posed by cost to consumers, Catherine Captain, then MSNBC.com’s senior vice president of marketing, told the Washington Post that the company planned to subsidize mobile streaming with advertising and offer MSNBC video content for free to cellular customers (Noguchi 2006). This Action Engine–­powered service launched in beta in April of 2006, available initially only for Windows Mobile phones, though support for other devices was planned (Action Engine 2008; “MSNBC Goes Mobile” 2006). In 2008, the Action Engine startup dissolved (Duryee 2008); however, by this time MSNBC.com’s ordinary, non-Action Engine mobile website had become more popular as the iPhone debuted and audiences began adopting it, as well as other smartphones with superior mobile web browsers. Soon afterward, in 2009, MSNBC.com’s mobile site was supplemented with video, including breaking news clips and “branded video destinations for ‘Today,’ ‘NBC Nightly News with Brian Williams,’ ‘Meet the Press,’ ‘Dateline,’ ‘Countdown with Keith Olbermann,’ and ‘The Rachel Maddow Show’” (Transpera 2009). Despite the development of more advanced phone browsers capable

Tracing the Path  96

of rendering sophisticated webpages, there remained numerous discrepancies in the display dimensions of mobile phones, as well as the video formats they could play.9 For this reason, MSNBC.com made a deal with a company called Transpera, much as it had with Action Engine several years before, to help the website circumnavigate the lack of standardization in the mobile phone market. Transpera was an intermediary that took in a feed of videos from a publisher, converted those videos into a range of formats appropriate for display on a wide variety of mobile devices, and finally streamed each clip to users in a format appropriate to the device that was loading it (Roizen 2008). Transpera thus allowed MSNBC .com and other content providers to thread a needle between, on the one hand, the increasing opportunity and pressure to compete in the mobile marketplace, and on the other, the recalcitrant technical requirements of the phones and devices that had enabled this market in the first place. In the process, Transpera offset the costs to the publisher of distributing mobile video—and made money for itself—by packaging video advertising with the clips it sent to the user. In 2009, when Transpera began publishing MSNBC video to mobile devices, it ran the “largest mobile video delivery and advertising network in North America.” It distributed mobile video not only for MSNBC, but also for CBS News, the Weather Channel, AccuWeather, MTV, and Disney, among others (Plesser 2009b).10 Video quickly became a popular addition to the mobile site, with viewing through the mobile web interface climbing 207 percent over the first four months of its availability (as measured against the first monthly figure from March 2009; “MSNBC Mobile Views” 2009). Finally, in 2010, the mobile homepage and show fronts were redesigned to more closely mimic the feel of the desktop websites, a process that included integrating Transpera-powered video clips into mobile story pages and section fronts in ways much more like those used on the normal desktop site. While MSNBC.com continued to exist in some form for low-end phone browsers, this redesign only applied to—and could only be taken advantage of by—smartphone users, specifically those users whose phones used WebKit browsers, including the Palm Pre, iPhone, and Android-­ powered devices.11 The technical limitations of lower-end browsers would likely have been difficult to support, and smartphone users also had more expendable income to lure advertisers. As in many news and media markets, this is a case where the size of a potential audience rubbed up against audience valuation (Schudson 2011). As an AdWeek article put it: “The target audience is consumers who purchase cutting-edge smartphones. ‘They’re

Clip & Share  97

self-selecting into being passionate about content [delivered] via their devices,’ said [MSNBC.com product manager Jeff] Maurone. Though iPhone users make up a small portion of MSNBC’s mobile audience, they’re also the ones who are engaged and ‘spending far more time on our site than [someone with] a Nokia or lower-end phone,’ Maurone added” (Wong 2010). We can see, then, how the company dealt with the recalcitrance of the “mobile” category by segmenting the market and selectively expending resources on a lucrative subset of the potential audience, which was ostensibly identifiable by choice of device.12 Meanwhile, on larger mobile devices like tablets the desktop website could be, and was, displayed in nearly its original form, with minor adjustments to support differences in functionality between touchscreen and mouse interfaces (for instance, the inability to “hover” over menu items when using a touchscreen), as well as to compensate for lack of Flash support on devices like the iPad. As the functionality of mobile devices increasingly mirrored that of the desktop website, the mobile web became a more and more popular means of accessing video, beyond even specialized mobile applications. With the help of emerging HTML5 standards to supplant Flash and its more limited mobile market, MSNBC.com ultimately hoped to obviate the need for users to download specialized apps in order to get an optimized experience of the site’s content (Plesser 2011a). THE ROLE OF MOBILE APPS

Despite such attempts to make the website compatible across devices, MSNBC.com continued to actively support the development of some mobile applications, particularly for the iPhone and iPad, though as the Android market grew the company began to focus attention there as well.13 Presumably because of their idiosyncratic and one-off nature—which will be explored shortly—the construction of many apps was outsourced to an intermediary, mobile application designer Zumobi, though MSNBC.com did go forward with building some mobile applications in-house. Mobile websites for various shows were often highly modular—one developer noted that across the mobile show fronts for Today, Nightly News, and Dateline, “the only thing that is different. . . is a few CSS overrides” (Nguyen 2010).14 But mobile apps were a different story altogether. MSNBC show apps were often tailored very specifically to the program for which they were designed and the particular audience watching it (Plesser 2010). For instance, market research showed that the Today Show’s primary demographic—women ages twenty-five to thirty-four—consumed

Tracing the Path  98

a particularly large amount of web video.15 This was reflected in the Today iPhone app, which prominently and predominantly featured clips from the show, grouped into a wide variety of categories. The Rachel Maddow Show, by contrast, had one of the most social media-savvy fan bases of any MSNBC cable show, and its iPhone app catered to this audience, aggregating feeds from Twitter, Flickr, Wikiquote, and Newsvine (Butcher 2009; Plesser 2011d). Although that app contained a playlist for watching clips from the show, playing video was only one among many options offered on the application’s main navigation interface. Clips displayed in mobile apps were shepherded through the tricky waters of mobile video distribution by the familiar company Transpera, with its built-in ad revenue stream and its ability to customize video formats by device. The executives at MSNBC.com might have liked to have reached every user everywhere, but the company ultimately forged a selective path to some audiences and some of their gadgets. It was a route influenced heavily by market considerations—the sort of audiences MSNBC. com needed to turn a profit; how much users were anticipated to be able to pay for access; which audiences ultimately chose to pay; the availability of advertisers to subsidize these costs to end users; and which users advertisers themselves were interested in. We also see how the path that video took to users’ phones required wending around various technical limitations, some imposed by device and software manufacturers, others a product of ongoing technological development. The route between these obstacles was equally influenced by the various deals and alliances that MSNBC.com was able to cobble together among intermediaries like Action Engine, Transpera, or MobiTV, and the various players—advertisers, cellular carriers, and so forth—with which those companies then dealt. Choices were also made at each juncture about which browsers, operating systems, devices, and consumer media software were worth developing for and interoperating with. The path that put video in users’ pockets bore the impression of heterogeneous actors ranging from the W3C, the consortium that developed the HTML5 web standards, to PepsiCo, which sold advertisements through Transpera, to the regional zoning commissions that were edging forward the mobile broadband market by voting on the installation of new cell towers. The path was also precarious. Not only was the assemblage of actors and relationships in continual flux, many of the television programs that users watched for free on the web and on their mobile devices were—and

Clip & Share  99

at the moment still are—there at the pleasure of cable providers, from which cable channels receive the lion’s share of their revenue. This relationship means that many content providers have been reluctant to put shows online for free. The following comes from a 2010 exchange between online video trade journalist Peter Kafka and Mark Marvel, then senior director of video at MSNBC.com:

pk: When you put that show up for free on the iPad that [cable providers] are paying you for access to, what do you think their reaction is going to be? mm: . . . You’re testing a new market. . . . The fact is, when our stuff is embedded and put on another site, it’s always a challenge . . . looking at the economics. . . . I wouldn’t be surprised if we hear some pushback from them, but my guess is that it’s going to take a little bit for this market to get going. But it’s going to be big. (Plesser 2010) This quote could be read as suggesting that cable providers’ “pushback” was less relevant to MSNBC.com than the opening of new markets, and that may have been the company’s interim position. As is the case with many web companies affiliated with television networks and channels, web video provided MSNBC.com with some opportunity to profit by disintermediating cable providers, at least when it came to on-demand content. Marvel’s diplomatic language in discussing the matter, however, was no doubt due to the recognition that cable providers still had a major interest in online video distribution and some big cards to play. For one, online distribution brought in modest revenues compared to cable distribution. For example, Viacom at the time was earning as much as $2 billion every quarter in television revenue (James 2011), while Hulu, which carried content from the networks and cable channels owned by all of Viacom’s major domestic competitors, earned just $260 million in revenue over all of 2010 (Lawler 2011b), which it then had to split across all its corporate owners. This sort of inequity was apt to make content providers much more attentive to the needs of their traditional distributors, including the cable providers, than to their online distribution partners. What’s more, cable providers might well have asserted contractual rights in the matter, to the effect that the privilege they were paying for, to distribute subscription television content, extended to online distribution. And in fact they were doing so with regard to live television ­programming— in April 2011 Time Warner Cable went to court, seeking a declaratory judg-

Tracing the Path  100

ment as to whether cable companies could deliver live TV to the iPad and other tablets (Lawler 2011a).16 In other words, MSNBC.com was free to test the waters when it came to providing time-delayed video on the iPad, which the various parties, for the moment, had tacitly agreed fell into a different distribution “window” than live programming. But the ground was shifting quickly, and as traditional cable providers developed their own online distribution strategies, Marvel and others were well aware that they might begin applying tremendous pressure, both legal and financial, to limit the ability of clients like MSNBC to become competitors by way of their own online operations. In other words, the cable providers, too, were engaged in aggressive heterogeneous engineering. In the case of MSNBC.com, this became particularly apparent as the massive cable company Comcast purchased part, and later all, of the rights to distribute MSNBC and NBC News content online—a subject that will be revisited in the Epilogue. A VIDEO VOLTA

One of the premises of the conversation economy is that digital distribution of media depends not only—and perhaps not even primarily—on what happens on a news site itself. I have described MSNBC.com’s efforts to reach users wherever they were in physical space by piping its offerings to cell phones, iPads, media players, and ATMs. Related, and equally important to the developing picture, were its efforts to reach users elsewhere on the web. But in tracing out this part of our developing volta, it has become apparent that the boundaries between “on-site” and “off-site” distribution were frequently not as clear-cut from a heterogeneous engineering perspective as they first appeared: a single web domain like MSNBC.com could consist of multiple sites and “micro-sites,” and moreover a single site might run on multiple publishing platforms and technological infrastructures, while technologies ostensibly designed for off-site distribution could become essential tools for on-site publishing. Law’s description of the Portuguese navigation route, or volta, is filled with recalcitrant agents that simultaneously served as both resources and shaping forces for the system builders as they created a navigable pathway between Europe and India. In the same way, we’ve seen how the seemingly simple goal of making televised video available on the web or pushing it to mobile devices requires the enlistment of myriad actors and technologies, as well as creative responses to the limitations imposed by them. The

Clip & Share  101

resulting route taken by video, or video volta if you will, now begins to feel like a remarkable achievement. It is not necessarily a stable achievement in the sense normally thought of by sociologists of technology as a “black box,” however; as I write this, the associations and technologies shared between MSNBC, NBC News, Microsoft, and NBCUniversal and their various staffs and project teams are already dramatically different than they were during the time when I conducted my initial fieldwork. These shifts are just one part of a larger ecosystem of vendors, publishers, developers, device manufacturers, advertisers, cable providers, cellular providers, and technological standards whose relationships have entered a persistent state of flux. The volta is being continually rerouted and reassembled. We are dealing with what Hughes called “postmodern systems” and system builders, who work less with an eye toward permanence and stability than toward continuous change. The relationships involved are successful enough and stable enough at any given moment that, from the consumer’s perspective, video just plays. How and where video plays, and for whom—where the volta goes—is nonetheless still contingent on the nature of these relationships. In mapping out the path of video, I’ve also begun to show how MSNBC is not a single, monolithic system builder filled with teams all in pursuit of a “common system goal” of the sort originally envisioned in the earlier work of Thomas Hughes (1987, p. 51) or various other sociologists and historians of systems. In fact, MSNBC is not even two system builders (of TV and web) but an assemblage of numerous project teams and groups, including show staffs, teams of web producers, video editors, development teams, PR staff, and more. The needs of many groups within the organization become imprinted on an object like MSNBC’s embeddable video player and so influence the flow of content—the delivery of news. And video is only one aspect of this system—just one sort of content put online by MSNBC television properties that during the period of my fieldwork also produced articles, blog posts, photos, audio clips, and social media content. As far as we’ve come conceptually with the example of video, we’ve yet to truly grapple with this shift in perspective from a view of the organization as a firm in pursuit of singular goals to one that sees it as an assemblage of teams and actors whose provincial interests—at times in concert, at times competing—have as much influence over the volta as the recalcitrant actors and forces we’ve traditionally conceived of as existing “out there,” beyond the boundaries of the organization.

This page intentionally left blank

6//breaking apart the monolith considering contemporary media organizations

The promise made in the previous chapters, to demonstrate how news work has become enmeshed with distribution practices, has been only partially fulfilled. We’ve seen how editorial staff, like MSNBC.com’s television-dedicated web producers and video editors, formed a vital part of the system that emerged for putting television content online. But we’ve yet to adequately explore in any depth how content other than televised video was made available on the Internet, or how various teams and shows within MSNBC.com and MSNBC TV managed networked distribution. Based on what I had seen from the outside of MSNBC and other news organizations’ online efforts, when I arrived at each of my field sites I expected to find software developers and news professionals adapting in sophisticated ways to digital distribution and the conversation economy, developing a keen sense of what social network strategies worked best for their material: which stories to forward to political junkies at the Huffington Post, which to pass along to mommy blogs, or which might interest health sites or Media Bistro. And indeed all these things were happening. What caught me off guard initially, however, was the amount of work that was going into getting content to travel within the media organization —the effort involved in persuading editors on the Blue Site to post a link to a Nightly News clip, in turning television commentaries into web articles, or in carving out a blog space for a new cable news show. It quickly became clear that MSNBC was not a single entity, or even two, but rather full of staffs and project teams with distinct aims. In fact, a growing body of literature suggests that when it comes to 105

Out of One, Many  106

large media conglomerates, expansion, partnership, and vertical integration of media companies do not mean simple homogenization of production cultures or content. Rather, large media organizations are full of editorial subcultures with highly provincial and at times competing agendas. And as more and more pieces of these organizations become publicly visible, taking on responsibility for distribution, connecting with users and audiences through niche blogs and social media accounts, what becomes apparent is not just the necessity of taking a network view of the external environment in which information is made to circulate, as emphasized in previous chapters, but also the importance of appreciating the multiplicity of actors and heterogeneous engineers inside a large media organization like MSNBC. Classic media sociology has often treated news organizations as monolithic and journalists as interchangeable, holding contemporary journalism to be a remarkable achievement of professional socialization (Schudson 2011). Herbert Gans (1980), in his classic ethnography Deciding What’s News, suggested that there were few remarkable differences between the newsroom cultures of CBS News, NBC News, Newsweek, and Time, remarking that he was “impressed by the similarities” among them, and that “news organizations are . . . sufficiently bureaucratized that very different personalities will act much the same way in the same position” (p. xxiii). He defines a journalistic culture that spans not only different newsrooms, but also altogether different media, indicating that “the stories which different news media select are sufficiently similar to suggest that technology is not a determining factor” in decision making (p. 80). But perhaps nowhere is the claim of a uniform national journalistic culture more evident than in the literature on news values, where many scholars have taken news judgment across individuals and institutions to be reducible to a set of knowable, enumerated variables like “celebrity” and “negativity,” which are said to be engrained in the customs, if not the ­psyche, of news workers. Here journalists are often said to be so well socialized that they cannot step outside their collective value system to comment on it (Bell 1991; Hall 1973; Hall et al. 1978; Warner 1970). According to Allan Bell (1991), news values “approximate to the—often unconscious—­ criteria by which newsworkers make their professional judgements as they process stories,” while Stuart Hall (1973) calls news judgment “a ‘deep structure’ whose function as a selective device is un-transparent even to those who professionally most know how to operate it” (p. 181).

Breaking Apart the Monolith  107

While some classic studies, like work by Jeremy Tunstall (1971) on special correspondents, have argued for some diversity within journalistic culture, the emphasis of much of the classic sociological work on the news media has been on professional socialization that spans institutional boundaries and the resulting monism of journalism and news work. This is perhaps doubly true of sociological work examining the relationship between the media and protest movements, where the news media are painted as examining the world with a singular, well-defined ideological lens, which is contrasted with and said to violently distort the picture of a diverse social movement struggling to find and speak with a coherent voice (see, for example, Gitlin 2003; Ryan 1991). THE MONOLITH

Over the past two decades, concentration of media ownership has also increased, resulting, by some estimates, in as few as five corporations owning the majority of American legacy media outlets (Bagdikian 2004; Deuze 2007). Consequently, many scholars have shown renewed interest in the possibility that diversity exists in the press—if only to warn us that it is evaporating as transnational media companies distribute their risk and gobble up their competition. Eli Noam (2009) takes issue with some of the more drastic figures produced by scholars and activists, cautioning that much of this literature on media concentration has been “stronger in [ideological] commitment than in empirical evidence” (p. 21). Even so, media concentration does indeed appear to be on the rise (Deuze 2007; Noam 2009). Moreover, while many observers of media continue to argue that the ascension of the web as a popular medium, with its low barriers to publishing, will lead to a diversification of the media environment, there is increasing evidence to the effect that ownership concentration among online media providers is advancing in a manner not unlike the various forms of mass media that preceded it (Dwyer 2010; Noam 2009). Meanwhile, industry critics like Nick Davies (2008), and scholars like Michael Schudson (2011), Mark Deuze (2007), Pablo Boczkowski (2010), and Rena Bivens (2014), contend that as news distribution speeds up to nearly instantaneous, whether through the ascendance of live twenty-fourhour cable news or the development of instant publishing mechanisms online, a culture is rising within the news media that prioritizes constant monitoring and imitation of the competition, including a penchant for pack journalism. Others argue that the diversity of news content has dissi-

Out of One, Many  108

pated as reporting resources have diminished and journalists have become increasingly desk-bound and reliant on the same information subsidies, institutional sources, and competing news outlets for information (Fenton 2010). Without idealizing the past or calling for a return to a “golden age,” the common thread among these scholars is a focus on a declining diversity of content in the news media. Once again, these studies focus on production, rather than distribution, of the news, but they share a common message about news organizations. If one of the key tasks of sociology is to locate and evaluate the interplay in society between structure and agency (Mills 1959; Duneier 1999), a substantial effect of the twin foci outlined earlier—socialization and the reproduction of professional values in journalism on the one hand, and the consolidation of media ownership and homogenization of news content on the other—has been to emphasize structure and the exercise of structural power within news organizations at the expense of the agency of individuals and groups working within those organizations (Bivens 2014; Cottle 2007; Hemmingway 2005). In a sociology ruled by structure, institutional change goes largely unexplained (Giddens 1984). Many scholars and industry observers, however, have noted that journalism and many of the major media industries are now in a period of tumultuous change, owing in no small part to the widespread adoption— not just by organizations, but also by publics and individuals—of technologies for digital production and distribution (for examples, see Deuze 2007; Noam 2009; Dwyer 2010; and Singer 2010). This has led to a major disconnect between much of the core literature in journalism studies and contemporary news practices, as Amy Schmitz Weiss and David Domingo (2010) have noted: The research on the sociology of journalism produced from the 1970s to the 1990s opened up the black box of journalistic craft, but its research interests and theoretical approach made innovation and change a blind spot of its inquiries. We learned about the routines that construct news, but little about the evolution in formats and the progressive introduction of digital technologies. (p. 1158) Keeping pace with rapidly transforming production and distribution environments has led to substantial shifts in the way that legacy media institutions are organized. Consequently, while relatively static structural characteristics like professional socialization do persist within news organizations,

Breaking Apart the Monolith  109

we should be increasingly wary of treating them, in isolation, as defining characteristics of media industries. Rather, as Deuze (2007) has argued, large media conglomerates are in fact assemblages of numerous distinctive organizational subcultures—a situation that has probably long been true to some extent, but is intensifying as these corporations retool to keep pace with the rapidly changing business environment brought on by online production and distribution.1 In other words, as I have previously suggested, if we intend to develop a network perspective on digital distribution, we must extend our gaze not just outward to the myriad actors that affect distribution “in the wild,” but inward as well, recognizing that large media organizations are not single actors with singular objectives, but complex and dynamic assemblages of individuals, groups, and technologies whose interrelations all influence the flow of content. Taking such a stance not only produces a fuller picture of digital distribution, but also suggests that organizational boundaries themselves are in part constructions that do important work, at times obscuring the complexities of the social/technological assemblages around which the bright line is drawn (Child and McGrath 2001; Couldry 2008). In the coming chapters, I will lay out several examples of groups within MSNBC.com and MSNBC TV whose distinctive interests, competencies, and limitations had major effects on the online distribution of television-related content across MSNBC. First, however, I would like to introduce a number of helpful areas of scholarship that will aid in this exploration, including areas related to all of the conceptual problems I have introduced here—the seeming paradox of concentrated ownership and decentralized authority, and the evolution and troubling of professional hierarchies and organizational boundaries within the media industries. OUT OF BOUNDS: MEDIA AND HETERARCHY

I’ve described how MSNBC was bifurcated by its corporate ownership structure, and how it consisted of many teams and editorial subcultures whose provincial goals contributed to the overall workings of the organization, but were seldom perfectly aligned. I’ve also explained how the flexible and at times fuzzy nature of its organizational boundaries was reflected in the routines, artifacts, and technical infrastructures in use throughout the joint venture. These features—which are particularly common among high-tech media firms—have not escaped the attention of organizational scholars. There has

Out of One, Many  110

been a good deal of work in this field on contemporary and emerging organizational forms, which are variously referred to as “postmodern” (Chia 1995; Child and McGrath 2001), “post-industrial” (Turner 2006), “heterarchic” (Girard and Stark 2002; Stark 1999, 2009), “permanently beta” (Neff and Stark 2004), and “postbureaucratic” (Kellogg et al. 2006), among other monikers. A common thread among these studies is that the increasingly tumultuous economic landscape of media in general, and new media in particular, has led organizations to cope with perpetual uncertainty by assuming novel forms, wherein workers and teams are arranged in ways that align with neither the “dependent” relationships characteristic of traditional organizational hierarchies, nor the level of independence once commonly associated with contractors operating in a free market. These new organizational structures, or “heterarchies” (Stark 1999, 2009)—to use one of the favored terms—are characterized by: 1.  Flatter hierarchies. The reporting structure characteristic of conven-

tional organizations has become more fluid, with workers now responsible to many different individuals and groups, even as they often have no clear boss (Girard and Stark, 2002). 2.  Blurred organizational boundaries and the rise of the “project” as a dominant organizational principle. Corporate ownership structures have become ever more complex and overlapping. Meanwhile even relationships between distinct firms have become increasingly fluid, as media products from websites to ad campaigns take the form of proj­ ects involving multidisciplinary teams with resources and members from numerous cooperating firms, who may work together repeatedly in different configurations (Grabher 2002). This interpermeability of organizational borders has led to the notion of the “boundaryless” firm among organizational scholars (Child and McGrath 2001). 3.  Teams and actors that coexist within companies but have provincial goals and logics. As we’ve already begun to see with MSNBC, contemporary media firms can house a great number of different organizational subcultures and professional identities. While the sort of provincialism this produces can occasionally be counterproductive, it can have notable advantages, too (Kellogg et al. 2006). In particular, a number of organizational scholars have argued that just as genetic diversity can ensure the survival of a species in rapidly changing environmental

Breaking Apart the Monolith  111

conditions, the coexistence of numerous organizational logics within a firm can help it to adapt quickly to tumultuous economic conditions of the sort that so often characterize the contemporary media landscape (Girard and Stark 2002; Neff and Stark 2004; Stark 1999, 2009). As Turner (2006) highlights, while such “post-industrial” organizational logics have come dramatically to the fore in the age of digital media, these changes are part of a much longer evolution in the industrial and cultural landscape that began in the 1940s and has been a topic of discussion among organizational scholars since at least the early 1970s. In recent years, the preponderance of wide-ranging and often technologically mediated interdependencies among different professional teams, firms, clients, and end users with distinct and frequently divergent interests has led many organizational scholars to lean on language and concepts from the history and sociology of sociotechnical systems to explain these developments (Chia 1995; Girard and Stark 2002; Jackson et al. 2002; Kellogg et al. 2006). Of particular interest has been the facility of systems scholarship in dealing with how technical artifacts circulate between different social groups and contexts. Indeed, as the very term “heterarchy” indicates, much of the challenge of studying new organizational forms lies in accounting for their heterogeneity—the “increasingly dense and differentiated layering of people, activities, and things,” within and across contemporary organizations, “each operating within a limited sphere of knowing and acting that includes variously crude or sophisticated conceptualizations of the others” (Suchman 2000, p. 6). PROVINCIALISM AND THE SOCIOLOGY OF TECHNOLOGY

One object of this chapter is to describe a local, situated—that is, nonmonolithic—framework for considering how organizational factors influenced digital distribution at MSNBC. There have been a number of calls in the literature on new media generally, and journalism in particular, to “provincialize” studies of digital tools and digital culture in this way. Gabriella Coleman (2010), for instance, in reviewing the ethnographic literature on new media, notes that academic work on online culture tends toward broad overgeneralizations, with scholars frequently assuming that digital media are used in a similar fashion everywhere, and with similar consequences. Drawing on myriad examples from digital ethnographies, she convincingly demonstrates the manner in which the technologies of

Out of One, Many  112

the Internet are both adopted and put to use in highly divergent patterns by different groups in different social contexts, concluding that “the fact that digital media culturally matters is undeniable but showing how, where, and why it matters is necessary to push against peculiarly narrow presumptions about the universality of digital experience” (p. 489). Coleman implores researchers to cast additional light on the “local contexts and lived experiences of digital media” (p. 489), and to take a provincial view of media use and effects. Plesner (2009) draws a similar conclusion concerning much of the extant literature on new media and journalism, suggesting that researchers need to pay much closer attention to contextual and provincial uses of Internet technologies in newsrooms, rather than continually seeking to construct a “general theory of email communication in newswork” (p. 620). A number of theoretical frameworks from the sociology of technology are useful in this regard.2 The first one worthy of discussion is generally referred to as the social construction of technology (SCOT). Pioneered by Trevor Pinch, Wiebe Bijker, and Ronald Kline, among others (Kline and Pinch 1996; Pinch and Bijker 1984), SCOT holds that the shape taken by technological artifacts is a negotiated one reflecting the interests of multiple social groups, each with distinctive views of what the technology is, represents, and what it should or shouldn’t do. Such groups may refuse the technology altogether, or use it in very different ways; for this reason technologies are said to have interpretive flexibility (Kline and Pinch 1996; Pinch and Bijker 1984). Kline and Pinch (1996), for example, in taking a SCOT approach to chronicling the early development of the automobile, were able to show how its emergence was influenced heavily not merely by the (sub)urban motorists celebrated in contemporary car culture, but also by opposition groups determined to prevent car adoption and rural users who modified their vehicles to power washing machines, pump water, and shell corn. “SCOT emphasizes multiple social perspectives, social construction, and the use of tools in specified contexts” (Gay and Hembrooke 2004), and when used as a framework for organizational studies, holds out the possibility that organizations are not monolithic, but made up of multiple subcultures, each with the potential to influence technological development and adoption (Jackson et al. 2002). Ursula Plesner (2009), for her part, points to actor-network theory (ANT) as a framework that is well-suited to both local and contextual anal-

Breaking Apart the Monolith  113

yses of journalism, and to describing changing workplaces.3 Many of the tenets of ANT will seem familiar, since Law’s framework of heterogeneous engineering, introduced in previous chapters, is in part an actor-network reading of the sociology of systems. Broadly, ANT emphasizes heterogeneity of the actors involved in human enterprises, emphasizing not merely social arrangements among people, but also the work done by numerous nonhuman actors from machines to laws in the creation of the social and technical environment in which we live. Much like Edwin Hutchins in his theory of distributed cognition (Hutchins 1995), ANT theorists take note of the various ways in which human tasks are automated by—or, to use the ANT term, delegated to—nonhuman actors. Actor-network theory asserts that once we recognize that a piece of layout software does the job of a human typesetter, it becomes harder to rationalize the omission of such nonhuman actors from our accounts of social reality, both because the choice to automate a particular process is significant in itself, and because we have admitted the notion of material agency—that artifacts do work in what are traditionally thought of as social systems (Latour 1988). Two of ANT’s founders, Michel Callon (1986) and Bruno Latour (1988, 1993), in criticizing traditional sociology for its omission of material agency, have famously pressed this insight to its logical extreme, proposing a tenet of general symmetry that, for analytical purposes, refuses the admission into actor-network studies of any distinction between humans and nonhumans. For this reason, ANT is often portrayed by its founders and champions as an oppositional sociology, a corrective to traditional sociology’s emphasis on the human social element. Researchers outside of ANT’s inner sanctum, however—perhaps recognizing some degree of irony in criticizing a field founded on the social for emphasizing the s­ ocial—have tended to treat general symmetry as a methodological insight, rather than an ontological claim. ANT has traditionally been applied to both the construction of scientific knowledge—how we create networks of resources that turn a messy world into a set of neat equations—as well as to the construction of tools and machines, to examine how a tentative technological solution to a vaguely defined problem similarly becomes a stable, even mass-produced artifact (see, for example, Latour 1987, 1990; Latour and Woolgar 1986; Law and Callon 1988, 1992). Where I have employed the notion of “recalcitrance” from Kenneth Burke (1965), ANT theorists generally talk of translation to similarly emphasize the work that must be done when the affordances and interests of one actor fail to fit neatly into

Out of One, Many  114

the plans or routines of another. When the relationships between networks of actors and the epistemological or technological solutions they support ultimately become stable and routinized, they are said to have been blackboxed.4 On balance, the media sociology literature that concerns itself with actor-network theory is predominantly aimed at comprehending changes in media production rather than investigating how heterogeneous assemblages of actors contribute to changing patterns of news distribution.5 Such sociological accounts nonetheless highlight a number of contributions of ANT to understanding media work more generally, even as it pertains to distribution. Several authors, in referencing actor-network theory’s focus on the marshaling of resources, as well as the discovery of changing and newly established relationships among actors, have conceived of ANT as a theory that privileges agency; they champion it as a corrective to traditional media sociology’s focus on the reproduction of structure, as well as a method for studying emerging, as opposed to established, roles and practices in news work (C. Anderson 2010, 2013; Hemmingway 2005, 2008; Mould 2009; Plesner 2009; Schmitz Weiss and Domingo 2010). Others appreciate ANT’s facility with issues of material agency, claiming that it offers a theory of technology long absent from journalism studies and sorely required in an age of computer-assisted reporting and Internet publishing (C. Anderson 2010, 2013; Domingo 2008; Plesner 2009; Schmitz Weiss and Domingo 2010). Emma Hemmingway (2008) and Nick Couldry (2008) similarly suggest that communication and media studies have long tended to focus either on the structural conditions of media production or its output and audience effects, with little attention paid to the actual process of mediation—ANT, with its focus on the machinery of knowledge production and the role of technology in the translation of both motivation and information, is said to provide a valuable remedy to this state of affairs. A number of authors also treat ANT as something of a field theory, not unlike that by Bourdieu (1983); the manner in which the structures of entire actor networks can shift in response to new entrants is said to provide a lens for examining how new technologies have helped to reshape journalistic practices (Domingo 2008; Hemmingway 2005; Schmitz Weiss and Domingo 2010). Finally, and perhaps most valuable to the present discussion, a handful of media theorists, including Fred Turner (2005), Oli Mould (2009), and C. W. Anderson (2013), have painted ANT as a tool for studying project-based, nontraditional, and more broadly postmodern

Breaking Apart the Monolith  115

forms of media work of the sort that are largely absent from traditional studies of media sociology: To date, journalists and almost everyone who studies them remain wedded to a deeply modern understanding of the profession. In keeping with a world view that routinely and firmly separates the natural from the mechanical and the actor from the action and its consequence, journalists and those who study them routinely separate news and newsmakers, reporters and audience, press and politics. These distinctions have begun to break down in practice though. As new media technologies have begun to invade the journalists’ domain, they have brought with them a need for new theories with which to make sense of the production and circulation of public discourse and for the role of what Latour and others might call socio-technical hybrids in the process. Actor-network theory (ANT) offers a powerful resource for this project. (Turner 2005, p. 321) As mentioned previously, changes to traditional organizational forms have not escaped the attention of social scientists. On the one hand, much has been said about the emergence of newly influential forms of commonsbased peer production (for example, see Hippel 2005; Benkler 2006; Boyle 2008; Bruns 2008; Coleman 2010, 2013), while on the other, the “flattening” in the structure of more traditional organizational forms, wherein contracting and project-based forms of teamwork have begun to supplement or supplant traditional hierarchical management strategies, has fascinated scholars not only from organizational studies, but also from systems sociology and media economics (consider, for example, Hughes 1998; and Powell 2001). More traditional media sociologists, too, have joined the fray. The work of Neff and Stark (2004), for example, which highlights the heterarchic or “permanently beta” nature of web development firms and their products, straddles media sociology and the sociology of the firm. Likewise, as discussed earlier, media scholar Mark Deuze (2007) has made a wide-ranging argument that contemporary media firms operate on assumptions of constant and permanent change, driven in part by technological development, concluding that “whether real or perceived, a structural sense of constant change and permanent revolution is the strongest guide or predictor of the human condition in the digital age” (pp. 234–235). His analysis brings us full circle to the notion of systems building, an enterprise that Hughes (1998) contends changed drastically between the beginning

Out of One, Many  116

and end of the twentieth century, ultimately resulting in “a commitment by industry to change-generating projects rather than to long-lived processes” (p. 5).6 Hughes continues, System builders presiding over prewar projects assumed that they would create a production, transportation, communication, or energy system that would remain virtually unchanged for decades; those presiding over technological projects today expect the systems they build to evolve continuously and to require new projects to sustain the evolution. This attitude is characteristic especially of . . . the computer and communications industry. In a front-page article on 19 August 1996, the Wall Street Journal finally recognized the trend. Alluding to the increasing number of nomadic computer consultants who, moving from one project to the next, solve computer problems for various organizations, the article stated that the nomads point the way “toward something significant: More and more of the work in America is project oriented.” (pp. 6–7; ellipsis in original) The existence of myriad, loosely coupled teams existing within and across organizations, alongside more traditional managerial hierarchies, is a hallmark of the “postmodern” style of systems building and management described by Hughes, in which “discontinuous change is the expectation of project professionals” (p. 302). It also characterizes the various heterarchic organizational forms documented by scholars like Monique Girard, David Stark, and Gina Neff. Once again, heterogeneous engineering offers a useful framework for drawing together many of these insights. It is itself a variant of actor-­network theory and offers up the analytical advantages enumerated by the scholars who have sought to bring ANT to the fore of media sociology. At the same time, while heterogeneous engineering employs general symmetry as a methodological insight, it is not an oppositional sociology. Rather, in laying out the framework, John Law (1987) sought to build on insights of SCOT—particularly the manner in which the interests of many groups become inscribed in the ultimate shape of technological artifacts, with the final form often being one reflecting the interests of those powerful groups that were best able to marshal resources in their favor. Law also built heavily on Hughes’s notion of systems, especially the observation that systems engineering is a heterogeneous enterprise that involves not merely machinery, but also science, law, commerce, and so-

Breaking Apart the Monolith  117

cial influence. Finally, while heterogeneous engineering eschews specific levels of analysis (such as dyads, groups, or organizations) it nonetheless connects to more traditional organizational studies: in particular, it has been offered up among organizational scholars like Robert Chia (1995), as well as Michele Jackson, Marshall Poole, and Timothy Kuhn (2002) as a methodological framework for understanding the structure of alternative and postmodern organizational forms. As Chia (1995) puts it, as organizational forms take on discontinuous and unexpected guises, [John] Law insists that if we want to understand social phenomena such as organization, it is important not to start out by assuming that which we wish to explain. Thus, we cannot begin by assuming the unproblematic existence of social entities such as “individuals,” “organizations” or “society.” Instead, we should begin by assuming that all we have are actions, interactions and local orchestrations of relationships. From this we might then begin to ask how it is that some kinds of interactions appear to “succeed”’ in stabilizing and reproducing themselves, thus generating “effects” such as “individuals” or “organizations,” whilst others disappear completely. (p. 595) SOME ADDITIONAL CONCEPTUAL TOOLS

Before returning to the case of MSNBC, I would like to distill from the ideas introduced thus far a few more conceptual tools to carry with us into the remainder of the book. Consider, first, Law’s emphasis on the entropy inherent in systems, the manner in which they are filled with recalcitrant actors that are always in tension and require continual work to hold together. If Law is right and systems are in a constant state of entropy, what needs explaining is not change, but the fact that there is ever stability, or even the appearance of it. Let me cull this observation and give a name to it: through this lens, a system that appears stable or static from the outside is actually under a great deal of isometric pressure. In other words, there are many powerful pressures on the system, applied by different parties, that cancel one another, resulting in the illusion of calm. Like the contraption from Apollo 13 discussed in Chapter 3, all of these parties may be pushing in different directions, but they are assembled in such a way that they collectively hold together. As soon as one actor relents or changes direction, however, the entire system becomes dynamic, veering and shifting to accommodate the new balance of pressures. The same might be said to

Out of One, Many  118

happen when new actors are introduced or old ones disappear. Isometric pressure, or isometric stability, here refers to the temporary stability of a collection of actors as they hold an arrangement that responds well to each of their energies. This is a different way of conceptualizing what Pablo Boczkowski (2004) and Monique Girard and David Stark (2002) have referred to as a “provisional settlement.” To illustrate isometric pressure with a familiar example, MSNBC.com would no doubt rather have published a live stream of MSNBC shows to the web, while cable providers would rather have live content be available online only through their own distribution system or not at all. The isometrically stable arrangement involved MSNBC.com making complete show content available online through a time delay—a state of affairs that could easily become dynamic again, should the business model of online video distribution continue to change, or in the event that MSNBC TV, MSNBC.com, or any of the cable providers with a significant financial stake in MSNBC TV content were given a reason to become more aggressive about their own position. And in fact, as I indicated previously, there are signs that this is already occurring. As Internet distribution continues to become more lucrative, both cable providers and cable channels have been releasing their own ­subscription-only services for viewing live television content online. These sites and services require users to log in before viewing a live stream, using details that prove they already pay for the channel through a cable subscription. This new model, dubbed by the industry “TV Everywhere,” is widely viewed as having the potential to upend free content models online as cable providers move to monetize their access to content across platforms (Albrecht 2009b, 2009a). In 2011, for example, CNN.com began putting live television content online via a TV Everywhere model (Ferenstein 2011). And indeed, in 2013 in the wake of MSNBC’s acquisition by cable giant Comcast, MSNBC TV and NBC News Digital—the Comcast-era successor to the MSNBC Digital Network—followed suit, releasing a live stream of MSNBC television programming online to users willing to authenticate themselves as cable subscribers. Moreover, while at the time of this writing briefly delayed cable content is still placed freely online in much the same manner that it was during my 2010 fieldwork, at least one recent MSNBC television special’s online placement has been permanently restricted to authenticated cable subscribers (Femia 2014), indicating that MSNBC TV and/or NBCUniversal executives may be experimenting with the idea of levying

Breaking Apart the Monolith  119

yet more restrictions on the availability and timing of televised content online. It should also be noted that the same special was quickly and predictably bootlegged to YouTube, garnering thousands of unauthorized views —a fact that online and television executives will no doubt weigh carefully when considering what direction to take in the future.7 They may ultimately choose to thread yet another needle by, say, emulating the sort of metering pursued by online newspapers—that is, they may continue to make clips freely viewable, but require the user to authenticate after watching some predetermined portion of a program or minutes of content.8 A second important property of postmodern systems that deserves a name is vantage point. In describing modern (as opposed to what he dubbed “postmodern”) systems, Hughes (1987) painted them as nesting, like so many dolls, inside one another. A large system, like a national power grid, has within it many distinct subsystems—local grids and so forth— that can, for analytical purposes, be described as systems in their own right. This gives the observer a choice of where to stand in relation to the system. To push forward with another physics analogy: it is a very classical view of the universe of systems, suggesting one global frame of reference. We can choose which piece of the chart to study, but it remains part of the same branching hierarchical tree. As organizational researchers have noted, however, many of today’s organizational structures bear little resemblance to traditional hierarchies. And as ANT scholars like Callon (1998) and Star and Griesemer (1989) have illustrated, what a complex network or system looks like depends entirely on where the observer stands. In other words, we ultimately live in a relativistic, not a Newtonian universe of systems.9 MSNBC can be seen as the system that enrolls contractors like YuMe and Transpera to distribute its media products. But it can just as easily be seen as a component of a system put in place by YuMe, as a mere content provider to be sold to advertisers. Or we may view advertisers as system builders who enroll YuMe and MSNBC as mere tools for putting clients’ commercials in front of potential customers. This is why vantage point matters—as with frames of reference in physics problems, there is no single correct one, but many waiting to be defined in relation to the interest of the observer. What’s more, this is as true within an organization as outside of it. MSNBC can be seen as the system builder that enrolls a charismatic host like Rachel Maddow as a means of generating ad revenue. It can also be seen as the tool enrolled by Rachel Maddow in a system designed to reach audiences.

Out of One, Many  120

Figure 6.1. Area (A) shows a system under isometric pressure and stable. Area (B) shows a system that is in flux under divergent pressures from the actors involved. And area (C) illustrates how the stability of a system is partly dependent on how its boundaries are defined—that is, on the vantage point from which the system is described.

Once we recognize this relativity at the heart of organizational and systems analysis, it ceases to be surprising that large media organizations are full of groups with different interests, which at times are in concert and at other times are in conflict. Each is a potential system builder. Each is a potential actor in, or force imposed on, another’s system: it all depends on the vantage point. Moreover, whether a system is itself stable or unstable—whether the pressures on it are isometric—all depends on the vantage point from which the edges of the system are drawn (Figure 6.1). As organizational scholars like Robert Chia (1995), Gernot Grabher (2002), and Katherine Kellogg (Kellogg et al. 2006) have come to understand, this realization ultimately challenges the very concept of the firm. If the edges of the system and of the organization are not the same, organizational boundaries become actor categories, rather than analytical tools: it is important to understand the work that they do for (or against) the system builder, but they are not themselves explanans. These observations are important keys, I believe, in understanding the nature of contemporary media systems generally, and online TV news distribution in particular.

Breaking Apart the Monolith  121

Moreover, they are best understood by way of exemplars, and it is with this in mind that I wish to examine a number of case studies of specific groups within MSNBC as it existed in 2010—to explore the same network from the vantage point of different system builders, and to thereby gain a better understanding of how a postmodern media organization (in our new understanding of the term) operates.

7//friends of mine, friends of theirs the origins of newsvine

Newsvine.com is a news-themed social network founded in 2005 by a startup company of the same name. Its story is closely intertwined with the online presence of NBC News and MSNBC TV. The tiny startup was purchased by MSNBC.com in 2007, and was the first-ever acquisition by the joint venture. By 2010, while the original Newsvine.com site remained relatively unchanged, the company’s role within MSNBC.com had expanded substantially. By 2011 technology developed by Newsvine was powering all of the blogs and nearly all of the user-community features on MSNBC .com, as well as on the extended MSNBC Digital Network. And by 2014, Newsvine’s technology would be one of the primary pieces of infrastructure behind a newly redesigned MSNBC.com. Among the pages that Newsvine was already powering in 2010 was a collection of blogs that came to constitute the homepages of most of MSNBC TV’s primetime shows. In this and the following chapter, I will describe several phases in the evolution of this system of blogs and interactive technologies from the vantage point of Newsvine, then swap vantage points to discuss digital distribution from the perspective of the staff working on MSNBC TV’s the Rachel Maddow Show (known internally by the acronym TRMS) in New York. Unsurprisingly, TRMS’s online distribution efforts were not limited to Newsvine technologies, and the digital distribution system will look quite different from this alternative vantage point. Along the way I will also throw in additional views and accounts from other parts of MSNBC.com and MSNBC TV for added context. In doing so, as one of my first graduate mentors, Charles Bosk, put it, my job 122

Friends of Mine, Friends of Theirs  123

will not be to illustrate which account is true and which are false, but how each can be true in its own way. NEWSVINE’S BEGINNINGS

To understand what Newsvine was and what it did, it is necessary to know something about where it came from. Many things about Newsvine changed between the time it was conceived and the time it became a functional website. According to co-founder Calvin Tang, the initial concept for Newsvine was born on a dive trip that he and founder-CEO Mike Davidson took to the Bahamas in 2004 (Tang 2005, 2006b). Newsvine.com could have been brought to you by Disney. The site was, in fact, launched by a small team comprised mostly of developers who had worked on prominent Disney-owned web properties—including ESPN Fantasy Sports, ABCNews.com, and Disney-themed sites for kids—all of whom left the company in 2005 to form their own startup. In an October 2010 interview Davidson described the decision to attempt the project independently, rather than as part of his job as manager of media product development at the Walt Disney Internet Group: I think the moment that I realized I had to leave Disney to do this was when we helped ABC News out with a redesign. . . . My team spent several long nights leading up to the release [working on their site]. They launched on a Friday, and I posted an entry on my blog saying how great it was—how this new site just launched, and how we were using [the latest] CSS and web standards. It got . . . over a hundred comments about how great the site was. Then I flew to Las Vegas for the weekend, and I came back, and [a senior executive at ABCNews.com] had . . . insisted that I be fired. . . . [A]ccording to him, employees are not allowed to speak about the products that they release. Everything is supposed to go through Disney PR. He was mad that somebody who had worked on the product was writing about it. And so, if you think about how bizarre that is, even back then in 2005, [and] . . . how much more bizarre it is today to prevent an art director or an engineer or a designer from writing a glowing post about work that you had just produced, it’s pretty silly. When that happened, I thought to myself, “That’s my signal that this company is going to stay pretty far behind the times. And it’s not worth it for me to try to develop this idea I have for Newsvine inside the company.”

Out of One, Many  124

Origin stories like this one are inevitably incomplete, from a point of view, and colored by memory. Even so, I would suggest that vantage point, as a research tool, prompts us in such a case to look at what a particular system builder is trying to accomplish, which involves tracing the system out from his or her perspective. Davidson’s position at Disney was a recalcitrant resource—that is, although from his standpoint it barred the creative control necessary to launch the web venture he envisioned, it did offer him experience and connections. Whatever his rationale for ultimately leaving Disney, Davidson did run the idea for the site by several of his co­ workers there before the group ultimately decided to leave and incorporate on their own (Tang 2005). Lance Anderson, Newsvine’s director of frontend development, described the earliest phase:

la: I was working at Disney. . . . Mike . . . was my boss and came to me one day in October of 2004 and presented me with the idea of Newsvine. It was really very different from what it ended up being. . . . And we talked through it and after a few months decided to raise money and go for it. jb: So what was the original concept, then? At least the way it was explained to you. la: It was very undefined. It was this idea of comments and being a place for citizen journalists to write and to break news stories. . . . There was the shooting a couple days ago over in West Seattle. It’s like if somebody was there . . . and could break a story and be the first to break it, rather than relying on the traditional news sources. Newsvine was founded amid a span of heavy experimentation with social news among technology startups. Following the notable success of several general interest and niche news-aggregation sites at the turn of the millennium, including Google News, Yahoo! News, Slashdot, and Topix, news provision became a major area of online investment and innovation in the mid-2000s (Kopytoff 2006; Peterson 2006). Between late 2004 and early 2006, a host of popular and/or critically acclaimed sites in addition to Newsvine, including Digg, reddit, memeorandum, the Huffington Post, and Findory, emerged. When Newsvine incorporated in 2005, with just over $1 million in backing from the Seattle-based venture capital firm Second Avenue Partners (run by another ESPN.com alum, Mike Slade, among others), the startup consisted of five staffers, including Mike Davidson as CEO, Calvin

Friends of Mine, Friends of Theirs  125

Tang as COO, Mark Budos as CTO, Josh Yockey as director of technology, and Lance Anderson as director of front-end development (Cook 2005; Tang 2006b; Thelwell 2007). The concept of the project crystallized into an idea for a site where citizen journalism published by users would be aggregated and discussed alongside professional news content. Users would be able to rate the quality of articles and their votes, when tallied, would determine the placement of stories on the site’s cover and section fronts. The result would be a high-rotation news site on which the prominence of articles was gauged by quality, rather than source, leading to exposure for the site’s “citizen journalists” that would far exceed what they would generally be able to obtain by publishing one-off stories independently. Whether the site would be dedicated to general interest or niche-specific news was a matter of some debate early on. For example, the team of former ESPN.com developers had a deep interest in sports, and the early success of several sports-themed web startups of the time, including Seattle’s own Scout Media, seemed to indicate that a site promoting citizen sports writing could hold a good deal of promise (Cook 2005). The debate dragged on over just what sort of news site Newsvine should be, but because much of the code that allowed users to write and share stories would be the same irrespective of the subject matter, the early development process didn’t force the issue. As Anderson recalled, “It wasn’t until we were several months into development where we got to the point where we had to make a decision: Okay, is this going to be sports only? Or is this going to be a more broad approach? And we talked with our investors, and tossed it around and ultimately decided to make a more broad approach.” The site entered private alpha testing in December 2005 (Tang 2005; Newsvine 2005), and was released to successively larger waves of beta testers before opening to the public on March 1, 2006 (Tang 2006b).1 During the beta testing and in subsequent months, the features and workings of the site took on the forms that would largely be familiar to Newsvine users in 2010. “Viners” could write their own blogs on the site, called “columns,” and post links to news stories from across the web as entries in their columns; posting a link in this fashion was referred to in the Newsvine vernacular as “seeding” a story. Viners browsing the site could then vote on the quality of stories—both original and seeded—as well as the quality of individual comments. The votes were aggregated by the Newsvine software, which used the

Out of One, Many  126

resulting rankings to lay out the front page and other sections of the site with popular articles, as well as to boost the status of users who had contributed well-received material. To ensure that the site contained the latest breaking news (rather than waiting for users to post links to stories), articles from the Associated Press were also automatically posted to the site. But as with user-contributed posts, the visibility of AP stories was at least partially determined by member votes. THE SOLUTION STACK: ALL THE FREE STUFF, BUT WE DIDN’T WANT TO SKIMP

Putting a system like this in place required a good deal of heterogeneous engineering, especially because startup sites are usually constructed by cost-conscious developers with an eye toward using their limited resources as efficiently as possible. To understand how this thriftiness translates into the infrastructure and design of a site, some brief technical background will be necessary. When discussing contemporary web applications, developers will often talk about a “solution stack,” which is a way of referencing the various software components required for the application to operate. What are these components? First, just like a personal computer, any machine that hosts a website requires the presence of an operating system before additional software can be run. Second, the same host computer must also run an additional software application that allows it to operate as a server, returning HTML files and other resources (images, style sheets, and so forth) whenever they are requested by a visitor’s web browser. Next nearly all contemporary web applications also store data, such as usernames, passwords, comments, story content, and so forth in a database, which has to be accessed and written to via a third component of the solution stack known as a “database management system.” Fourth, and finally, web applications make use of a “server-side scripting language” to assist in writing data provided by users and administrators to the database, and to display stored information in new and interesting combinations. While at the advent of the web developers and designers generally wrote out by hand the HTML markup—the code containing the bulk of the content—for individual webpages, today this is generally done programmatically through the use of scripting languages. These are programming languages that can tailor the content of a webpage based not only on user input (for instance, they can automatically add a message such as “Thank you for submitting your comment”), but also on system varia-

Friends of Mine, Friends of Theirs  127

bles (for example, by displaying the current time), database information (such as by showing the five most recently added stories), or other sources (by featuring the latest stories from a specified RSS feed, for e­ xample). For a given web application, each layer of the stack can be filled by a number of available solutions.2 For example, Apple and Microsoft both make server editions of their computer operating systems. Microsoft, IBM, and numerous other corporations make server software as well. There are multiple database management systems and variants of the popular SQL database language produced by different vendors. And there are numerous server-side scripting languages.3 In addition to the proprietary solutions offered by Apple, Microsoft, and others, there are also free, open-source solutions available at each level of a stack, one of the most popular combinations being a Linux operating system, Apache server software, the MySQL database management system, and any of a number of scripting languages, including but not limited to PHP, Perl, or Python.4 Owing to the fact that all the parts are free to use and modify, this combination of tools became one of the most popular solution stacks of the last decade, earning it the beguiling acronym LAMP (Kay 2006). And while it may never quite have been true that, as former MySQL CEO Marten Mickos put it, “everyone on the web. . . that matters runs [LAMP]” (Brodkin 2011), it certainly became, and remains, a common standard. Around the time that Newsvine was getting off the ground, however, LAMP—while it had some high-profile adopters, including parts of Google and Amazon—was far from the juggernaut it ultimately became, and was still considered by some to be a risky platform for running sensitive applications (Kay 2006). Newsvine, as we’ve come to expect of heterogeneous engineers, charted a middle course in choosing its own solution stack, striking a balance between the cost-effectiveness of open source on the one hand, and the need for reliability on the other. Josh Yockey, Newsvine’s director of technology and one of its founding members, explained, SQL Server [Microsoft’s proprietary database management system], actually, was the only part of the initial stack that we paid for. Obviously when you’re starting a startup, you want to be cost-conscious. And so we did all the free stuff. We had Linux. And we ran PHP and Apache and so on. But for the database, we really didn’t want to skimp. And particularly at the time . . . MySQL was not particularly stable and refined.

Out of One, Many  128

And so we actually paid for a SQL Server license. We ran SQL Server at ESPN at pretty large scale for fantasy games. And so we were quite happy with that. The decision to build Newsvine on this solution stack also had other implications going forward, which will be detailed later. In the meantime, for posterity, it’s worth noting that large-scale web applications, including Newsvine, also frequently involve several other components beyond just the solution stack that I’ve initially described here. To keep the load on the database manageable, there is often a caching layer—a bit of code that periodically caches and then parrots to users’ browsers the results of common database queries, so that every user who loads a page isn’t querying the data­base. Nearly all contemporary web applications also involve JavaScript, code included with HTML pages that is executed by the user’s web browser, rather than the server, to allow sites to respond to user input in real time as a desktop application would. Several popular JavaScript libraries exist for writing sophisticated JavaScript features. It is now commonplace for Javascript to be used to pull new text, photos, and other information off of servers without forcing the user to refresh a webpage—a functionality generally referred to as AJAX. THE INITIAL USER BASE: GETTING THE RIGHT SORTS OF PEOPLE INVOLVED

It has been noted repeatedly that the culture and social norms of an online space are critical to its success in accomplishing the builder’s objectives (Lessig 2006; Reagle 2010), and many arguably successful sites become popular for reasons other than those the builders originally envisioned. For instance, danah boyd and Nicole Ellison (2008), in discussing the initial growth of MySpace and its attachment to the independent music scene, noted that “while MySpace was not launched with bands in mind, they were welcomed” (p. 217). Similarly, celebrated venture capitalist Fred Wilson has remarked that “Twitter wasn’t planned, it just happened,” in reference to the company’s continual efforts as a startup to adjust to trends set by their user base and third-party developers (quoted in Rao 2011). Newsvine developers recognized early on that, in an environment where anyone could publish, it would be essential to start out with a community of thoughtful, pro-social users. This informed their selection of the initial user base, which began using the site during the alpha and beta

Friends of Mine, Friends of Theirs  129

tests. Davidson recalled, “One of the things about Newsvine is that it was a self-selecting community from the start. The people who were there during the alpha were friends of mine. The people who were there during the beta were friends of theirs.” The importance of starting with a selectively arrived-at user base was emphasized on the Newsvine blog, where the staff posted announcements and development news. Upon the announcement of the alpha test, the blog read, “Quality of conversation is often tough to maintain in open expression forums (see ‘Slashdot’ :) ), but as long as we’re judicious about getting the right sorts of people involved, discourse should be more like a college classroom and less like a college hockey game” (Newsvine 2005). As is now relatively common with closed betas, a month later, when the initial beta test began, users were given a limited number of invites to share with friends, each of which would admit a new user to the site. Through the invite limit, and more explicitly through the Newsvine blog, users were instructed to be selective in who they invited to the site. These restrictions were coupled with a monetary incentive for inviting productive users by way of the site’s revenue-sharing program: “As the inviter of any new member, you’ll receive 10% of the ad earnings from any traffic they generate. Here’s the rub, however: by inviting someone into Newsvine, you are implicitly endorsing them as a potential positive member of the community” (Newsvine 2006d). Similarly the site’s COO, Calvin Tang (2006a), posted to his personal blog, read by many of the alpha testers, this message: “Since Newsvine is a meritocracy whereby people discover great things based on endorsement by other users, one of the most important things any user can contribute is a wise choice on who to invite into the Vine” (para. 1). At this point, we can begin to appreciate the amount of heterogeneous engineering that went into assembling Newsvine’s initial core user base. The developers relied on their personal social ties to find an initial high-quality group of users. Subsequently they used a combination of social engineering (verbal guidelines for invitations), technological restrictions (limited numbers of invites), and financial incentives (revenue sharing) to ensure both thoughtful postings by the expanding user base and a reasonable rate of growth. These steps in turn made it more likely that existing users would remain and become acquainted with one another, and thus become both a self-policing community and one that would naturally avoid the sort of bad behavior that often comes with online anonymity.

Out of One, Many  130

THE PUBLIC USER BASE: AN ASPIRATIONAL PHILOSOPHY OF USER BEHAVIOR

Of course, once a site oriented around user-generated content is open to the public, the builders face two major challenges: acquiring and bringing to the surface high-quality content, and keeping out malicious or illegal behavior. These are not unrelated problems—after all, if the featured content on a site is laughable, the whole enterprise is less likely to be taken seriously by users and critics, putting the site at increased risk of trolling. Extensive spam and trolling, in turn, can drive away users and their potentially valuable contributions, leading to a downward spiral in quality. Even during the closed beta period of the site these were pressing concerns among Newsvine developers. Lance Anderson recalled, “I ate chips and salsa” is a one-liner that we use around here. Because the first week into private beta—which was six, eight, ten thousand users, something like that—there was a story on the front page and the headline read, “I ate chips and salsa.” And you’d click on it, and then there’s one sentence and a picture of, like, chips and salsa. And it was, “Featured Writer,” you know? And it was like, “Oh, God. Our site’s going to—We’re just going to fail so miserably.” It was like a punch in the stomach: “Oh, no. We’re going to be so bad.” The development team strategized aggressively against this possibility through numerous avenues over the course of the beta and the site’s first year of operation. As indicated earlier, Newsvine initiated a revenue sharing program that not only gave users a portion of the ad revenue generated by the content they contributed, but also offered them a cut of whatever ad revenue was generated by the users they referred, thus encouraging viners both to contribute good content and to recruit other high-quality users to the site (Cook 2005; Newsvine 2006d). A fundamental technological element designed into the Newsvine site, as with other social news sites, was that it increased the visibility of stories and comments based on user votes. In the case of stories, this was done by populating the cover and front pages of sections (U.S., Sports, Politics, and so on) with highly rated material and, within stories, adding quick in-page navigation links to highly rated comments.5 The site also included technological countermeasures against abuse and questionable use cases. For instance, the Newsvine software contained a profanity filter, which found and replaced any of a library of objection-

Friends of Mine, Friends of Theirs  131

able words with grawlix whenever a user posted content.6 The software likewise prevented virtual shouting by prohibiting viners from typing in all caps. There were also technological countermeasures in place to weed out spambots and malicious users as soon as they attempted to join the site. Users who began their career at Newsvine by contributing links from a list of sites deemed “questionable sources” by Newsvine, by linking to the root of a site (for instance, http://mashable.com, as opposed to an article page on mashable.com), or by posting too many links to the same web domain in rapid succession, were flagged by the site’s algorithms as potentially malicious users. The site’s filtering software also included automated mechanisms that made it simple for users to report abuse and basic “content checks,” wherein the software scanned for words and message attributes common to spam. Developer Josh Yockey explained, If somebody comes in with just straight spam or something offensive, then it tends to quickly get reported. . . . We [also] do some user segregation by level of trust. So if you haven’t confirmed your email address with us you’re less of a trusted user and you can’t post any links. That gets rid of a lot of automated spam. [Most automated spam software] doesn’t bother going through any [additional] steps, because there are so many other sites where you can just post a link and get away with it. So even if you put up a pretty minimal barrier, they won’t bother. Or they’ll try and fail and not even notice. Most of that’s “fire and forget.” The “levels of trust” that Yockey mentions are in reference to Newsvine’s moderation technology, which contained a fairly nuanced array of “tokens” and “states”—corresponding to specific privileges and restrictions across the site—to which users could be assigned. The moderation system, however, was only one part of a composite and evolving solution for managing the user base. One important reason that heterogeneous engineering gives us such an apt vocabulary for discussing a system like Newsvine is that many of the most important components of user management on the site were not technological actors at all. When Newsvine began in private alpha among friends of the staff, it had few strict user guidelines. When the alpha was announced in the Newsvine blog, the first comment beneath the post began, “The only thing that I don’t really understand is what kind of posts are you expecting? Is there any kind of guideline? Can I write about anything or only [things that are]

Out of One, Many  132

news related?” Davidson replied, “We definitely need some posting guidelines, but we’re going to lean towards posting anything you *personally* find to be newsworthy. We’d like to minimize the amount of ‘I just walked my dog’ sort of posts, but you can’t really tell people what to post. I think once there are a ton of people using this, it will kind of sort itself out via the voting. As in, the more widely read your posts are, the higher up they will go in the vine” (Newsvine 2005). When guidelines were ultimately published, upon the public launch of the site in March of 2006, they took the form of a sort of legal-­philosophical document called the Newsvine Code of Honor. The Code of Honor was developed by the Newsvine staff in cooperation with users during the private beta. The initial invitation for users to contribute suggestions for the document projected that the Code of Honor would consist of “about 10 common sense guidelines for people to follow when interacting within the community” that would “help avoid the problem of people treating the community as an open playground” (Newsvine 2006b). Ultimately, the staff assembled a list of six pithy guidelines (Newsvine 2006c), written by COO Calvin Tang, and in 2008 five of these were fleshed out into the language that would remain on the site for years to come (see Box 7.1).7 The Code of Honor served three major purposes. First, it provided users with guidelines regarding the sort of activity that was acceptable on the site. Second, it was an easily referenced set of rules that the Newsvine staff could point to when moderating content, disciplining users, or deleting offending accounts. Finally, and perhaps most importantly, it was intended to provide the basis for a self-policing user community. Yockey explained, Early on we had a sort of aspirational philosophy about user behavior. . . . If you’re respectful to your community, then at least some critical mass of your community will be respectful back. Some sites are much more cynical about user behavior and they just start in the assumption that users are all a bunch of foul-mouthed trolls. The obvious [example] is comment moderation, where a comment won’t even show up until a human sees it. I think users are . . . insulted by that and think, “Well, if you think we’re just a bunch of foul-mouthed trolls, then I guess I will be [one].” Whereas on Newsvine . . . we put a lot of faith in our community guidelines. . . . [The Code of Honor] didn’t constrain everyone. There are obviously hardened trolls and there are people who don’t understand the thing and people who have short tempers.

Friends of Mine, Friends of Theirs  133

Box 7.1. The Newsvine Code of Honor as It Appeared by 2008

1. A  bove all else, respect others. Address issues and arguments and refrain from making personal attacks. If you see something disrespectful or inappropriate, report it--rather than further inflaming the situation. 2. N  ewsvine’s primary purpose is to provide a place for people to share and discuss topics relating to the news. Self-promotion, seeding links to your own site(s), and advertising are not allowed. 3. H  eadlines should be supported by the information presented in the article/seed, rather than used primarily as a means to draw attention. Articles and seeds must be published to appropriate groups and categories. Tags must be relevant to the article or seed. 4. A  s the host of your column, you are expected to foster healthy, open discussions by setting a good example. Be responsible for the content you submit and exercise impartiality when deleting comments and reporting abuse. 5. A  cts that run contrary to the spirit and purpose of Newsvine, including attempts to circumvent the Code of Honor & User Agreement, are not allowed. (Newsvine 2008b; Newsvine 2008c)

. . . But on average it was actually extremely helpful, because the community policed itself when it had . . . some guidelines to go by. . . . Once you codify those standards, then people are willing to defend them. So the non-technological side took care of a lot. . . . We were almost surprised at how eagerly people took to self-policing.8 While Yockey makes the case here for respectful treatment of users, it’s important to note that Newsvine’s design, whereby viners were allowed to publish instantaneously and objectionable material was taken down after the fact, was not just an editorial or regulatory philosophy—it was also a labor-saving strategy without which it would have been impossible to maintain the site with a staff of less than ten people. In my first phone conversation with Mike Davidson, he elaborated on the dual nature of this policy of moderation-after-the-fact:

Out of One, Many  134

If you’re going to carry out a conversation with other users online, you can’t wait four hours or . . . twenty-four hours for your comments to be published. We have threads where five or six users will be discussing something for three days, . . . publishing comments sometimes five minutes apart. . . . If you kill that, you kill discussion. I look at pre-­ approval of comments as giving up. It’s like saying, “We think we need to have this commenting thing, but we don’t like the quality of comments, so we’re going to approve things.” I don’t think it’s a good user experience [or that] users appreciate it. . . . I mean, we have one moderator on staff. That’s it. . . . We moderate tens of thousands of comments every day with one person, and the reason we’re able to do that is he isn’t . . . reading every comment. . . . He is responding to user reports of bad behavior on the site, and he’s checking out certain high-volatility threads, popping in, slapping people on the wrist, giving people suspensions, giving people encouragement—all sorts of things. . . . That’s a lot more useful way to encourage the sorts of behavior that we like to see around the site. This form of moderation depended on outsourcing the discovery of unacceptable and malicious behavior to the site’s myriad users—a social form of the open source developer’s mantra, “given enough eyeballs, all bugs are shallow” (Raymond 2000). At this point we can begin to understand the Code of Honor and the policy of after-the-fact moderation as heterogeneous actors in a system intended to make a user base that included over a million registered accounts manageable by a single moderator.9 But to enroll users in the task of moderation requires additional policies and technological actors that we’ve yet to explore. First of all, users had to have a way to report abuse on the site. When users were logged into Newsvine, they could delete offensive comments on their own stories. Moreover, every story and every comment they saw was accompanied by a form button labeled with an exclamation point, which allowed users to flag the story or comment as a violation of the site’s policies. Clicking it brought up a context menu of possible conduct violations, which once selected would send an automated abuse report to the Newsvine moderator. Users could also use the site’s general-purpose contact form to file reports of abuse and malicious behavior on Newsvine. Moreover, even when crowdsourced, moderating after the fact only makes sense if a site’s proprietors don’t knowingly let malicious actors into the site to begin with. Assuming good faith is an admirable policy for le-

Friends of Mine, Friends of Theirs  135

gitimate columnists and commenters, but it quickly breaks down when applied to spambots, trolls, and black-hat SEO marketers. To keep these pressures at bay, the site required users to register before posting, and as we’ve seen employed a number of algorithmic filters to flag potentially abusive users and content. But by themselves these filters would not have been enough to stop malicious accounts—for instance, even if a malicious account was flagged for review after one post, a few spambots creating accounts en masse could still have overwhelmed the site with junk content one post and one account at a time. To prevent spam algorithms from taking over the site, and to add an obstacle for users who would register for the sole purpose of posting something unsavory, the Newsvine staffers created a quarantined section of the site, which they dubbed the “Greenhouse.” In principle (exceptions were introduced, as we will see), any user who registered for Newsvine was initially untrusted by the software. This meant two things. First, any blog posts they published to their Newsvine column would be skipped over by the algorithms that normally promoted content to the cover and section fronts across the site based on user votes. These posts would, however, be featured in the dedicated section of the site called the Greenhouse. Second, while untrusted users were allowed to comment anywhere on Newsvine, any links included by these users in their comments were automatically removed by the system. Newsvine’s full- and part-time moderators regularly visited the Greenhouse, deleting posts and accounts from blatantly malicious users and spambots, at times blocking their IP addresses when these agents attempted to register repeatedly. At the same time, moderators would elevate apparently pro-social users to trusted status on the software, and explain Newsvine’s Code of Honor to individuals whose contributions were clearly not malicious, but still unacceptable by the site’s standards of conduct. Existing Newsvine users could also visit the Greenhouse and report abuse there or vote for articles they liked; enough votes from trusted users on a novice’s seeds and articles would also tell the software to graduate the new user from the Greenhouse. The Greenhouse accomplished several things for the Newsvine system builders. First, it kept malicious accounts and any content they generated from wreaking havoc on the site’s article recommendation system or otherwise severely impairing areas of the site frequented by the public or relied on by trusted users. Second, because it minimized the exposure of malicious posts and intervened in the posting of spam links to the com-

Out of One, Many  136

ment threads of existing content, it reduced the incentive for spammers and trolls to target Newsvine in the first place. At the same time, an automated system that divided users into trusted and untrusted groups could presumably be created without a section of the site dedicated to featuring content from untrusted users, which begs the question, why have this portion of the site at all? The answer is again that it contributed to the desired social environment on Newsvine. The Greenhouse was promoted to users as “a section where new users are featured for the community to discover and endorse” (Newsvine 2008a). When it launched, the Newsvine blog declared it one of the few positive outcomes of the unfortunate need for a trusted user system: “The upside of this, however, is that we can now feature all new writers in one area: . . . The Greenhouse is the best way to help identify new positive members of the community and even add them to your Watchlist if you so choose” (Newsvine 2006a).10 As Tyler Adams, Newsvine’s staff moderator put it, “The Greenhouse is . . . pretty valuable as a training ground” for new users. Ideally, being featured in the Greenhouse gave new and still relatively anonymous users an elevated level of attention and positive feedback upon joining the site, and provided a safe area where new users could become acquainted with the norms of Newsvine before their content entered the community at large. It also put a positive spin for new users on what might otherwise be considered a delay in their exposure akin to the sort of pre-moderation that Davidson had derided. In addition to the Greenhouse, Newsvine also created events on the site as another way to inculcate users into a desirable—and regulable—­ normative framework. For instance, in 2010 Newsvine organized semi-­ formal debates between users as models of civil conversation in an initiative called “Newsvine Next Level” (Tang 2010). Others took the form of live events—over the years, users organized meetups across the country, which the staff encouraged and at times attended. The developers also created a sophisticated karma system, called “Vineacity,” that, like similar systems on other sites, totaled the amount of positive feedback a user’s stories and comments had received, but—unlike most of these other systems—also programmatically enumerated other types of achievement that the Newsvine staff saw as desirable, such as courtesy to others and connectedness within the community (Figure 7.1). Most of the Vineacity badges (or “branches” in the Newsvine vernacular) that users earned through their participation were awarded automatically by the site’s software. But this was not true of all branches, as Davidson explained:

Friends of Mine, Friends of Theirs  137

Figure 7.1. An illustration of “Vineacity.” A smaller version of the icon explained in this illustration appeared on each user’s profile in 2010, displaying those branches that the user had earned. This image and the included text are a recreation of a screen-resolution diagram by Newsvine. NBCUniversal.

The sixth branch is the only one that’s not automated. That’s the branch that we call the “Random Act of Vineness.” We give this out editorially for moments of greatness on Newsvine, or within the Newsvine community. We had a user organize a cancer walk around a track in the middle of the night in Ohio, and he raised thousands of dollars for cancer research. . . . We had another user get their name on the ballot for the Ohio Senate race, and he documented the entire thing on Newsvine. He got a branch for that. We try to create this culture where people are concerned with their reputations around the site. We do not want a drive-by culture where people think they can just pop into Newsvine, sound off and then pop back out again. Here we again encounter heterogeneous elements, wherein technological components like registration of user identities and the Vineacity karma system were paired with a policy initiative—the Random Act of

Out of One, Many  138

Vineness—to promote a desirable normative structure for the site’s user base. For instance, Random Acts of Vineness were awarded for contributions to the Newsvine community, such as to a group of users who systematically helped to report spam postings to the moderation staff. The award was also given to users who embodied the site’s aspirations to citizen journalism. The story of the 2007 Virginia Tech school shooting was in fact broken by Newsvine via Chris Thomas, a user whose wife worked at the school; he later received a Random Act of Vineness award, as did Sandy Frost, a Seattle user who, through Newsvine, published an investigative piece on the Shriners that received recognition from the Society of Professional Journalists. Moreover, Newsvine intentionally limited its user base in correspondence with its ability to moderate and socialize them. The sixth original clause of the Code of Honor, which admonished users to post only in ­English, was not eliminated, but rather ultimately moved to the site’s more labyrinthian user agreement. As Yockey explained, the prospect of a multilingual Newsvine was actually quite thrilling initially, but ultimately unsustainable from the staff ’s perspective: “We wrote [code that was] reasonably internationally friendly. And without our planning it, some guy wrote an article in Arabic—which I was super-for. . . . But eventually I got overruled. Even though [moderation is] mostly user-driven, we [as the staff] do provide some moderation function. We try to keep flame wars and spam out, and we just couldn’t do that in Chinese or Arabic. So rather than lowering our moderation standards, we just stuck to English. But I thought it was cool to see some Chinese and Arabic articles in the first couple days.” This decision to limit the potential growth of Newsvine in accordance with the ability of the developers to understand what its users were writing underscores the importance of social engineering on Newsvine. If aggregation of votes was enough to make the site work in accordance with the builders’ wishes, then concern over having a multilingual user base would be moot provided enough users existed for a particular language to moderate corresponding articles effectively. But the technology of Newsvine, as we’ve seen time and again, depended on the establishment of a prosocial user culture, which involved a much broader array of interventions. As John Law (1987) and other scholars of sociotechnical systems would point out, it ultimately makes no sense to separate the gadgetry of Newsvine from its social predicates—or its financial or policy ones, for that matter. Only together did the various parts constitute a working technological system—this, again, is the essence of heterogeneous engineering.

Friends of Mine, Friends of Theirs  139

Of course, unruly behavior did occur on Newsvine, and as Yockey explained earlier, moderation was necessary. The Newsvine staff employed diverse mechanisms to create a social environment in which the users themselves took on much of the work of policing the community. Parts of this process were automated—for instance, when enough users reported a comment on an article as abuse, it was automatically “collapsed” by the system, meaning that the text of the comment, while not removed altogether, was hidden by default (users who wished to read a collapsed comment could click its heading to reveal the text). As the site moderator, Tyler Adams, noted, “You can also bring down articles with enough reports. And the amount of reports changes as the article gets [more exposure]—it takes a lot of reports to bring down an article with a lot of attention and eyes on it, because there’s a fair ratio.” It’s worth noting that the comment-collapsing algorithm itself is a heterogeneous product. When debates get heated, some comments may simultaneously be voted up by some users and reported as abusive by others. Creating a collapsing algorithm that intelligently weighs the balance of votes for and against a comment—and operates on a sliding scale as a discussion scales in popularity—is a tricky challenge, one simultaneously technical and social. Law (2002), in examining the engineering equations behind contemporary aircraft design, underscored the manner in which equations that on their face purported to simply explain the physics of lift, wing contouring, and airspeed were ultimately as much or more about the comfort and safety of crews and pilots. In his story, a neat, concise equation for acceptable values of “gust response” appears deceptively simple and technically oriented at the end of the day, hiding the range of concerns that went into its production. “Removed from the flat space occupied by the formalism, we find ourselves in the sweating world of the aircrew. We discover pilots who flew their creaking aircraft too low, pilots who worried about whether the wings would break off, pilots who were thrown about their cockpits, pilots who climbed shaking from their aircraft at the end of these flights. If we are imaginative, then perhaps we can smell the fear, feel the sweat on the bodies, the taste of vomit” (p. 123). Bruno Latour (1990) refers to this process through which the messy and diverse world of experience is reduced to an equation, an engineering diagram, an algorithm, as deflation. While Newsvine is far removed from the life and death world of experimental aviation, we can similarly understand moderation algorithms as concealing a great deal of hard-won experience, and containing within them a set of working assumptions about

Out of One, Many  140

fundamentally social problems: the extent to which controversy should be tolerated and/or valued in discourse, what a constructive debate looks like in aggregate as opposed to a messy argument, as well as the nature and values of the community that is flagging and voting on those comments. They are thus heterogeneous technologies. Automated moderation systems at times may seem simple, but thinking about them in this way reveals a surprising amount of complexity at work. The same was true of the site’s algorithm for featuring comments that had been rated highly by users—setting a threshold for how many user votes make a comment “highly rated” takes a good deal of knowledge about the site’s dynamics. Set the threshold too low on a high-traffic site and the “highly rated” moniker ceases to have meaning. Set it too high and there won’t be much to feature. Still, even with the myriad financial, technological, and administrative choices aimed at socializing a responsible and productive user base, some moderation had to, in the end, be done by human staff. The Newsvine staff had a number of moderation implements at hand, including all the standard tools you’d expect, such as the ability to delete comments, articles, and user accounts. The moderator could also manually feature a comment, adding it to the list of those highly rated by users, or manually collapse it. And he could manually restore comments and articles that had been collapsed by the community. In addition to these basic tools, the moderator also possessed a more nuanced set of options for dealing with problematic users.11 He could grant or revoke trusted status from users; suspend a user from logging into the site for a specified period of time—a day, a week, or a month; or ban them indefinitely, in which case their existing content remained intact, but they were unable to use the site. After banning a user indefinitely, the moderator could also, if he chose, delete the person’s account, removing all traces of that content from the site. Alternatively, users could be “quarantined,” meaning that their trusted user privileges had been revoked. Their activity while in this state was limited to their own column (for instance, the viner could comment on her own stories, but not those written by others). Users could also be assigned “tokens” by the moderator, which granted or revoked specific privileges on the site. For instance, there were tokens to give users access to specific beta features not yet available to the larger user base. For viners who abused the site’s user-moderation features, there was also a token that prevented them from deleting comments or filing abuse reports. A staffer could also use the moderation interface to send emails to of-

Friends of Mine, Friends of Theirs  141

fending users, explaining why changes had been made to their accounts. This email feature included the option of sending an automated blast specific to a given action or offense, or writing a more personalized note. The moderation interface also aggregated a range of information about viners, collating the abuse reports they’d filed as well as those that had been filed against them, and displaying over twenty statistics about the user’s account, including the number of stories the user had published and how many had been censored or deleted. Newsvine as a sociotechnical system didn’t work perfectly to engineer user behavior, of course. Tyler Adams, the site’s full time moderator, described problems from time to time with what he jokingly referred to as “Newsvine gangs”—cliques of likeminded users who coordinated through back channels to systematically game the site’s collapsing algorithm and bring down the articles and comments of their ideological opponents. As moderator in these cases, he would often have to engage in a fair amount of detective work to discern who was participating in such bad faith behavior and ban or suspend them from the site. Moreover, users at times would attempt to game not only the site’s technology, but also its policies. Adams recounted numerous ways in which viners would at times pay close attention to the letter of the law when it came to the Code of Honor, while violating its spirit. Some users would write exaggerated titles for links they seeded to news stories around the web, skirting Newsvine’s requirement that the headline given by users to a story accurately reflect that article’s content. Others would assign dozens of tags to an item they had submitted in an attempt to give it exposure across many of the site’s topic pages. Adams also pointed to cases that required particular discretion on the part of the site’s human moderators. For example, the Code of Honor forbade “self-promotion” in the sense of users posting links back to their own blogs and homepages elsewhere on the web. But when it came to posting links back to their own Newsvine contributions in comment threads across the site, “That’s tricky to handle, because sometimes [their contribution is] relevant. . . . So it’s a matter of whether they’re overdoing it. . . . One user was commenting on a hockey column and linking to their politics [essay]. And I was like, ‘Okay, this is officially advertising.’” Careful discretion on the part of the moderator was also required when intervening in disputes between users. While the Code of Honor dictated “above all else, respect others,” Adams noted: “The people who are most effective at poking other users until they react are the ones who recognize that it’s okay to attack other peo-

Out of One, Many  142

ple’s content. But then they take it too far and repeatedly write paragraphs upon paragraphs about how stupid the content is. Then people take it personally and fight back.” Against cases like these, you may recall, the Newsvine staff were careful to include the Code of Honor’s final, catch-all clause: “Acts that run contrary to the spirit and purpose of Newsvine, including attempts to circumvent the Code of Honor & User Agreement, are not allowed.” At the other end of the spectrum, however, were users who had become fully invested in the norms of Newsvine and worked to enhance the quality of its community. One user wrote a browser plugin that automatically filtered Newsvine comment threads, preventing low-quality comments with characteristics of spam from appearing in the user’s browser. These “client side” filters proved effective enough that the Newsvine developers considered adopting them into the server software as well. Other users formed a group, “Newsvine Anti-Spam,” that was dedicated to searching out spam postings across the site, aggregating them, and bringing them to the attention of the Newsvine moderators. An Australian user went so far as to build an independent website called GLoS, or “Giant List of Scum,” dedicated to fighting Newsvine spam. It was accessible only by Newsvine moderators to whom she or he had given special login privileges, and contained a searchable database stretching back to 2006 of tens of thousands of user accounts that the site owner had flagged as malicious. The list of flagged accounts was categorized by type of violation (“spambot,” “troll,” “seo,” “advertising,” and so on), and it contained notes on the malicious activities in which the user was engaged, a status message indicating whether or not the issue had been addressed, and the date the account had been flagged. It also contained links that Newsvine moderators could use to jump directly to the moderation view of each offending account, allowing Newsvine staff moderators to quickly access the account deletion tool on Newsvine for each user. While Newsvine’s moderators independently reviewed the violations alleged by the Anti-Spam group and GLoS, they found them to be so reliable overall that scanning the recommendations of these users became a major part of the staff ’s regular moderation routine. Still other users created initiatives, such as the “Summer Writing Challenge,” to encourage valuable contributions to the site. And while attending to the Greenhouse, in practice, largely fell to moderators rather than users— partly as a result of its role as a buffer for marketers and s­ pambots—many users nonetheless assisted in educating new registrants. The Newsvine staff

Friends of Mine, Friends of Theirs  143

even added an extra bit of technical infrastructure to further encourage this behavior, creating a “Newsvine Guide” token and a corresponding badge, visible to other users, denoting holders as being particularly experienced and helpful. Adams recounts, Guides was something I’d been pushing for for awhile when we implemented it. One week I saw that one of those users [who was later named a guide when the program was instituted] had dropped off to the side. And I went back and there was a string of [instances] where she was really giving great advice to new users and the new users were like, “Don’t tell me what to do.” And I said, “You know, that person really deserves more than that.”

Elsewhere, he explained further, [The people who were named Guides] were doing what they’d been doing now for years. We’ve had long-time users who are spam busters, who really care about the welfare and wellbeing of Newsvine, who go around teaching and evangelizing. . . . We wanted some sort of designation that would alert new people, or people who would be more likely to brush off advice, that this was someone you should listen to, someone worthy of respect.

Guides is yet another excellent example not only of heterogeneity in the sense of how technologies were used to support and encourage desirable social behavior on Newsvine, but also of how this process was dialogical. Technologies were not simply put in place to direct the emergence of desirable norms on the site; they were also invented to support and perpetuate desirable behaviors that emerged unbidden. NEWSVINE’S BUSINESS MODEL

At this point, we’ve seen the diversity of tools and strategies implemented by Newsvine aimed at building its user base into a self-policing community, manageable by a small staff, that consistently delivered palatable, and at its best, high-quality, content. We’ve come to understand that the system never achieved perfect results, but at the same time delivered some impressive outcomes. The site broke national news. Enough users became exemplars and teachers of the site’s norms to warrant a program that recognized their efforts. According to its founders, Newsvine developed a reputation for being one of the most difficult sites on the web to abuse

Out of One, Many  144

with malicious posts and automated programs. And by the fall of 2010, one full-time moderator and one part-time staffer were managing a user base of as many as a million registrants. Newsvine.com was in many ways a remarkable achievement of heterogeneous engineering. Chief technical officer Mark Budos recalled: “One of Mike [Davidson]’s biggest fears from the start was he didn’t want to wake up in the morning and come in and see a bunch of spam links on the site. And that, to this day, has never happened. It’s actually one of my favorite things, is that you can go home, go to bed, get up the next morning, and bring the site up. And no one’s been doing any [moderation] for eight hours. And the site has got fresh news from the morning. It’s amazing to me. It just runs itself. It’s really cool.” As we’ve seen “runs itself ” is an exaggeration—Adams noted that “things tend to go really badly when there’s no moderation presence”—but at the same time it was true enough that moderators could, for instance, take vacations, and the site had survived for extended periods with little or no moderation. Finally, and perhaps more importantly to a startup, “runs itself ” was true enough in the period after the site’s founding that the company could sell ads next to its content and enter into deals with traditional news corporations. While noble ends in themselves, ultimately the point of keeping the site free of malicious activity and surfacing high quality content was to turn a profit and pay back investors. In addition to ad sales, after its founding Newsvine sought partnerships with mainstream news providers. The most notable of these came in December of 2006 when the New York Times website, NYTimes.com, added the first-ever content-sharing tools to its site and selected Newsvine as one of three social networks (with Digg and Facebook) that would have “share” buttons next to New York Times content.12 Of the three chosen networks, two—Newsvine and Digg—were approached by the Times, and Digg was already a major driver of traffic in 2006. Facebook approached the Times for inclusion after it added its News Feed feature. At the time, Davidson told the Seattle Post-Intelligencer that Newsvine was “the up-and-comer” of the group (Cook 2006); it was a far smaller website than either Digg or Facebook, but it was growing quickly and according to Davidson, the Times editors appreciated the elevated level of discussion that took place on the site. The New York Times itself had no discussion threads on its stories at the time, and sending users to Newsvine was effectively a method of outsourcing the community features of its site.13 The press at the time described

Friends of Mine, Friends of Theirs  145

the partnership between the paper and Newsvine as strictly a technology deal (Cook 2006)—no revenue sharing was involved—but there were two potentially huge advantages for Newsvine: inclusion on the Times website promised to give Newsvine a great deal of exposure, driving traffic and ad revenue its way; and the deal would potentially lead to additional partnerships with the Times, and perhaps other sites as well, that could prove lucrative to the start-up in similar or additional ways. Newsvine built a share button for use on NYTimes.com, but its work on the arrangement did not stop there; it needed to modify its own site as well. Under the existing design of the Newsvine site, new users—precisely the commodity Newsvine hoped to gain from its deal with the Times— would enter the site as untrusted users and be relegated to the Greenhouse, as would their newly seeded New York Times content. That would have two negative effects. First, it would limit the exposure of NYTimes.com content in ways unlikely to be attractive to the paper. Second, unlike the early adopters who made up the bulk of Newsvine’s user base at the time, many users coming from NYTimes.com would simply be looking for a place to discuss the story they’d shared, and dropping in as they did from “outer space” were less likely to understand the restrictions placed on them as new registrants to the site. So Newsvine added an exception to its algorithm. Users whose first act on Newsvine was to seed a New York Times story skipped the Greenhouse entirely, which in turn made their content immediately eligible for inclusion on the front page and section fronts across the site.14 Along with the numerous other social, technical, commercial, and ­policy-oriented interventions we’ve encountered thus far, the modifications to the Greenhouse system demonstrate how the creation and socialization of the Newsvine user base was central to Newsvine’s ability to draw in a major partner. It’s also possible to see how that partnership reciprocally impacted the design of Newsvine and its Greenhouse. The modifications made to the Greenhouse were an attempt to carefully balance the interests of the existing Newsvine community, the traffic and community needs of the New York Times and its audience, and the financial interests of Newsvine itself. As it turned out, the technological solution that they devised for lashing together these recalcitrant resources would have a major influence on the next phase of the company’s existence, as well as on the company that was about to acquire it: MSNBC.com.

8//different things to different people newsvine and msnbc.com

In 2007, Newsvine began work on its first major site upgrade. The aim of the updates to the site’s codebase—an architectural revision the team dubbed the “M2 framework”—was to make Newsvine customizable by users. At the time, other social networks like MySpace were allowing users to alter their profiles using “modules” or “widgets”—customizable displays of personally relevant information that could be added to, removed from, swapped out, and rearranged on profile pages by users themselves. Similarly, several services along the lines of Netvibes and Pageflakes had begun to spring up, offering users the opportunity to create a personalized dashboard-style homepage by mixing, matching, and even creating widgets that displayed personalized headlines, weather, email, calendars, and more. At the time, the ability to customize a profile or a homepage seemed to the Newsvine team to be a growing trend worth capitalizing on, and one they hoped would help to increase the popularity of the site. The M2 framework, as envisioned by the startup, would allow users to customize their Newsvine homepage. Rather than being stuck with the site’s default settings, users could instead replace the various sections of the front page with their own choice of headline modules—feeds of news from various sections or groups on the site, particular news sources, regional headlines, local weather, and so forth. They would also be able to customize their own Newsvine profile pages in a similar fashion, choosing what information to show off to visitors. Not only would the content of pages be customizable by viners, but so would the layout—the M2 framework would allow users to drag and drop the various modules, arranging them in any order they saw fit.1 146

Different Things to Different People  147

Figure 8.1. The Newsvine homepage in 2011. Note the icons in the corner of each frame, indicating the frames are modules that could be swapped out by users for custom content. NBCUniversal.

The M2 update began with an initiative the developers called the “Tax Man Project,” nicknamed for its intended release on April 15, the U.S. tax deadline (it ultimately launched on April 24, 2007). The Tax Man Project, known to users as “Newsvine 2.0,” culminated in a widgetized redesign of the front page that looked more or less like the one still in use at the time of my fieldwork in 2010 (Figure 8.1). The developers started writing a variety of modules for use with M2, and even briefly considered the idea of allowing users to write modules themselves. Work also began on the M2 profile pages, which ultimately became available to users upon request as a beta feature, but were never rolled out across the entire site (Figure 8.2). Newsvine, as it turned out, was about to be acquired by one of the largest news sites in the country: MSNBC.com. As Lance Anderson, Newsvine’s director of front end development, described it,

Out of One, Many  148

Figure 8.2. M2 customizable profile page. The profile subject had beta access to the M2 profile; she was also hired as a part-time staffer. This profile appears with the permission of the user. NBCUniversal.

We put so much work into modules [“widgets”]. We had a little module directory where you could add modules [Figure 8.3]. And before even imagining other people creating widgets, I thought we’d have more resources and more time where we could create more cool widgets [ourselves] for our users. . . . [For instance,] Josh [Yockey] has this cool algorithm that, based on your viewing on Newsvine, can recommend stories to you. [I’d imagined more] widgets like that that we could have put on the front page and we just never did. And I think there’s a host of reasons for that. One of the big ones is that a month after we launched that front page, MSNBC contacted us and we went into negotiations with them. And that changed our whole course. . . . We were running out of money and I think we realized that creating a dozen cool widgets on our front page wasn’t going to quadruple our user base. [laughs] So we were trying to look at things that would help give us a jump in [user] registrations. Acquisition by MSNBC.com, then, became a solution for Newsvine as a system builder. It provided an answer to Newsvine’s flagging finances

Different Things to Different People  149

Figure 8.3. Newsvine module directory in 2011. NBCUniversal.

and gave it a way to pay off investors without engaging in another round of financing. It was a deal that allowed a startup, one that by its founders’ own admission was low on reserves, to become what an industry observer deemed “one of [VC firm] Second Avenue’s most successful investments.” The acquisition should not be seen as an act of desperation, however. The startup’s leaders had talked with corporate suitors in the past (Newsvine 2007a), but come to the conclusion that MSNBC.com would offer several unique advantages. First, as one of the top three news sites in the United States, MSNBC.com had a remarkable base of 29 million monthly users, which meant that enormous amounts of traffic would be funneled to Newsvine (A. Johnson 2007; Newsvine 2007a). In addition, MSNBC.com had the potential to offer Newsvine substantial infrastructure improvements to help handle all the additional traffic—with the help of Microsoft, it operated a world-class data center, which would give the social news site “vastly more server power and reliability, which Davidson said . . . had been the site’s most significant challenge” (A. Johnson 2007). And MSNBC.com was headquartered nearby in the Seattle suburbs, which meant there would be no need for Newsvine to relocate or for the staff to fly cross-country on a regular basis. The new corporate parent would also provide an additional source of professional news content to the site beyond the existing Associated Press wire stories, and its financial resources would allow Newsvine to grow its own development staff (Newsvine 2007a). Additionally, MSNBC.com would allow Newsvine to continue operating independently as its own brand. As Newsvine put it in the acquisition announcement on their blog,

Out of One, Many  150

As desired by both companies from the outset, Newsvine will continue operating independently, just as it has been since launching in March of 2006. Msnbc.com is committed to maintaining and growing the community and features that have made Newsvine what it is today. In other words, Newsvine will be the same Newsvine you’ve always known, only stronger. Over the next few years, Newsvine technology and content will make its way onto msnbc.com, and vice-versa where it makes sense. Our users’ columns and content will remain as they have been, as we work hard with msnbc.com to implement improvements and enhancements. (Newsvine 2007a) Out of a sense of fairness—and to grow its user base—Newsvine had always promised members revenue sharing on advertising, and while this spreading of the wealth didn’t extend to ownership of the company, it framed user contributions as valuable commodities. A few users, as a result, developed a sense of financial entitlement toward the site, and when Newsvine was bought out for a large, if undisclosed (A. Johnson 2007), sum on the basis of their content, some inevitably took umbrage with the deal.2 To minimize this sort of reaction, news about the acquisition was carefully framed by the founders. At the time of the announcement, they were sincerely effusive in their thanks to the community. On his own blog, COO Calvin Tang (2007a) wrote a note of appreciation in which he stated that, beyond the technical features provided by Newsvine, “the open dialogues, the free and creative expression of ideas and the genuine manner in which all of you participate on the site are some of the foremost reasons that msnbc.com found Newsvine to be an attractive company to partner with” (para. 1). Mike Davidson expressed similar sentiments in his comments on the Newsvine blog (Newsvine 2007a), and in the post announcing the acquisition he was careful to emphasize the benefits of the deal for the Newsvine community, including: Increased exposure for Newsvine writers. Remember when Killfile broke the news of the Virginia Tech shooting 22 minutes before the Associated Press? What about when Corey Spring got an exclusive interview with Dave Chappelle? When important moments like these occur on Newsvine, why shouldn’t they also be put in front of 29 million people on msnbc.com? What about when a Newsviner builds up an audience for a weekly entertainment column like Steve Watts’ Lost in the Vines? Why shouldn’t great content like that be put on an even bigger stage? We think it should, and although Newsvine and msnbc

Different Things to Different People  151

.com will remain independent brands, we’re going to spend the next several months figuring out ways to get the best content in front of the biggest audiences possible. (Newsvine 2007a) The announcement and the ensuing comment thread, in which the Newsvine founders reiterated their promise of increased traffic and potential placement on MSNBC.com and carefully responded to the questions, concerns, and criticisms of users, represent acts of heterogeneous engineering, as the startup went about getting its essential and carefully constructed—but also recalcitrant—user base on board with the company’s plans for expansion and financial stability. The notion of giving users a bigger microphone via MSNBC.com was obviously intended to help sell the idea of the acquisition to the Newsvine community, but it was not a disingenuous promise. For example, several prominent users received press passes to the 2008 national party conventions to report for Newsvine on behalf of MSNBC.com. And two of News vine.com’s prominent members, Viki Gonia and Chris Thomas, were put on the air shortly after the acquisition at MSNBC TV to describe their, and the community’s, reactions to events at the Democratic National Convention (Newsvine 2007b; Tang 2007b). We can see, then, how Newsvine’s acquisition placed it at the center of revenue sharing and crowdsourcing labor debates that persist even today.3 The sharing of ad revenue, while it helped Newsvine avoid the negative connotations of crowdsourcing as a form of exploitative unpaid labor (Terranova 2000; Fuchs 2009) or digital “serfdom” (Friedl 2006), proved to be a somewhat sticky issue in the new context of acquisition, and did not entirely exempt the company from criticism by its users. As we’ll see, this was only one of the carefully refined aspects of Newsvine’s system that would prove recalcitrant under the new context of MSNBC.com o ­ wnership. MSNBC.COM AS A HETEROGENEOUS ENGINEER

So far we’ve looked at MSNBC.com as an actor in the system that Newsvine was constructing. As an initial exercise in vantage-point switching, it’s useful to briefly consider what MSNBC.com was hoping to do with Newsvine. When the dot-com crash hit in 2000, MSNBC (at that time all one company) had “shelved ambitious plans to go public and acquire complementary properties” (A. Johnson 2007). After MSNBC.com became independent of the MSNBC cable channel, however, it gradually returned to the idea of growing its offerings through acquisitions.

Out of One, Many  152

During this period, online news providers were also increasingly fashioning themselves as providers of public infrastructure, to use a term from Mike Ananny (2013), opening up spaces for public participation and conversation online in relation to their offerings. Newsvine was to become the joint venture’s first acquisition in its eleven-year history, and one of the first acquisitions by any mainstream media company of a social news site (Arrington 2007; Stelter 2007). Since its launch in 1996, MSNBC.com had become a traffic juggernaut, but according to the site’s own report on the merger, “While msnbc.com has long ranked among the three most popular news sites on the web, it has been late to the game in expanding its offerings in user participation and non-professional reporting” (A. Johnson 2007). For instance, MSNBC.com had limited comment functionality at the time; by contrast one of the company’s leading competitors, AOL, allowed comments on every story, and another, CNN.com, was heavily touting its iReport citizen journalism initiative (Stelter 2007). Social networking functionality on websites was also becoming a major trend. At the time, MSNBC.com president Charles Tillinghast told reporters that Newsvine offered “a lot of interesting features” that the parent company could “evaluate and port to msnbc . . . rather than us creating a feature they already have” (A. Johnson 2007). In an interview in 2010, he recalled, “Social networking was really hitting its stride at that time. And we felt we needed a position in that space. It was a question of buy versus build. We didn’t have anybody who was an expert on social networking who worked for us. So ‘buy’ meant hire, then build, as opposed to just buy. In buying Newsvine we picked up technology that we were able to integrate throughout all of our sites for commenting, blogs, ratings—and we also picked up the people who could help us to think through how to integrate social networking into our products.” Similarly, at the time of the acquisition, one of the MSNBC.com executives involved in the deal wrote on his personal website, “We plan to leave Newsvine alone—learn from it, integrate little pieces of it, watch it grow. . . . [M]eanwhile, we will incrementally find sensible ways to integrate the ‘social thinking’ of Newsvine into the ‘big media thinking’ of MSNBC.com” (Sorgatz 2007). MSNBC.com executives also appreciated that Newsvine was a small organization at the time of its acquisition and that it was nearby geographically. According to the parent company’s own news story on the acquisition, “‘Newsvine is local, small, nimble—they don’t come with a lot of

Different Things to Different People  153

things you don’t want,’ [Tillinghast] said, such as complicated partnerships and contracts. ‘There isn’t a lot to rearrange.’” At this point, we’ve begun to see what the acquisition of Newsvine accomplished for the heterogeneous engineers at MSNBC.com. With its purchase, MSNBC.com was attempting to play catch-up with its competition in terms of social engagement with readers, as well as the technology behind such engagement. It had acquired a group of experienced developers with a mature social networking platform, and without taking on the same amount of organizational inertia and baggage that would have come from the purchase of a larger firm. Several sources also noted that Newsvine offered not just the technology and space to facilitate community, but also a particularly valuable existing user base. At the time, Newsvine traffic was at 1.2 million visitors monthly and already growing at a rate of 46 percent a quarter. Meanwhile, its users were highly engaged, viewing an average of twenty-one pages per day, and contributing tens of thousands of comments and hundreds of thousands of votes each month (MacManus 2011). This group of highly engaged users was potentially a very lucrative one for MSNBC.com, but at the time, MSNBC.com had relatively little to offer it. Newsvine was, in many senses, intended to be a solution to this predicament. Mike Davidson described the demographic, dubbed “news explorers,” that MSNBC.com was pursuing through this acquisition: News explorers are the very high-involvement people who have an above-average interest in news, who are very likely to forward stories to friends, who are very likely to comment on stories themselves, who aren’t satisfied reading a quick, three-paragraph AP blurb about a topic. There are plenty of people out there who only care about reading a three- or four-paragraph AP blurb about a topic. In fact, I would say that number is well over 50 percent of all news consumers. But if you take the rest of those people [the news explorers], they are the ones that generate a ton of page views and spend an inordinate amount of time on your site. We think of ourselves as the site within the company [MSNBC.com] that is specifically geared to deal with those sorts of people and to make the most out of them. AFTER THE ACQUISITION: NEWSVINE ADAPTS

Looked at from the vantage points of their respective builders, neither Newsvine nor MSNBC.com was a completely stable system in the time

Out of One, Many  154

leading up to the acquisition. Newsvine needed a buyer to help it stabilize its finances, while MSNBC.com needed some form of community offering in order to capture a highly engaged and potentially lucrative subset of the news market that its competitors were beginning to pursue aggressively. But much had been achieved, through the heterogeneous engineering that the companies had done to that point. Newsvine had developed a productive, pro-social, and rapidly growing user base of “news explorers” and “citizen journalists.” MSNBC.com was dominating the more traditional news market online. In other words, large pieces, if not the whole, of each system had been stabilized from the vantage point of the respective builders prior to the acquisition. As we’ve seen, however, the appearance of stability is in fact maintained through isometric pressure. Returning to the vantage point of Newsvine, we have come to understand how much continual effort went into maintaining this user base, and how many different recalcitrant actors—­ technological, financial, policy-oriented, and social—had to be lashed together and constantly maintained to attain this “stability.” In taking on MSNBC.com as a corporate parent, Newsvine had lashed another recalcitrant actor into its evolving system, and the stable pieces of the Newsvine system would soon become dynamic under the new balance of pressures. As Domingo (2008) and others (Hemmingway 2005; Schmitz Weiss and Domingo 2010) have pointed out, adding an actor to a system in this way will often lead to a shift in the relationships in the existing network, between components of the system—in our terms, it will necessitate further heterogeneous engineering to restabilize. In what follows, I will look at how the addition of MSNBC.com as a corporate parent affected Newsvine and the user community it had so carefully and relentlessly engineered. We’ve discussed the advantages that Newsvine hoped MSNBC.com would provide as an actor in its system. But as we’ve learned, all actors within a system are recalcitrant—they solve some problems and create others. What advantages did MSNBC.com ultimately provide and what changes were necessitated to cope with its recalcitrance? The Solution Stack: The Only Unix Guy at MSNBC

Newsvine, as a frugal startup, built its entire system on a solution stack that was predominantly free and open source. In selling to MSNBC.com, the Newsvine staff hoped to take advantage of the parent company’s tremen-

Different Things to Different People  155

dous resources for handling heavy web traffic. MSNBC.com and its data center, however, were built on a proprietary Microsoft solution stack. As Josh Yockey explained, We built up a large codebase in PHP, and some Apache. . . . And then early on when we got acquired the big question was, “Do we want to migrate all of our functionality to what they [MSNBC.com] did?” Which is C# [a Microsoft-developed programming language] running through IIS [Microsoft’s proprietary server software]. And we decided the amount of time it would take to do all that could be better spent on making new features. So we stayed on Linux-Apache the whole time. Which has been an issue, because they don’t have operational personnel that know anything about it. So we ended up retaining a much, much larger amount of operational responsibility than we wanted to. You know, when we were a startup we understood if the database server has a bad drive or something it’s our job as engineers to drive out and fix it, because there’s nobody else there. But our hope when we got acquired was, “Oh, finally I don’t have to get woken up at one in the morning because there’s a RAID [hardware] failure or something.”4 But in fact three years later we still get woken up at one in the morning because there’s a RAID failure. Because the operational team over there [at Microsoft] just doesn’t have the expertise. To fill this breach, Newsvine eventually made a new hire—enrolled a new actor in its system—an experienced Linux system administrator named Eric Flamoe, referred to jokingly as “the only Unix guy at MSNBC.”5 His job was to keep the hardware and low-level systems at Newsvine running smoothly, as well as to gradually educate the tech support team for MSNBC.com, Microsoft Operations Support (MOS), on how to handle common problems with Linux servers. As a self-described “liaison” between the Windows experts at Microsoft and the Linux-based world of Newsvine, Flamoe also learned a great deal about the way MSNBC.com’s data center worked. We can see here the sense in which Newsvine found MSNBC.com to be a recalcitrant actor—it did not, in the end, provide a simple plug-andplay solution to the startup’s need for scalable web hosting. But hiring a dedicated Linux systems administrator, who would double as a Microsoft liaison, was only one of the potential solutions imaginable for dealing with this issue. After all, from the vantage point of MSNBC.com and the system

Out of One, Many  156

it was trying to build, Newsvine and its alternative solution stack were the recalcitrant actors, and we can also imagine a world in which Newsvine solved its problem by rewriting its software to run on MSNBC.com’s solution stack. To fully appreciate how Flamoe’s hire threads a needle between all the interests involved, we have to understand why this alternative solution wasn’t adopted. “On the technical side,” said Josh Yockey, I would say we’ve repeatedly been asked, “Why don’t we just run our stuff on C#, on .NET? And that would make everything interact much better because that’s how everything runs at MSNBC.” And that’s certainly possible. I wouldn’t say we’re totally inflexible on it, but obviously anyone who requests that is dramatically underestimating the amount of work it would take. . . . We might replace PHP with s­ omething. . . . [But] the OS and the core platform [are] pretty set. Other than that, we can pretty much do anything. The development team at Newsvine prioritized speed to market and prided itself on the ability to respond flexibly to requests. Rewriting at least parts of their software, then, was never out of the question, but doing so with a small development team—remember, Newsvine had fewer than ten employees—would have quickly put a strain on resources. On the one hand, all of the developers felt that being part of a small group was advantageous in many ways—every founding member of the team was intimately familiar with the software’s architecture, and the ability to fit the entire development staff into a small shared office suite eliminated much of the bureaucracy and the “meeting culture” that many saw as being lethal to productivity. Newsvine’s CTO Mark Budos explained, “I’ve been doing this a long time and a dev team that’s any bigger than five or six will fail. . . . You can’t have 10, 15, 20 people working on one thing. Your productivity just dies.” On the other hand, as we’ll see, upon its acquisition Newsvine became responsible for a great number of additional tasks, meaning that productive and flexible or not, the development team was spread very thin. Originally, Newsvine had hoped that in taking on a corporate parent, it would acquire the necessary resources to substantively expand its staff, both to accelerate development on Newsvine.com itself and to assist in development tasks for the parent company. Again, however, it encountered recalcitrance, not just from MSNBC.com, but also in ways related to the nation’s financial meltdown in 2008. Davidson explained, “Right after we got purchased, that was when the economy hit the skids. There were hir-

Different Things to Different People  157

ing freezes all over the media industry, and teams weren’t generally getting bigger; they were getting smaller. Instead of going from six people to twenty people, like maybe we had expected, we went from six people to eight ­people.” At this point we’re able to see how the decision to hire a Linux system administrator was a way of answering to a variety of pressures and interests. It is a response to the pressure levied simultaneously by the ­economy— instead of fifteen new staff members, the enrollment of MSNBC.com allowed the hire of only a handful of new employees—as well as by the proprietary systems at Microsoft and MSNBC.com. Because of these limitations, it was impossible to hire the number of new developers necessary to help shoulder the intense work required to rewrite Newsvine for a Microsoft solution stack. The recruitment, instead, of a single Linux system administrator took pressure off the developers to fix hardware problems, and allowed them to focus on new features and responsibilities. Flamoe’s efforts to be a liaison between Newsvine and Microsoft’s support team also alleviated some of the absence of operational assistance for Newsvine.com’s largely open source solution stack. We can also see the heterogeneous aspect of the concerns addressed here, which are simultaneously technical, financial, and policy-­ oriented (Microsoft Operations Support had developed standard policies and procedures for dealing with problems on Microsoft-owned websites that in some cases had to be amended to support Newsvine). The User Base: “That’s Not Really Acceptable Behavior”

For legacy news organizations, embracing the idea of “the public” is one thing; dealing with an unruly torrent of user-generated content often proves to be quite another. In 2007, for instance, CBSNews.com ultimately made a decision to temporarily shut down comment threads on all stories related to then-senator and presidential candidate Barack Obama. Mike Sims, the director of news and operations for CBSNews.com, told the press that the site could not immediately find a way to deal with the “persistence and volume” of racist comments that the stories were attracting. While some appreciated the move, it also led to criticism of the site, with some commentators claiming that not allowing readers to see and react to racist comments effectively hid the important issue of racism in the campaign (Seelye 2007). That a major network would invest so heavily in designing interactive features into its site only to shut them down drives home the profound challenge that comment moderation can pose. This is particularly true for

Out of One, Many  158

legacy media sites whose twentieth-century brand images carry deeply embedded cultural currency, as compared to their social media counterparts. MSNBC.com web producer Will Femia noted that a news site and the discussions it hosts will inevitably be “assigned a character by individuals who are on whatever crusade they may be on.” The news organization becomes a symbol—or rather a whole set of symbols with cultural currency—to which different individuals and groups impute different meanings: “NBC News means different things to different people, but it means something. And MSNBC means something. And Microsoft means something. And even within that, our sub-communities mean things beyond what we can control. It just doesn’t matter [what you do]. You’re going to get a Fox News guy in there somewhere. Or you’re going to get an anti-corporate guy in there. To some extent there’s nothing you can do.” These challenges extend to all the U.S. networks and cable news organizations. In 2010, ABCNews.com, for instance, found it necessary to retain a team of moderators whose sole job was to review every comment posted to the site. Until May 2010, MSNBC.com pre-moderated all of its blog comments, reviewing each and every one before it went on the site. This is commensurate with the evidence gathered by Alfred Hermida and Neil Thurman (2008) around the same period, who noted that after some initial experimentation with unmoderated comment spaces, U.K. news websites were likewise overwhelmingly shifting to pre-moderating their user comments, despite finding the task to be incredibly resource-intensive. The worry among legacy news providers has generally been that unsavory comments by users might tarnish the civil and/or objective brand image of the news provider. As Michelle Levi, a web producer at CBSNews.com, put it in 2009, “There’s a quote [placed between a blog entry and the ensuing user comments] saying ‘CBS News content stops here. We’re not responsible for these things.’ You are giving a little bit of a playground for people to say what they want. But it’s still on a webpage that has the name ‘CBS News’ on top of it. And no matter what, it will be associated with that news division.” Even when well-moderated, user comments don’t always constitute a conversation or a discourse—civil or otherwise. And poor moderation on a news site potentially exposes the organization to legal liability. In some cases libel laws apply, and there are other risks as well—if a commenter exposes an anonymous source, for instance. Moreover, it’s been noted that vicious comments on stories can in fact scare away journalists’ hard-won sources (Davenport 2010).

Different Things to Different People  159

Throughout the 2000s, at many news organizations, and especially national television news organizations, wariness of these risks led to what Hermida and Thurman (2008) referred to as a strong gatekeeping approach when managing the inclusion of user comments on legacy media websites, meaning that journalists in these cases tended to view themselves as arbiters of what content was fit for public consumption. Whereas Newsvine.com was a social network built to allow users to monitor and moderate one another’s contributions, the “gatekeeping approach” was in favor for quite some time at MSNBC.com. According to the Newsvine staff—as well as a number of editors and producers I talked with at the parent company—a preference among journalists for the gatekeeping strategy led to a fair amount of resistance to the inclusion of Newsvine’s style of community comment moderation on the primary MSNBC .com domain. At the time of the acquisition, MSNBC.com described itself as “racing to foster a community among its readers and to exploit the power of unmoderated user commentary and ranking of the news” (A. Johnson 2007). At the same time, though, it was not racing to have these features appear on its own flagship domain. Tillinghast during that period described the flagship site’s position on the inclusion of Newsvine’s functionality as “evaluate and port” (A. Johnson 2007), meaning that the site would make decisions carefully over time as to which Newsvine features would work for MSNBC.com. Initially, as opposed to adding Newsvine comment threads to all its stories, MSNBC.com added a link titled “Discuss on Newsvine,” which when clicked would send users away from MSNBC.com to a version of the story appearing on Newsvine.com. This approach was what Mike Davidson described as MSNBC.com’s “arm’s-length community concept”: We [at Newsvine] just don’t really do any filtering at all. And so, the reason why it’s nice to have a slight brand separation between the two [sites] is that we can open up comment threads on Newsvine for MSNBC stories, and because they’re branded Newsvine, it’s not as important as it would otherwise be to tightly control what shows up. As you know, from reading user comments on really any site that allows them, sometimes comments are awful, and those ones come down. But other times there are comments that are not worth deleting, but not really the sorts of things that you would want to show up on your flagship site, necessarily.

Out of One, Many  160

In other words, the moment when a user clicked “discuss” on an MSNBC.com story or jumped to Chuck Todd’s space on Newsvine, they were dumped—somewhat jarringly—into an entirely different site, with vastly different branding, layouts, and user interfaces. The setup took the small disclaimer mentioned by Michelle Levi, “our content stops here” and blew it up into a sort of virtual border crossing. Users were allowed greater freedom on Newsvine, but at the expense, perhaps, of their direct involvement with the network and cable channel, which shuffled them off of their flagship domain entirely. Initially, it didn’t appear as though this level of integration would be all that challenging. Newsvine had already developed an integrated button for seeding content during its partnership with the New York Times website. All that had to be done was to put the same button on MSNBC.com and change the wording from “Seed to Newsvine” to “Discuss on Newsvine.” Without realizing it, the first reader to click the link would actually be clipping the article to Newsvine, in much the same way that she would share it to another social news site like Digg or reddit. The primary difference was in how the action was framed for the user.6 When the “Discuss on Newsvine” link was clicked, the user would arrive at the article on Newsvine.com, with a comment thread below, not necessarily realizing that she had been the one to initiate its placement there. Unregistered users would, of course, be asked to create a Newsvine account during this process. When subsequent users clicked the “Discuss on Newsvine” link, Newsvine’s software would—as it would for any other article—identify that the link had already been seeded to the site and simply take the user to the existing Newsvine post and its comment thread. A fringe benefit of this, for MSNBC.com, was that stories no one felt like commenting on would not turn up on Newsvine, so all MSNBC.com content there would appear discussion-worthy. Here we can see how Newsvine was able to marshal its existing code for a new purpose. The move was an efficient use of the company’s limited resources, and it worked extremely well from a technical standpoint. But as we’ve already seen time and again, problems are never exclusively technical, and the “Discuss on Newsvine” button would prove recalcitrant. Several things happened as a result of its implementation that would require additional heterogeneous engineering. First, with tens of millions of monthly visitors to MSNBC.com, Newsvine opened itself up to a “firehose” of MSNBC.com seeds and traffic, which

Different Things to Different People  161

at times overwhelmed the site’s front page. Lance Anderson recalled, “We were getting so much traffic from MSNBC stories that our front page was becoming dominated by MSNBC content. And it was right after the acquisition. So, of course, a lot of our community users said, ‘Argh! They’re being taken over by MSNBC!’ [But the truth was], ‘No, really we just need to write some code to clean it up. It’s not this big conspiracy.’ . . . Now we have [MSNBC.com stories] all funneled down to one little module on the front page. All the other modules on the front page, they’re [no longer] allowed to have MSNBC-sourced content.” The additional code was thus part of yet another delicate balancing act, alleviating the concerns of Newsvine’s user base (and so lessening the potential impact on its citizen media brand image), while answering the desire of its new corporate parent for access to the site’s community. But even with the technical fix that relegated MSNBC.com content to its own small box on the front page, Newsvine began acquiring a great number of new users who were not “friends of mine and friends of theirs.” Many of them had never heard of Newsvine or the Code of Honor. And as you might expect, on a site that relied so heavily on a carefully engineered and largely self-policing user base, this began to have repercussions. The Newsvine blog entry from this time in mid 2008 reads, in part, “[The influx of new users] is a good thing, of course . . . but it’s [also] potentially a bad thing because as we all know, when new users come en masse to a Newsvine thread, they aren’t always aware of how we do things around here. We were all new at one time. Please welcome in our new visitors as you would a new guest to your own cocktail party, with diplomacy and patience. We hope that the Newsvine community puts its best foot forward when receiving new participants into the collective discussion.” The post listed several of Newsvine’s initial technical responses to this problem, including adding a visual “Greenhouse” designation to new users’ comments, to flag for the existing community which users should be given helpful advice and/or a wide berth. In light of the instant popularity on the site of MSNBC and NBC News personalities, Newsvine also coded an exception to prevent any user with a token that indicated s/he worked for Newsvine, NBC News, MSNBC, or MSNBC.com from appearing on the site’s “Leaderboard,” a page displaying the stats of Newsvine’s most popular users. But the post, and particularly the excerpt quoted earlier, are also interesting in the manner in which they lay out a heterogeneous solution to the problems incurred by the “Discuss on Newsvine” button. Techno-

Out of One, Many  162

logical fixes were being delivered, yes, but the post is also an entreaty to experienced Newsvine users to help socialize the unknown masses from MSNBC.com, to make them amenable to the site’s system of moderation. And interestingly, a technical fix—the visual Greenhouse designation for comments—was used to help identify users in need of social engineering. But the problems did not end there. MSNBC.com stories continued to be disproportionately popular on Newsvine, owing to the firehose of users coming from MSNBC.com. This flow only increased starting shortly after the acquisition in 2008, as the United States entered into a presidential election cycle—a time when news sites generally do their best traffic, and MSNBC’s political brand becomes especially valuable. These surges of traffic during the election cycle compounded a side effect of using the pre-acquisition story-seeding functionality to provide Newsvine discussion threads for MSNBC.com stories, which was that the first user to click “Discuss” on an MSNBC.com story didn’t just share it to Newsvine—they added it to their own blog column. This resulted in a number of major issues, for two reasons: Newsvine.com had promised users a cut of the ad revenue when they generated a popular column, and Newsvine users, under the Code of Honor, were responsible for moderating and maintaining civil conversation in their own column. As Newsvine’s staff moderator Tyler Adams explained, If you arrived on the front page of MSNBC.com at 8:30 in the morning and Sarah Palin had been selected as the Vice Presidential nominee, and you clicked on that, it went into your column. And that could have been you registering for Newsvine for the first time. And then you had a 30,000-comment thread in your column. One, you would be terrified if you were new. And two, you would make a bunch of money. And so there were people who would go around every morning and seed the top ten articles on [MSNBC.com], or try to. . . . Those people made a lot of money off of Newsvine and did no moderation. Even if they had tried to do moderation, they were signing up for reviewing 15,000 comments a day or something like that. The system, in short, terrified new users unexpectedly and encouraged those who became more familiar with Newsvine’s workings to game the system to their own financial advantage. It also took about half a year to fix. With their added post-acquisition responsibilities it wasn’t until January 2009 that the developers were able to roll out a workaround, which clipped

Different Things to Different People  163

“Discuss on Newsvine” stories from MSNBC.com to generic columns corresponding to their section headings on MSNBC.com, such as “Business” or “Entertainment” (Newsvine 2009). As we saw earlier, a major aspect of Newsvine’s heterogeneity was that many of the components that made it work—that allowed for community building and pro-social discussion—weren’t technological actors at all. Even without MSNBC.com’s robust data center, in many ways Newsvine’s servers and software performed admirably through the acquisition and increased MSNBC.com traffic levels. In its first two and a half years of operation, the service experienced only three extended periods of downtime (Newsvine 2008d). Much of this success was due to the development team’s experience at Disney. Unlike many startups, which experience major technical problems as they grow, Newsvine, from its inception, was built for scale. As Budos put it, “Josh [Yockey] and I are freaks. He worked on Fantasy Football at ESPN, which is extremely popular. . . . I worked on an ads system that served a billion ads a day. . . . We architected for scalability from the get-go.” Rather, if Newsvine was in many ways an isometrically stable system before MSNBC.com, the pieces of the system that came under the greatest divergent pressure after its acquisition had to do with its cultural components, as opposed to its technical ones. In the original system, Newsvine’s technology depended on Newsvine’s user culture to the extent that it made little sense to separate the two—they were both integrated and necessary parts of a heterogeneous phenomenon. But even after MSNBC.com stories had been prevented from overwhelming Newsvine’s front page and removed from the columns of individuals, MSNBC.com traffic and registrations continued to make it more difficult to maintain the user culture that had made the original Newsvine work. Mike Davidson remarked, The people that come over to Newsvine in that way [from MSNBC .com “Discuss” links] are not familiar with the ethos of Newsvine. . . . They feel like they are about to leave a comment on a major media site. Sometimes there are bad behaviors that go along with that. If you’re reading some article—or watching some video—by a pundit on MSNBC . . . [and] you want to leave a comment about it, flaming what you just saw, you’re not necessarily aware that that’s not really acceptable behavior on Newsvine. That’s not really what people do. And so, often those threads are the toughest ones to moderate, because you have people who don’t know the rules.

Out of One, Many  164

There were those in Newsvine’s core user base who answered the developer blog’s clarion call to help draw MSNBC.com users into the Newsvine community and help acquaint them with the site’s norms. But those users whom Davidson just described, people who viewed the Newsvine site as a normal comment thread on a mainstream news site, who then responded to—or worse, flamed—a single story and left, never to return, were opting out of Newsvine’s reputation economy and normative structure entirely. Although the growth of online communities frequently leads to escalating “conflict and coordination costs” (Kittur et al. 2007), and challenges to established norms—including flames—can at times lead to better articulations of a community’s existing values (Franco et al. 1995), what was happening at Newsvine was the introduction of a ceaseless, directed flow of users who not only didn’t share the community’s norms, but mistook it for something else entirely. Josh Yockey noted, At first the Newsvine users tried to interact with them as if they were part of the Newsvine community. So they’d reply to their comments and teach them about the Code of Honor if they looked like they were violating it. And I think eventually most of them just stopped. Because they realized these people weren’t coming back. . . . I think a lot of the Newsvine people just realized that it wasn’t worth going into these ten thousand comment-long threads and thoughtfully replying to each one of them. “Excuse me, you should educate yourself on the Code of Honor.” You know, they just got out. . . . [That was one of the] the major disruptions in the community. INVENTING THE FUTURE: BUILDING SYSTEMS FOR MSNBC.COM

You’ll recall that when Tillinghast described the acquisition of Newsvine, he mentioned that the company was in part purchasing a “buy or build” corpus of social networking and user participation features that could be applied or ported to MSNBC.com. When new editorially oriented development projects came up at MSNBC.com, a team within the company called Creative Development decided how they would be implemented, dividing up the work among its staff and deciding which parts to outsource or give to other internal development teams. Upon its acquisition by MSNBC.com, Newsvine became a unit within the MSNBC Digital Network to which these sorts of tasks could be assigned, and as requests rolled in they gradually became increasingly responsible for building features for

Different Things to Different People  165

MSNBC.com—especially to support its user-participation functions as well as those of NBC News and MSNBC TV. Over the years these features included the Nightly News Discussion Club, an online discussion forum accompanying the broadcast of Nightly News that prefigured many aspects of the contemporary efforts of network television at hosting a “two-screen experience” via social media like Twitter and Facebook. Newsvine also developed “question boards” for MSNBC .com as a replacement for the Blue Site’s standard user polling feature. Question boards allowed users to vote on a survey question, as well as offer a reason for their vote and to participate in a discussion thread about the issue behind the poll. Each of these different development projects could be the subject of its own substantial discussion. Presently, though, I’d like to focus on one major project undertaken by Newsvine—MSNBC.com’s blog redesign—because blogs became a major component of the way MSNBC TV content was distributed online. First, however, I must introduce two additional pressures on the Newsvine system that emerged from the time of its acquisition. The first of these was the rise of general-interest social media. When Newsvine began, and at the time of its acquisition, many startups, venture capitalists, and tech industry commentators believed that the future of social networking would involve the facilitation of niche online communities. Rather than a few massive social network sites, the majority of people’s participation would take place on smaller social networks dedicated to every imaginable topic and specialized interest—a vision popularized in former Wired editor Chris Anderson’s book The Long Tail. Mark Budos recalled the manner in which The Long Tail became part of the zeitgeist in the tech industry in the mid-2000s: “It was going to be great because you’d have all these sites that would be talking about all these little niche subjects. . . . Newsvine still has a place, which to me is, you’re still able to leverage those niche subjects. You’re still providing a forum for people that want to participate in that [way]. It’s not this purely vanilla, generalized platform. . . . We’re really trying to focus on people that really like the news, and then they can create their little pockets and communities throughout our product to satisfy their needs.” MSNBC.com, meanwhile, was developing a business strategy that involved targeting different segments of the news market with different products. On the one hand, this meant targeting the same consumers with a selection of products that would meet their needs in different contexts

Out of One, Many  166

throughout the day. As Tillinghast described in a 2010 press release, “We know that people want news in different ways in varying circumstances” (MSNBC.com 2010). On the other hand, this “one size does not fit all” strategy meant targeting different audiences with dissimilar consumption habits by providing a diverse selection of news products under different brands. The acquisition of Newsvine was one of the first salvos in this strategy of diversification. Newsvine.com was originally intended by the MSNBC .com system builders to be the parent company’s social network offering, a community space aimed at a small but valuable segment of the news market. In the years after its acquisition, however, the rise of massive general interest social networks like Facebook and Twitter transformed the online social network space.7 It’s “the massive decentralization of conversation,” remarked Davidson. “The concept of one place where people are talking to each other right underneath the content . . . [has] become less popular as Facebook and Twitter have really taken over.” He also lamented a trend among friends and colleagues of removing discussion features from personal blogs, under the assumption that “if somebody wants to leave a comment, they can just tweet about it. Or they can reblog it on Tumblr.” “Decentralization” is an interesting way of looking at this situation. While others would point to the manner in which Facebook and Twitter have concentrated online discussion beneath their respective roofs, in Davidson’s view, the rise of general-interest social networks has meant the breaking up of topical conversations into different, egocentric social circles, with each person discussing the story with their own friends or followers rather than as a larger public. However defined, nearly everyone I talked to at Newsvine, MSNBC TV, NBC News, and MSNBC.com saw the rise of general-interest social networks broadly, and Facebook and Twitter specifically, as a transformative development, one that marked a dramatic shift in the conditions under which people congregate online (Tryon 2013; Acland 2003). From the vantage point of MSNBC.com, it also in many ways altered what Newsvine was good for. “When we bought [Newsvine],” said Tillinghast, “Facebook didn’t offer all these tools to create, basically, social networks within your own site. So from a tools point of view, Facebook is really supplanting some of the need we had with Newsvine.” But while the value of Newsvine may have changed for MSNBC.com, it did not go away, and in fact it increased in many ways. The conversation economy, especially distribution through social media, has rapidly become

Different Things to Different People  167

critical to online media seeking audience exposure. As Tillinghast put it, persuading fans and social media followers to share content online is “an area where we have to get really good. Because that’s the nature of distribution these days.” And while Newsvine.com, or the “Green Site” as it came to be known internally, remained a niche product compared to the likes of Facebook and Twitter, the staff of Newsvine was intimately familiar with the technologies and principles behind social networks. In this environment, from the vantage point of many at MSNBC.com, the Newsvine office had the potential to become especially valuable as a design shop and a consulting team focused on social media technologies and strategies extending beyond the Green Site itself. Newsvine also remained important to MSNBC.com due to the parent company’s increasing need to revise its blogging platform. MSNBC was an early mainstream media entrant to the blogosphere, launching its first blogs in 2003. Its early efforts were experimental, rather than largescale, initiatives, and as a result the various MSNBC.com blogs that survived into 2010 were hosted on multiple blogging platforms. These platforms included TypePad from SixApart, and another commercial product called Community Server, which was created by Telligent and adopted as MSNBC.com’s primary blogging software when the company first began its larger-scale blogging efforts. The Telligent platform was also the one used by all of the NBC News and MSNBC TV programs that kept blogs through 2010. The graphic design of all the blogs on both TypePad and Community Server was becoming dated by 2010, as was much of the backend functionality when compared to more contemporary—or at least more recently updated—blogging tools. Moreover, at least in the case of Community Server, the software itself put limits on what a new blogging template could look like, which in turn limited the ability of designers to give MSNBC .com blogs a fresh look. While an upgrade and redesign of these tools was “sorely needed,” as one staffer put it, it made little sense from a resources standpoint to go on indefinitely upgrading multiple blogging platforms. Newsvine, meanwhile, had built a mature blogging platform to power users’ columns across the Green Site. In short, looking from the vantage point of MSNBC.com, the joint venture needed to simultaneously upgrade and consolidate its blogging CMS and conveniently owned a company that built just this sort of software.

Out of One, Many  168

This combination of the Newsvine team’s experience with social media fundamentals and the fact that they presided over a mature social blogging platform made the subsidiary a uniquely valuable and influential property within MSNBC.com. As Tillinghast suggested in October of 2010, the management at MSNBC.com had determined that “Newsvine, given their skill set, is really the unit we need to use to invent our future.” I THOUGHT THIS WAS ABOUT TELEVISION

So far we’ve been predominantly looking at system building from the vantage point of Newsvine, occasionally switching to the vantage point of MSNBC.com for the purpose of gathering additional perspective. This approach has given us several valuable insights. At this point, we can see there is no single hierarchical structure that explains this universe of systems. MSNBC.com can be seen as an actor enrolled in Newsvine’s system, providing financial stability to the social network Newsvine is constructing. Or Newsvine can be seen as an actor enrolled in MSNBC.com’s system, a property the joint venture uses to leverage one more segment of the news market. Which actor falls into which role is strictly a matter of the perspective from which we analyze the situation. From the discussion so far, we can also mine another analytical tool: local teleology. Which is to say, each system builder is working toward its own provincial goals, and each has a situated notion of how heterogeneous resources need to be (re)aligned in order to unfold its particular vision for the future. It is where these goals come into conflict with those of overlapping system builders that recalcitrance is perceived. One heterogeneous engineer’s sticking point, or “reverse salient,” to quote Hughes (1987), is often part of another’s local teleology. We’ve seen how Newsvine developed an isometrically stable social news platform. Another way of putting this is that as Newsvine set and met its various objectives, encountering and overcoming reverse salients, it was building in accordance with a local teleology. The goals, values, and assumptions tied up in that local teleology and built into that system— Newsvine’s “aspirational philosophy of user behavior”—were what came under such acute pressure upon Newsvine’s acquisition by MSNBC.com and its legions of drive-by commenters. Which is not to say that Newsvine’s philosophy or the system behind it collapsed. Rather, as we’ve seen, Newsvine responded to these new challenges by heterogeneously engineering solutions to preserve its local teleology or at times accommodate

Different Things to Different People  169

revisions to it, and to ensure that the system it was building continued to unfold as its founders intended. This conceptual understanding of local teleology and the relative nature of systems is also important in understanding the reasoning of this and the previous chapter. Why, after all, spend so many pages talking about Newsvine’s startup phase and its acquisition in a manuscript that’s ostensibly about online television distribution? What does Newsvine, particularly in this early period, have to do with TV? As we will see, Newsvine and its blogging platform came to be intimately associated with NBC News and MSNBC television programs. By 2011, Newsvine blogs would make up the primary web presence for many of these shows, to the extent that when MSNBC .com and MSNBC TV began considering the creation of a standalone site dedicated to the cable news channel, Tillinghast remarked that “it may make sense just to build the whole site on Newsvine.” Much of this story remains to be told, but what we’ve come to understand in rehearsing Newsvine’s early system-building efforts and its relationship with MSNBC.com is the provincial and fragmented nature of large contemporary media companies and postmodern system building. That in a span of five years a site dedicated to citizen journalism, with no video player, could end up providing the online publishing infrastructure for a national television news organization—generally considered one of the most centralized and professionalized media institutions of modern times —and do so while retaining its original system, speaks volumes about the heterogeneity (or as organizational scholars would put it, the “heterarchic” nature) of contemporary media conglomerates and the power of the kludge. MSNBC.com in 2010 was a quintessentially postmodern organization in the sense touched on by Chia (1995), Hughes (1998), and Deuze (2007). We can speak of its needs, its managerial decisions, and the systems it builds—we can punctualize it, to use the ANT term (Law 1992)—but in doing so, we must ourselves be reflexive about the fact that we are ultimately employing a construction. In point of fact, MSNBC.com was an assemblage of system builders, including but not limited to subsidiaries like Newsvine, all with provincial agendas and local teleologies that worked in concert often enough to be mutually beneficial, but which were just as often recalcitrant when viewed from different perspectives.

9//if we’re not online, we don’t exist the rachel maddow show builds a web presence

By 2010, within the cable news industry, The Rachel Maddow Show (TRMS) had already grown into a great success story of television on the web, consistently ranking at the top of many popular online audience metrics. As we’ve seen, online cable news sites had become one of the leading sources of news for Americans, far outpacing even the leading online newspapers in terms of web traffic (Joyella 2011b). Among these cable news sites, the Rachel Maddow Show—according to Experian Hitwise, a ranking tool eyed by many in the media business—consistently scored among the top three sites attached to cable news programs (Cohen 2011a, 2011b). According to both comScore and MSNBC.com’s own figures, the show’s audience consumed more online video than any other at MSNBC TV in 2011 (Plesser 2011d), and Maddow herself had more Twitter followers—well over 1.8 million—than any other television journalist (“Social TV Leaderboard” 2011). How did this network of distribution resources come to be? ENGINEERING TRMS

The classic ethnographies of television news production, including Deciding What’s News by Gans (1980) and Putting Reality Together by Schlesinger (1978), were written before the advent of twenty-four-hour cable news. Each portrays the assembly of an evening newscast as a massive undertaking, and it is. One of the substantial differences, though, between a broadcast and a cable news program is the size of the staff responsible for putting it together. Compare the Rachel Maddow Show, which by 2011 was already the premier evening news program at MSNBC TV, to its flagship counter170

If We’re Not Online, We Don’t Exist  171

part at NBC, Nightly News. Nightly was assembled in NBC News’s central newsroom, which took up a substantial portion of the General Electric skyscraper’s third floor. Nearly the entire staff of TRMS, by contrast, were seated at two small rows of desks in an office suite that held half of MSNBC TV’s television staff. Again, for the sake of contrast, there are many network newscasts for which the anchor is largely a figurehead, if for no other reason than that it is impossible to be personally involved in every aspect of the expansive news gathering operation. Network news programs—especially news­magazine shows—are also at times less a platform for the anchor and more a showcase for the work of network correspondents. American cable news anchors, by contrast, are frequently part of small staffs that work and travel together in a tight-knit group. They are often intimately involved with nearly all aspects of their own programs. One key to understanding the situation is that much of cable news commentary—and perhaps especially material on the Rachel Maddow Show—is closer in its editorial form to talk radio than to a network news broadcast, and the agenda for it is set as much by what the host wants to talk about as by the news of the day.1 In fact, Maddow got her start in talk radio, and carried many of the elements of her political radio show over to TV. Numerous producers gave very similar descriptions of the Maddow program’s editorial process, describing their jobs as revolving around locating stories that would interest the host. A representative passage comes from Bill Wolff, one of the show’s executive producers who doubled as MSNBC TV’s vice president for primetime programming: “The challenge of producing [the show] is to pore through the news, and then do your best impersonation of Rachel Maddow in your head, which probably won’t be very good. And then try to present to Rachel what you think will be the most compelling stories of the day to her.” Maddow herself explained her close involvement with the editorial direction of the show, noting that unlike some other MSNBC newscasts, with generic titles like Hardball, the Last Word, or Countdown, the program was prominently branded with her identity: “I need to be comfortable with whatever is represented as part of, or derivative of this show, because people conflate me and all that is the Rachel Maddow Show. So, if I am conflated as an individual human being with all editorial product that comes from this show—both the TV show and everything that we do online—it’s essentially putting my signature on everything that we do. And so I’ve got to be okay with it.”

Out of One, Many  172

The reason it’s important to understand the depth of involvement Maddow had—and has—in the program that bears her name is that it also helps us to grasp the extent of the influence she wielded at MSNBC. According to some sources, MSNBC TV executives were initially reluctant to give Maddow her own program (Pressler 2008), but TRMS’s debut in September of 2008 was the most successful show launch in the channel’s history, with the show garnering numerous accolades and awards during its first year (MSNBC.com 2011). By the end of its first month on the air, Maddow had doubled the ratings for her time slot at the cable channel (Hood 2008), and by 2011 MSNBC TV president Phil Griffin was touting TRMS as the channel’s “biggest show” (Kurtz 2011). The point of mentioning all this is not to aggrandize Maddow or the program, but again to illustrate her influence. Maddow was synonymous with her program, both to audiences and staff, and her show quickly became a highly competitive franchise at a channel that until that point had generally wanted for ratings. This early and sustained success gave her a great deal of clout within both the cable channel itself, where she had quickly become a hot commodity, as well as at MSNBC.com, which was also eager to cash in on her success. And one of the things that Maddow wanted most was a robust web presence for the show. Amid all the profiles and celebrity buzz that emerged in the months following the launch of the show, one of the more popular factoids to emerge was that prior to making the move from talk radio to cable news, Rachel Maddow did not own a television set (Battaglio 2008; Pressler 2008). Rather, she consistently described her information habits as revolving around the Internet (Battaglio 2008; Pressler 2008), telling interviewers “anything I want I can find online” (Couric 2011) and describing her avoidance of other television as “purposeful naïveté” (Pressler 2008) aimed at preventing the show from becoming derivative of other cable news. It would be fair to argue that her foregrounding of these facts to the press was about painting an interesting public persona. But rather than dismissing this persona, we should recognize, first, that it might be an accurate portrayal of her information habits, and, second, that the Internet-savvy public persona she put forward helped to create a web-friendly brand for herself and her show. As the head of MSNBC.com’s TV-web producer team, Sam Go, put it, Maddow is “web-savvy. So she also requires that of everyone around her.” Staffers at TRMS were immersed in print journalism—albeit online print

If We’re Not Online, We Don’t Exist  173

journalism—and the political blogosphere. Television journalists generally start their day by “reading in”—catching up on the day’s news and looking for story ideas by reading the papers, an aspect of their work that long predates the Internet (Hetherington 1985; Schudson 2011). For many politically oriented news programs, and especially for TRMS, this process had not only moved online, but also expanded to encompass a greater diversity of sources. Vanessa Silverton-Peel, for instance, a TRMS producer who had come with Maddow to MSNBC TV from a previous job in radio, described her daily reading-in as consisting of “six newspapers and probably thirty different blogs” ranging in their subject matter from politics to war to foreign policy. When I talked with Maddow, she again foregrounded her personal media habits and how these played out in her own—and consequently, the show’s—desire for a strong web presence. I think the reason that we have prioritized the web, . . . the reason that I am interested in the web is because the web is my only source of media. I don’t watch TV. And I don’t read anything that’s not online. Like at all. I really don’t have any experience of any media in terms of my understanding of what’s going on in the world other than what happens online. And so for me, us having a substantive presence online is almost an existential question. If it’s not online it doesn’t exist in my world. So if we’re not online, we don’t exist. Again, it’s worth recognizing that Maddow’s characterization of herself and her affinity for the web, here and elsewhere, while not false, are also part of an important mythology that has helped to propel TRMS’s online efforts. From the beginning of her time at MSNBC TV, she publicly foregrounded the web. Just what should constitute a robust online presence for the show, however, was a subject highly dependent on vantage point. As we will see, the development of TRMS’s Internet footprint is both a tremendous example of heterogeneous engineering and an important demonstration of local teleologies at work. “EVERYTHING HAS TO WAIT”

Maddow and her staff wanted to launch a blog as their program’s homepage, an idea that turned out to be unexpectedly difficult. After all, many shows at NBC News and MSNBC TV, from the Nightly News to Hardball, kept blogs on MSNBC.com. As far back as 2007, MSNBC.com had claimed

Out of One, Many  174

publicly that the Community Server blogging platform gave them the ability to spin up new blogs quickly and easily (Telligent 2007). To understand why a blog was not immediately in the offing—to see what was different this time around—we have to step momentarily into the vantage points of MSNBC.com’s Redmond development offices and Newsvine. You’ll recall from previous chapters that MSNBC.com had decided to move from its Community Server blogging platform to Newsvine-­ powered blogs. This in itself was not a reason for delaying the Maddow program’s blog. The issue was that the Blue Site was in the middle of a redesign not just of its look, but also of its underlying architecture—and this project ultimately took a long two years from start to finish. From an architectural standpoint, the new blogging platform should have been immune to these delays. It was built on Newsvine’s codebase and hosted independently using Newsvine’s Linux servers, rather than the Blue Site’s Windows system. But the Newsvine blogs would exist under the MSNBC .com domain, rather than as part of the Green Site, and the designers at MSNBC.com were concerned that the look of the new blogs be consistent with its new visual identity. That meant waiting to launch the new blogging platform until the redesign of MSNBC.com was nearly complete. According to Newsvine CEO Mike Davidson, this process should ideally have taken two to three months, but “because design ended up taking a lot longer on [MSNBC.com’s] side it dragged on for probably over a year. . . . As far as we [at Newsvine] were concerned we could have launched blogs six to nine months earlier.” Nor were Newsvine blogs the only development project caught in limbo during the redesign. Will Femia, the MSNBC.com web producer assigned to the Rachel Maddow Show, recalled that numerous projects were in the mix, including the site’s lucrative video player: “This happens every once in awhile, that we [at MSNBC.com] have this idea of what we’re going to do, so everything has to wait because this new thing is coming. But then the [launch date for the] new thing gets bumped. We’ve already been waiting six months, and now we have to wait another six months. . . . So we’ll have been waiting a year and not having advanced at all.” Because the redesign of MSNBC.com’s story pages and their underlying architecture was the number one priority, everything else had to wait. And as Femia pointed out, “everything” included more than just the launch of the Newsvine blogging platform. “Blogware wasn’t even second. Blogware was down [in the queue]. Video player actually makes money. Video player

If We’re Not Online, We Don’t Exist  175

was in there waiting.” In short, the Maddow program’s request for a blog was caught in the middle of a long production cycle. Community Server blogs were being phased out, and Newsvine blogs were not yet ready, since they were tied to the release of MSNBC.com’s larger redesign. Making matters more complicated was the fact that Maddow was displeased with the standard “micro-site” template for MSNBC TV shows. This template, which was dominated by a video player with a grid of links beneath it, made a great deal of sense from the vantage point of MSNBC .com. It prominently promoted video, which was a particularly valuable commodity for the web venture. Moreover, many shows and anchors contributed relatively little content to the web beyond the repurposed show clips that went into the player, and there were quite possibly television personalities who thought of contributing to the web as an unwelcome obligation or a distraction from putting out a newscast, as opposed to an opportunity for engagement or exposure.2 A number of shows in the past had started blogs, only to all but stop maintaining them. Cathy Finkler, an MSNBC.com web producer who worked with numerous cable shows, spoke enthusiastically about the online efforts of MSNBC TV shows throughout my interview with her. At the same time, she said, even after blogging software was again readily available, We don’t just snap our fingers and then this show has a blog. No, the show has to realize that when they have a blog it’s not just going to be my blog. They have to contribute. They have to come up with ideas. And as much as we want to say, “Refer to Maddow’s blog. Look at how great this turned out,” they also can’t just start copying Rachel’s blog from start to finish. So from there we have to try to figure out which shows really want a blog and which are going to say they want one and then a week later the blog’s going to be empty. That could be a problem. Making video the dominant feature of the micro-site templates used by MSNBC and NBC News programs was one way in which MSNBC.com was able to highlight the content that shows were guaranteed to contribute—that is, video—while downplaying potentially dormant blogs or other types of content that might not be as forthcoming. For that reason there had been shows that the template served very well. Moreover, one of the first MSNBC TV shows whose content was distributed online through this template was Keith Olbermann’s Countdown, the

Out of One, Many  176

micro-site for which was launched in June 2008 (Learmonth 2009; Plesser 2008b). Olbermann was active online, but his reputation for unapologetic partisan commentary made his web contributions the object of voluminous and frequently incendiary user comments, which spilled over onto forums and blogs across MSNBC.com. At the time the original template was designed, the dominance of the video player over blog posts and other forms of content with relatively greater affordances for user commentary may have been a design strategy aimed at minimizing the amount of belligerent feedback the site received. Finally, as discussed in Chapter 4, the show micro-sites themselves were one means by which MSNBC.com was able help surface MSNBC TV’s “op-ed television” content while keeping the bulk of it distinct from the content of the Blue Site “cover” (front page) and the latter’s nonpartisan editorial image. In other words, from the vantage point and local teleology of MSNBC.com there were numerous heterogeneous factors having to do with finances, branding, editorial philosophy, and human resources that played into the design and implementation of the show templates. From the vantage point of Maddow and her newly assembled show staff, by contrast, the original show template and the numerous delays surrounding the blog instead appeared to be “reverse salients,” elements holding back the progress of the rest. Maddow recounted, in terms very consonant with Law’s notion of heterogeneous engineering, how the program went about creating its web presence (Hughes 1987). Her account is worth quoting at length because it conveys a very different view of the system built by MSNBC.com than that described thus far: We realized pretty early on that MSNBC.com was not our website. And therefore if we wanted to have anything useful online, that we were going to have to do it ourselves. And so we essentially set about corralling the resources that we could that were already being made available to us and maximizing any other resources we could apply to the problem so that we would have a web presence that made sense. What MSNBC.com built for us was essentially a generic single-frame website with a video player in it. And [facetiously:] a place for links! It was like 1998. It was 1998 with hinky flash video. It was really, really totally useless. And so we wanted to have something that actually had some content on it, that wasn’t just a competitor with our TV show for ratings because it was a video provider. For us to actually have written word-based content, which is what I think all the most useful social

If We’re Not Online, We Don’t Exist  177

media is, we built it ourselves. And so we went through a monthslong fight with the powers that be here in order to essentially get the freedom to do our own thing and have it go up alongside our official website, which now still has the better URL and is totally functionless. So we have a [blog located at a] worse URL that we built on our own that’s all our own functionality, all our own design, and has none of their stuff on it, run by all our own people. And we exist alongside the corporate structure that they won’t let us shirk. Maddow’s description of the process by which the new website for her program came to be is a profound lesson in local teleology. That one show staff could define its online distribution requirements and ambitions in a manner so differently from not just a partner organization within the same media conglomerate, but also other shows working out of the same office suite, helps to drive home the heterarchic nature of today’s large media organizations, which are far from homogenous or monolithic in their culture or in their objectives. This is not a criticism of MSNBC.com or MSNBC TV, nor any of their parents or subsidiaries. As organizational researchers studying new media firms (such as Girard and Stark 2002; Neff and Stark 2004; or Kellogg et al. 2006) have found, it’s equally simple to find examples of divergent local teleologies not only within other television networks, but also—and perhaps especially—among born-digital media companies.3 Moreover, as discussed previously and suggested by researchers like ­Girard and Stark (2004; Stark 1999, 2009), the coexistence of different and even competing organizational logics within a larger enterprise may well be highly adaptive over the long term in tumultuous and unpredictable industries like today’s media sector. For its own part, Maddow’s critique may be somewhat overstated— there were in fact designers and engineers in Seattle and Redmond who were intimately involved in helping to design and build the new show site to which she refers. But as we’ll see, her emphasis on the amount of work involved, and the number of different resources that had to be enrolled to create what became one of the most successful cable news sites ever, isn’t disingenuous. Thomas Hughes (1987) points out that even ostensibly helpful or useful components of a system can become bottlenecks that limit its overall efficiency if they are out of sync with the larger goals and advancements made by the system builder. Outdated hardware or software, obsolete policies, or dated customs may once have been useful to a system’s architects—and gen-

Out of One, Many  178

erally still have useful aspects—but become regarded as uncooperative reverse salients until such time as they can be updated, engineered around, or brought into phase with the rest of the system. And what’s “outdated,” “obsolete,” or difficult in other ways is very much in the eye of the beholder; such terms often get attached to alternative visions of progress or efficiency that have been marginalized in the eyes of the current system builder. Where there are multiple system builders and organizational logics in play simultaneously, however, what is central and progressive versus what is marginal and obsolete is often largely a matter of where the observer stands. Our discussion to this point has illustrated that reverse salients are seldom merely obstacles, but rather recalcitrant aspects—when seen from a particular vantage point—of actors that are otherwise important resources. The Rachel Maddow Show’s website would likely not have achieved its popularity without MSNBC.com’s seamless video delivery system and its powerful data centers, but the site’s success also depended as much or more on the show staff ’s well-developed capacity for creating and showcasing compelling content across different media. Rather than trying to reach a specious conclusion about which system builder had the “right” idea when it came to assembling the web presence for the Maddow program, we would do better to recognize the presence of different local teleologies simultaneously at work, trace out the various overlapping systems in play from different vantage points, and learn something about how they interacted and influenced one another. HIRING A FULL-TIME WEB PRODUCER

As we saw in previous chapters, from MSNBC.com’s perspective, MSNBC TV and NBC News were only two—albeit an important two—of a growing range of content providers in the “MSNBC Digital Network” in 2010. In recent years, television content on the Blue Site had been handled by a small team of MSNBC.com web producers based out of the web venture’s New York offices in Rockefeller Center. Their job was primarily to help repurpose televised video for the web and to maintain the various microsites dedicated to MSNBC TV and NBC News shows. They were also frequently responsible for helping to maintain—in whole or in part—the content of various blogs, mobile applications, and social media accounts started by or on behalf of the TV shows and their staffs. While the micro-site templates were generally very similar to one another, shows tended to differ greatly in their embrace and use of these other

If We’re Not Online, We Don’t Exist  179

platforms. Some shows had numerous social media accounts, while others had few. Some were filled with personal interaction while others were primarily used to post headlines. And as discussed in Chapter 5, mobile applications, where they existed, were relatively diverse in their design and content. Sam Go, who led the New York team of MSNBC.com TV-web producers, described how the workflow for the different shows was similar overall, though various programs could be more or less demanding: “The shows are different; some of them are more involved than others. But at the end of the day they very much want the same things, which is the content to be on the site, anything related to be easily findable on the web, and social distribution.” In other words, while various programs had unique requirements, the differences between the tasks required of Go’s team for one program versus another should not be overstated. Owing to limited resources, most of them were even done by the same people. In 2010, one person, for instance, handled the web content for all of the shows that aired weekdays on MSNBC TV between 11 a.m. and 7 p.m., while others took similarly large chunks of the remaining TV roster. To keep up with the demands of such an expansive schedule, the different web producers often pitched in to help with one another’s tasks. The exception to this block-scheduling of web producers, as you may have guessed, was the Rachel Maddow Show, which was able to procure a single, dedicated web producer from MSNBC.com in the form of Will Femia. Femia had been working with MSNBC since the late 1990s, managing websites, chat rooms, blogs, and online communities for television and the web since long before the term “web producer” had become common parlance among journalists. Not only was the Maddow program acquiring a full-time producer from a web team with a thinly spread staff, it was also getting one of MSNBC.com’s most experienced television professionals in a major reapportioning of resources. From the vantage point of the show, Femia appeared to be an under­ utilized resource, someone overlooked by MSNBC.com. As Bill Wolff, MSNBC TV’s vice president of primetime programming and an executive producer of the Maddow program, explained: Will’s been around for a long time. . . . And he’s really good at [his job]. That I already knew. . . . And we thought . . . Will’s supposed to be an MSNBC.com guy doing stuff for MSNBC TV shows. Only nobody

Out of One, Many  180

really capitalizes on him. . . . Everybody [at MSNBC TV is] working really hard to get their jobs done, and doesn’t have time to think about the Internet or doesn’t understand it. . . . And then he gets ignored and nothing comes of [his ideas]. And we just said, “Well, why don’t you come to the show meeting every single day? Why don’t you just be part of the show, the show’s operation?” Meanwhile, from the vantage point of MSNBC.com and its East Coast television staff, Femia was far from underutilized. Femia recounted how the decision to dedicate his time to the Rachel Maddow Show put a strain on the resources of MSNBC.com: [The Rachel Maddow Show] was still new. And this is actually kind of funny. I hadn’t met Rachel. She went out to Redmond. And . . . I met her by teleconference when she was in Redmond, rather than twenty feet away from me in the building. I was there basically to chime in about our tools, just to be a nerd in the room. But someone had made the mistake of saying, “Whatever you want, let us know.” And she said something like, “Who’s your best guy?” And they said, “Will.” She says, “Okay, I want Will.” And . . . they’d already said, “Whatever you want.” [laughs] . . . And it’s not like we had the resource to spare. It has been a good experiment, and a good way of showing what it’s like when a show has a dedicated web person. But for the amount of work that Sam [Go]’s team has to do, to say one guy is devoted to a show is a poor use of resources from that standpoint. . . . If you said, “Oh, Sam, you have . . . all the shows except Today Show.” So, Meet the Press, Nightly News, Way Too Early, Morning Joe, Jansing & Co., NewsNation, Andrea Mitchell Reports, Dylan Ratigan, Hardball, Ed, Countdown, Maddow, Lawrence O’Donnell. You have two staff members [for all those shows] and people who do Nightly, Meet the Press, and Dateline exclusively. How do you want to use those two staff members? Put one exclusively on Maddow Show? That’s crazy. The point is, having a dedicated guy for one show is weird. Importantly—and this may not be clear without additional context— Femia was not critiquing the apportioning of resources or his placement with the Maddow program, but rather illustrating the magnitude of the decision. In the terms we’ve been using, the Maddow program’s recruitment of Femia—during the month in which she doubled MSNBC’s Nielsen

If We’re Not Online, We Don’t Exist  181

ratings—put rending pressure on what, from the vantage point of the MSNBC.com’s East Coast television team, had been an isometrically stable system.4 In light of this, it’s interesting to consider from the vantage point of MSNBC.com why they made the concession and what they gained. Maddow, through her radio show, her guest appearances on MSNBC TV, and her other work in media, online and off, had developed a substantial online following of people who were generally thrilled by MSNBC’s decision to give her a show. When announcing that Maddow would be receiving her own program, Keith Olbermann recounted on his show the reactions of several major progressive blogs and websites, in which Maddow was referred to variously as “a gift” and “a star,” “radiating competence and a deft, bright careerism,” and in which her hire by MSNBC was dubbed—with more than a bit of hyperbole—the “single smartest move ever made by a television network” (Olbermann 2008). This sort of online reception was obviously welcomed by MSNBC .com, and the company was thus willing to shuffle a few resources to capitalize on it. Randy Stearns, MSNBC.com’s East Coast deputy editor, recalled, When [Rachel Maddow] joined the MSNBC lineup, we courted her pretty heavily because she’s a digital native; she and her EP [Bill Wolff] get it. To her credit, she and her executive producer came out to Redmond and met the news team, talked about how we could work together. We designed a site for her. We dedicated Will [Femia] to work with her team—the first time I’d ever dedicated a producer to a single cable show. (We’ve never had one person dedicated to a particular show, because we couldn’t afford it.) And they’ve done great stuff—on TV, on the web and, more recently, for devices like the iPad. Moreover, both Stearns and Go emphasized one aspect of Femia’s earlier comment, which was that assigning Femia to the Maddow program was a demonstration of what was possible for the cable shows online, if only more resources were dedicated across the board. As Sam Go said, It’s a strategic decision on our part to have him spend more time on that show. You need that model, a prototype for how it could be done. And that’s spoken well of our company, when we [the East Coast offices] deal with MSNBC.com. It’s like, “We didn’t treat them all equally.

Out of One, Many  182

We are trying to experiment aggressively with one show that does more of this kind of content than the others.”

She continued: I think it would be a less productive use of time to devote those resources to Hardball, because they’re not as committed to producing for the web, or they’re already understaffed for their main priority: television. So you put your resources where they might make a difference, considering you have a few. Ideally, would I want one person for every show? Yes, of course.

Here we can see the duality—what Bruno Latour (1987) calls the “Janus face”—of actors within systems, how they are seldom strictly a resource or an obstacle, even from the vantage point of a single system builder. Femia’s dedicated assignment to the Maddow team might have been seen as a strain on the systems put in place by MSNBC.com or by Go’s team, but it was reframed as a resource in each case—the capture of Maddow’s online brand and tech savvy audience for MSNBC.com, or the opportunity to demonstrate the potential benefits that would come with the addition of resources to Go’s team. To the extent that Femia’s new assignment became an imposition on other parts of these systems, it was regarded (in our terms) as recalcitrance—a difficulty to be engineered around through reapportioning of the remaining staff ’s duties and other means, for the sake of the larger benefits that Maddow brought to these systems. USING OUTSIDE SERVICES

Even after the Rachel Maddow Show had a dedicated web producer, it still didn’t have a blog and its micro-site didn’t offer many opportunities for interactivity. To get around these problems, Femia, at the behest of the show, helped in kludging together a number of workaround solutions. Some of these involved MSNBC.com resources—Maddow, for instance, participated in Q&As on Newsvine, where users were invited to discuss with her the arguments she made on a regular segment of the show titled “Talk Me Down.” Moreover, in addition to the video-player-centric micro-site template, cable shows were given the option of maintaining an informational page on the tv.msnbc.com micro-site, dedicated to the cable channel. While this page was originally designed to contain relatively static content, Femia

If We’re Not Online, We Don’t Exist  183

updated it continually, appropriating its slightly more flexible template as a place to aggregate information on Rachel Maddow and the show from across the web, including that contributed by users and fans. On this page, he cobbled together various features that would ultimately be assumed by the blog, including a section for primary documents related to stories on the show, links to podcasts, a listing of Twitter accounts related to the program, links to Maddow’s guest appearances on other programs, as well as links to press clippings and profiles featuring her. Elsewhere, the Rachel Maddow Show and Femia went outside the normal resources of MSNBC.com, starting an account for the show on Flickr to create embeddable slideshows and accept user-contributed photos, for example, and another on a site called Dipity, which generated interactive timelines. When the producers discovered that fans of the show were frequently posting quotes from the program to their blogs and Twitter accounts, Femia created a page on Wikiquote where audiences could assemble all their favorites in one place. These sorts of specialized third-party web accounts multiplied over time, and for a variety of reasons. One reason was that, while the blog was still in limbo, there was little infrastructure on MSNBC.com to host a centralized user community around the Rachel Maddow Show—at least not without committing to a solution, like the moribund Community Server platform, that would end up being dismantled in a matter of months. It made more sense to seek out communities of users around the web that were already engaging productively with the show, then to highlight, and where possible facilitate, their contributions. This strategy not only outsourced the community functionality of the as-yet-nonexistent blog; it also promoted the circulation of the program’s content in different conversational spaces across the web, helping to extend the conversation economy surrounding the show. In this spirit, Femia added to the ostensibly static informational page for the program prominent links to Maddow fan sites, samples of user-submitted art and commentary, as well as links to Wikiquote and other interactive Maddow Show account pages around the web. The top of the page prominently featured a slideshow of the latest TRMS-related photos from Flickr, submitted by both users and show staff. The page created for the program by Femia was thus a pointer to distributed activity around the web, rather than a host to it. As we’ll see, these impulses are not mutually exclusive—once the Maddow Blog ultimately launched, it continued to embed and point to many off-site resources. Until that time, however, the lack of community affor-

Out of One, Many  184

dances on MSNBC.com meant that only one of these options was available to the show. A second reason that the Rachel Maddow Show started numerous discrete social media accounts had to do with a strategy by Femia for engineering the quality of user contributions. Accounts on specialized services—for instance, a quotes wiki or a timeline tool—served as framing devices for user submissions. More than one web producer I talked with at MSNBC .com noted that belligerent behavior by users could be minimized by giving them a suggested course of action wherever feedback was requested or allowed. Ian Sager, for instance, who at the time was the MSNBC.com web producer for weekend on NBC News programming, noted that he and other web producers had begun to tack a question addressed to users onto the links they were posting to Facebook. As an example, he mentioned a story the network aired about the experiences of a man who had lived on coupons for a year: “We posted the story yesterday on Facebook and along with [it] the question, ‘Could you actually live on coupons for a year with no cash, no credit?’ Just so the first response was not, ‘What a dumb story.’ And of course there are people who are going to say that, too. But it also cuts down on that. A lot of people write back and say, ‘God, no I could never do that.’” In other words, these rhetorical attempts to suggest in advance the sorts of contributions that were welcome helped to reduce the amount of moderation and surveillance that were necessary across a show’s web presence. Femia regarded the Maddow show’s niche social media accounts as another way of going about this, commenting, I really feel like the new community model is “constructive engagement,” not “open engagement.” And by that I mean give your people stuff to do, and you get more relevant return. Open up a comment thread and you get bullshit. And what’s more is, you have to waste time dealing with the bullshit. It can’t just be, “Well, there it goes.” Suddenly it’s your bullshit that you have to deal with. That’s why we had Flickr for images and the Wikiquote page for quotes. And [Maddow’s] celebrity appearances were a whole separate page I was maintaining. Her fans would spot stuff she would do and submit it. . . . There were just more facets [to maintaining Maddow’s web presence]. I didn’t have a hub at all, which is what the blog is now. Instead I had all these crazy . . . third-party things. [Emphasis in the original.]

If We’re Not Online, We Don’t Exist  185

Finally, the third-party accounts at times simply gave the show’s web presence functionality that it did not have through the MSNBC.com website, or that would constitute more work for less benefit to obtain through MSNBC.com. For instance, while MSNBC.com had no easily embeddable slideshow widget at the time, the Flickr account not only provided the show with such a tool; it also came with social features and a simple-to-use interface that was easily accessible, not just for MSNBC.com web producers, but also to audiences and to the show’s television staff who had seldom worked with the website’s publishing tools before. As we’ll see, this trend toward using outside services to supplement the functionality of MSNBC .com and extend the reach of the program’s content continued long after the Rachel Maddow Show received a blog. MAINTAINING FAKE BLOGWARE

The Maddow page updated by Femia on tv.msnbc.com served as a library of original TRMS web content, as did the link area on the Maddow program’s micro-site, where Femia would post the show’s guest list, along with links to other show content on MSNBC.com, Twitter, and elsewhere. He also regularly posted links to blogs, newspaper stories, and other sources mentioned on the show, and eventually consolidated these into what was billed on the site as a “link blog.” Notably, the link blog was actually built not on any blogging software, but rather on a standard MSNBC.com storypage template that Femia would manually update, and on which he simulated as much of the functionality of a blog as possible. Nearly all the functionality of the various pages that Femia was creating and/or maintaining on the Blue Site was eventually assumed by the blog when it was built. In the meantime, however, it fell to Femia to bridge the gap between TRMS’s expectations of a blog and MSNBC.com’s slow delivery of the new blogging tool by manually accomplishing much of what the blog software would eventually do. He referred to these tasks on the link blog, and across TRMS’s presence on the Blue Site, as “faking blogware”: “The most awkward thing that [we] on Sam [Go]’s [TV-assigned web] team have to do as liaisons . . . is explain the development schedule of MSNBC.com to the TV side. The TV side can be impatient anyway, but even when they’re completely reasonable sometimes our schedule just isn’t. Sometimes our priorities are different. . . . But we [the Maddow program] wanted a blog. And as it is, here’s me faking it. That’s what it was.

Out of One, Many  186

I’m writing this fake blog with fake timestamps, fake archives, anchor-tag permalinks, everything.” The ease with which Femia slips between possessives and pronouns like “our,” “their,” and “we,” which place his group affiliation variously with MSNBC.com and the Rachel Maddow Show, is not coincidental. All the MSNBC.com staff in New York who dealt with television had to run interference at times, when the local teleologies of their dot-com employer and the television staffs with whom they worked were out of alignment. When Femia was enlisted full time by Maddow, he became one of the individuals for whom this situation was most pronounced. As MSNBC.com’s East Coast Deputy Editor Randy Stearns put it, “Will [Femia]’s become part of [the Rachel Maddow Show] team, even though he still reports to [MSNBC .com].” Television-affiliated web producers at MSNBC.com dealt with the tensions inherent in their jobs in different ways. Femia described his situation with the Maddow program as a provider-client relationship: “[I] think of it as, ‘They’re my client. Everything that my client wants, they’re going to get. Either you’re going to give it to me, MSNBC.com, or I’m going to go get it for them.’ I don’t see how there could be any other way of thinking about your job.” At first glance, this self-framing by Femia of his work role may appear to put the onus for providing resources largely on MSNBC.com, but more accurately, it paints the MSNBC Digital Network as only one of a larger universe of available service providers and provides one rationale for why the Maddow program so aggressively sought out third-party tools as part of its web presence over the years. When it came to the blog, however, the situation was not so clear-cut. MSNBC.com was, after all, promising a blog “soon,” and hosting it elsewhere would in effect have been outsourcing what the show staff considered their homepage—which is not to say the group didn’t consider it, as we’ll see in a moment. Faking blogware was not something Femia had been directed to do by any particular party. While it was clearly a task he took on at the behest of Maddow and her producers, it was also a compromise he helped to create in order to knit together two local teleologies that were, if not competing, out of sync in their unfolding. Ultimately, though, the standard templates that Femia used to simulate blogware proved to be another recalcitrant resource, not just in the amount of labor they required to maintain, but also in their presentation of information. The story-page template, in particular, had been designed for delivering breaking news, not blog posts, and

If We’re Not Online, We Don’t Exist  187

it showed. Whereas a blog ordinarily lists all the author’s recent posts in reverse chronological order, the publishing tools designed by MSNBC.com mimicked the format of a newspaper, continually presenting the latest headlines, while older items disappeared into the site’s archives, as opposed to simply being pushed down the page. Femia underscored how the traditional blog format was different from the MSNBC.com platform’s logic of “These are the headlines now. Boom. These are the headlines now,” illustrating how blogging via the standard news platform at times “made things more disjointed” from a user’s perspective. Circumventing the limitations of MSNBC.com’s story template would ultimately require starting an actual blog.

10//the maddow blog is born

Between 2008 when her show debuted and early 2010, Rachel Maddow quickly became a celebrity and an influential figure inside and outside of MSNBC, with profiles appearing in publications ranging from the New York Times to People Magazine. And as one online staffer put it, “As far as she was concerned, she didn’t have any allegiance to dot-com. Her allegiance is to making sure her shit looks how she wants it. And that she can do online what she wants to do.” Moreover, at that moment in time, the gulf between the affordances offered by MSNBC.com and what Maddow wanted out of her web presence was considerable. In addition to her reportedly disliking the appearance of the blogs generated by the obsolete Community Server software, Microsoft’s branding and visual identity guidelines, mandated for all MSN-affiliated sites, put limits on what even a new Newsvine-powered blog design could look like. Such limitations were especially unwelcome given the ease with which it was possible to spin up a custom theme with the desired look for a blog powered by a third-party tool like WordPress or Tumblr. By all accounts, Maddow and her staff were actively—and vocally— eyeing such third-party workarounds. Moreover, because MSNBC.com’s contractual rights to MSNBC TV content were largely limited to video, and because Maddow and most of her team were employed by NBC Universal, not MSNBC.com or Microsoft, the online company could do very little to prevent the show from launching its blogging efforts elsewhere. As Femia recalled, “What [were they] going to do? Complain to Phil [Griffin, president of MSNBC TV]? Phil’s going to take his new star and scold her 188

The Maddow Blog Is Born  189

for doing her own [blog]? There’s just nothing you can do. . . . That was the situation. That’s where we [at the Rachel Maddow Show] got to really push and say, ‘Do this or we go.’ It was an interesting spot to be in.”1 By now, we can see much of the Maddow program’s heterogeneous engineering on display. The show leveraged Maddow’s online celebrity to obtain a (nearly) full-time experienced web producer from MSNBC.com, something unprecedented for an MSNBC cable show. In the absence of a permanent blogging tool, it was able to make use of Femia’s skills and familiarity with both MSNBC.com’s CMS and a variety of third-party tools to mimic the functionality of a blog on the Blue Site until a dedicated blogging platform became available. TRMS also made use of its staff ’s lack of contractual obligations, in combination with Maddow’s influence and celebrity, to get MSNBC.com to press a permanent blogging tool into service more quickly. The show made another move unprecedented within the cable channel when it hired an additional web producer onto the television organization’s payroll, marking the first time that MSNBC TV had hired its own internal web staff independent of MSNBC.com. The Rachel Maddow Show had begun life in 2008 with a skeleton staff and continued to be run by a small team, even by primetime cable standards, throughout most of its first year. Following the show’s surge in the ratings and its critical success, however, the program was offered the opportunity to expand its producer staff and began to hire new people. In late 2009, when one such opening came up for a new television producer, as the New York Observer put it, “Maddow and her team decided they’d rather spend the money on someone who would focus primarily on the web” (Gillette 2010). In the following interview passage, Bill Wolff explained the motivation behind the move: We needed somebody who was going to work for our show and not have a conflict of interest with working for MSNBC.com. . . . [MSNBC .com staff are our] partners, but they have different concerns, they’re driven by different profit motives, editorial motives. It’s a different business from our show. And they’re not to be blamed for that. But we needed somebody—it sounds a little bit harsh, but—who answered to us, and whose motives were our motives. . . . We had a couple of open positions, and . . . I [decided] we were going to spend a television producer [position] on a web person. Because that was the only way our

Out of One, Many  190

show was going to have a sophisticated Internet presence. No one from MSNBC.com was going to have the time or the ability to be dedicated to making our web presence what it needed to be. That person needed to work with us every day, be part of the show. Because what was happening on the web needed to be an extension of the show, hand-inglove. It just needed to be the same thing, except a different format, a different platform. It’s worth taking a moment to contextualize this passage a bit further. First of all, I’ve already emphasized how Will Femia came to be a full-time web producer for the Rachel Maddow Show, which begs the question of why Wolff would suggest that “no one from MSNBC.com was going to have the time or the ability” to build out the show’s web presence. While every member of the TRMS staff with whom I spoke had nothing but plaudits for Femia’s work, referring to him as a “critical” and “brilliant” component of the show’s online presence, there were several limitations of his position that from their perspective warranted the hire of an additional full-time web staffer to the show’s own payroll. First, while Femia was dedicated to the show full time, many of his duties revolved around the repurposing of television content for the web—putting video online, for example. In other words, much of his schedule had been filled before the blog was even launched. Second, while according to his employers Femia’s schedule was ostensibly dedicated only to the Maddow program, in practice this wasn’t perfectly true. Owing to the limited resources of the MSNBC.com web producer team responsible for handling television content, Femia also retained responsibility for the Countdown with Keith Olbermann micro-site. While the micro-site had been made as static as possible to minimize the amount of incendiary feedback it invited, and was therefore less demanding than the web presence for some other shows, it still required regular updating, which consumed a good portion of Femia’s day. And as with all the other producers on Sam Go’s team, Femia would pitch in on the duties for other shows when necessary. Owing to all these additional responsibilities, Femia was not physically embedded with the Maddow program. While he attended the show’s regular editorial meetings, he did most of his work in the MSNBC.com satellite offices, which were separated from the MSNBC TV offices by a hallway and a locking door. The new TRMS web producer, Laura Conaway, by contrast,

The Maddow Blog Is Born  191

was to be involved in assembling television segments and pitching in, along with all the other television producers, to ready the Maddow program for air each night. In running the blog, then, she could speak for the show in a way that Femia could not. In fact, while Femia regularly wrote for the blog, he noted to me on several occasions that he generally limited himself to topics like science, photography, or humor where there was less risk of inadvertently contradicting Maddow or the television staff by stating, say, political opinions that later turned out to cut against those appearing on the show. Most of the various TRMS television producers, whose job frequently involved trying to anticipate what Maddow would be interested in each day, discussed the difficulty of predicting what she would have to say on a given issue. In deciding what to contribute to the blog, Femia faced the same challenge without the benefit of working alongside the host. “Rachel is so independent that even the people who do sit over there often aren’t quite sure what she’s about to do,” he remarked. “So it’s hard to stay connected that way.” This was what was meant by Wolff when he talked about the desire for the blog to fit with the show “hand-in-glove.” One early idea for the blog was to regularly feature posts from television guests and other invited commentators. This concept, however, was quickly jettisoned in favor of a blog that would reflect the interests, discoveries, and opinions of the people who actually produced the show. Shortly after it launched, Maddow told the New York Observer that, rather than serving as an additional obligation for an already-busy television staff, the blog was conceived as a way to capture the insights and conversation that went into producing each program, and “ended up being an engine for the show. It’s a new way to collect information. We post every morning a bunch of stuff we’re reading to get ready for the show. People get online and point us to new stuff. We’ve had a lot of story ideas for the show generated by people commenting on the blog” (Gillette 2010). Conaway expanded on what she felt the relationship should be between the blog and the show, which was that there should be little distinction between them: [My work online] is not about the blog. . . . It’s about, “Is the show, which now includes the blog, healthy?” Is it doing things that surprise? Is it breaking the form? Does it have more personality? Does it have more personalities? Did it find something that nobody knew today that made its way [into conversation]? Did it talk to people? . . . [When the

Out of One, Many  192

show produces an interesting segment], did that live [on] anywhere? . . . The blog helps make all that possible. But it’s really not about the blog. . . . The blog is hugely important, but it’s not the point. The point is the entire enterprise. In other words, in Conaway’s local teleology—the one she was hired to pursue on behalf of the Maddow program—the blog and the show would become complementary outlets allowing multiple platforms for the TRMS staff ’s journalistic pursuits: more ways of publishing, more ways of collecting and assembling information, more ways of engaging with audiences, and so forth. The blog was (ideally) a way of increasing support for and engagement with the same editorial enterprise, rather than a separate project to be judged on independent merits. It’s significant, too, that Conaway was not hired at random. She had personal connections to the show in the form of an existing friendship with one of the program’s executive producers, having previously worked with Bill Wolff ’s wife and another future Maddow producer, Tricia McKinney, on an NPR enterprise called the Bryant Park Project. The Bryant Park Project was a short-lived, experimental live news radio program named for the location of NPR’s New York studios adjacent to Bryant Park. Its goal was to reach out to younger and nontraditional NPR listeners by experimenting with the live news format, and ultimately by creating a bustling online presence, the hub of which was a blog run by Conaway that developed a reputation both as an interesting online destination and as an unusually interactive site for a broadcast organization. The blog contained a great deal of exchange between the show’s staff and its listeners, while both the show and especially its website made use of users’ online suggestions and contributions. After the Bryant Park Project ended, Conaway went on to help found the award-winning site for Planet Money, a new show at NPR that prominently covered the 2008 economic meltdown and also at times broke news based on users’ online contributions. In other words, Conaway’s approach to integrating blogging and broadcasting was far from unknown when the Rachel Maddow Show created a position for her. “Wherever I’ve been I’ve always done web, plus whatever the main medium was doing,” she said. “So, in a way, what marks my work is being both those things. And what that tends to make possible is the web is able to feed the product, whether it’s a newspaper or a show. And the show is able to feed the web. And hopefully it makes a more unified, interesting, expansive thing.”

The Maddow Blog Is Born  193

From the vantage point of the Rachel Maddow Show, the level of integration among the show’s various products promised by Conaway would likely have been difficult to achieve without hiring a dedicated web producer into the television office. Moreover, as Wolff ’s earlier quote suggests, the program’s staff were not merely enthusiastic about the generic idea of an internal web producer: rather, they were pursuing a particular person with a known editorial outlook. As part of this arrangement, Conaway was afforded a good deal of editorial independence with the blog on a show that was known for the close editorial oversight of its anchor. “I’m not sure it would be that way were it not for Laura,” said Wolff. “She is a very experienced person. She is a very responsible person. And I think she’s a very simpatico person in terms of her editorial outlook with Rachel. So then, in terms of the story selections, I just trust her implicitly.” Stepping back once more to take stock of what the show’s staff achieved to this point as system builders, we can see how the program staff not only employed heterogeneous means to maximize their resources within MSNBC.com, gather an assortment of complementary third-party tools, and push the web venture to give them a blog as quickly as possible—they also drew on their personal connections and redirected financial/­personnel resources given to them by MSNBC TV to acquire a new producer whom they trusted uniquely to manage their expanding web presence. The cumulative result of all this heterogeneous engineering was that the Rachel Maddow Show received early access to Newsvine’s tools and launched a blog on the new platform months in advance of the rollout of new blogs across the rest of MSNBC.com. Back on the West Coast during this period, around the end of 2009, MSNBC.com had actually already launched one Newsvine blog: the weblog for NBC News’s Elkhart Project, an experimental reporting initiative in which the network news division opened a bureau in Elkhart, Indiana, one of the cities worst hit by the recession of 2008, to report on the economic crisis and elusive recovery from a local perspective. The template for the Elkhart Project blog was the subject of some back and forth between designers at Newsvine and MSNBC.com’s Creative Development department, which was responsible for the look and much of the front-end functionality on the Blue Site. The final template for the blog came from Director of Creative Development Ashley Wells. Not entirely satisfied with the Elkhart template, Wells subsequently decided to hire an outside designer

Out of One, Many  194

to create the standard template that would be used in the mass rollout of Newsvine blogs on MSNBC.com. In late 2009, when the web venture decided to give an early launch to the Maddow Blog, the new designer, Martin McClellan, was in the middle of this redesign project that was now tied not just to the creeping release schedule of the new Blue Site story pages, but also to an ongoing upgrade of the Newsvine software architecture. With the M3 architecture, as the new Newsvine code was known, the Newsvine CMS went from being highly specific to the original Newsvine.com social network—a.k.a. the “Green Site”—with its specialized features and customizable widgets, to a CMS capable of being quickly retemplated to power a broad variety of sites and blogs. But M3 wasn’t ready at the time, and consequently neither was McClellan’s new blog template. Instead, a template was rapidly spun up by Newsvine CEO Mike Davidson based on the Maddow program’s specifications and using the previous M2 architecture, which had been employed for the experimental Elkhart Project blog on the MSNBC.com domain, but otherwise only ever rolled out in beta beyond the front page of the Green Site. The result of this rapid deployment was a site that looked far more contemporary than the various Community Server and TypePad blogs that adorned the Blue Site at the time, but could at times prove buggy or unpredictable from the perspective of its proprietors at the Rachel Maddow Show. While Newsvine worked to iron out these problems with the M2-based Maddow Blog, which was still one of the first-ever blogs they’d deployed for MSNBC.com, they were also rushing to complete development of the M3 architecture in time for the site-wide release of the “Martin Blogs,” as the columns bearing McClellan’s template were known internally at Newsvine. This dual assignment slowed Newsvine’s issue-response times substantially, and the developers expressed an awareness that this—along with the fact that the Maddow Blog had been in many ways Newsvine’s “guinea pig”—had probably frustrated Will Femia, Laura Conaway, and other East Coast staffers working with the show. Newsvine generally prided itself on its speed to market and abided by a “release early, release often” (Raymond 2000) development strategy, in which features were shipped quickly, warts and all, with the understanding that bugs would be eradicated with equal rapidity. Squashing bugs rapidly became more difficult, however, as the staff ’s workload mounted. And from the vantage point of TRMS, which had campaigned for a blog for nearly a year, the Newsvine software they received seemed like the worst

The Maddow Blog Is Born  195

Figure 10.1. The Rachel Maddow Show micro-site. Aside from a few cosmetic changes, the micro-site for The Rachel Maddow Show remained largely unchanged from its launch in 2008 until the redesign of MSNBC.com under Comcast in 2013. The key difference introduced during this period was that the standard space for links at the bottom of the page was replaced with a clickable forwarding message encouraging users to visit the Maddow Blog.

of both worlds—sometimes-buggy software that had taken months and substantial maneuvering to acquire. Despite any initial discontents, however, the Maddow Blog did everything required of it. All of the permanent links—to Twitter accounts, fan pages, and so forth—on the tv.msnbc.com page were moved to the blog’s sidebar, and all of the “fake blogging” activity that had been manually maintained until the launch was continued on the Maddow Blog with the help of the new Newsvine blogware. Within weeks Femia had closed out the temporary landing page on tv.msnbc.com, redirecting its traffic to the blog, and similarly wiped away the grid of links on the MSNBC.com micro-site, replacing it with a large graphic directing users to the blog for any purpose beyond viewing clips (Figure 10.1). Moreover, the new Maddow Blog template itself had links in the sidebar to all of the latest video clips, which meant that viewers no longer needed to visit the micro-site (Figure 10.2). At that point, virtually all that remained of the Maddow program’s original

Out of One, Many  196

Figure 10.2. The Maddow Blog as it appeared in 2011. NBCUniversal.

MSNBC.com web presence was the original video player page, which was otherwise devoid of content, but—as Maddow pointed out—continued to have the more intuitive URL (rachel.msnbc.com as opposed to maddow blog.msnbc.com) and was still where viewers landed when they clicked on any header navigation link on the Blue Site. Even so, Maddow and her staff viewed the new site not as a supplemental blog, but as their new website “run by all their own people.” BEYOND THE BLOG: A DISTRIBUTED WEB PRESENCE

At this point, we’ve witnessed the creation of the Maddow Blog from the vantage point of TRMS and the extensive heterogeneous engineering that took it from an initial idea and request for a blog to a complete website run, in many ways, independently of MSNBC.com’s Blue Site. But the local teleology behind the Rachel Maddow Show’s web presence does not stop unfolding here. While Laura Conaway managed the blog and was the only

The Maddow Blog Is Born  197

TRMS hire in 2010 to be dedicated primarily to the web, web savviness remained a job requirement for everyone affiliated with the show. Vanessa Silverton-Peel, for instance, who came with Maddow to MSNBC from radio, had previously helped to manage Maddow’s web presence for Air America. And the show invited another producer hire, previously at NFL Films, to interview with them after they noticed and thought highly of his Twitter account. The candidate, Jamil Smith, also submitted ten personal blog entries as some of his primary résumé items. Femia, with the help of Conaway and others, continued to develop the program’s active presence on Twitter. And the show continued to use many non-MSNBC.com services for presenting content. To take just one example, in addition to Flickr slideshows, which continued to appear periodically on the blog, the show made particularly heavy use of the videosharing service Vimeo to host its “video tweets” (videos created by the show staff and promoted on Twitter), other behind-the-scenes footage, and at times even news reporting. There were several reasons for this. One was that posting to a social site like Vimeo made TRMS content discoverable in another venue. Posting a video to Vimeo was also frequently much faster and simpler—particularly while traveling—than going through the more formal process of flipping a video through MSNBC.com’s CMS. And finally, the MSNBC.com player, with all its branding and accompanying playlists full of highly produced news content, could at times carry with it a formality that the producer wanted to avoid with a clip that was exceedingly brief or intended to be light. As an example, Conaway mentioned a clip she posted to Vimeo of Maddow waking up on an airplane during a reporting trip: “I didn’t have time to deal with flipping it to anybody else. I just flipped that thing in Vimeo and called it a day. . . . If I have to flip it for MSNBC, that means all of a sudden MSNBC has to own this video as a corporation. And they didn’t necessarily go out and assign this video of Rachel in bed-head. It’s not like a finished segment that it has to live alongside of. It’s just this silly thing.” In a McLuhan-esque turn, the informal aesthetic and associations carried by Vimeo, a service used primarily by artists and hobbyists, meant that the player could also be its own message to users, conveying a sort of “between us friends” intimacy antithetical to the corporate image put forward by MSNBC.com’s video player. Conaway recalled when the show was recorded in front of a live audience during a reporting trip to Anchorage,

Out of One, Many  198

Alaska. Seconds before the program went to air, she published a clip to Vimeo of the audience chattering in anticipation of the broadcast: [I] didn’t even blog it. . . . I just tweeted it. It was a way of saying, “Look!” to this little dedicated audience who help us all the time. . . . “This is what you’re about to see.” In a way it’s a pitch for the show, and also it’s a way of just saying, “Hey, you guys, you got my back, right? You’re looking at this the same way I am.” And that doesn’t need to go on MSNBC.com. That process takes longer and there’s no reward for that. I tend to like to put things on Vimeo and put them on the blog in Vimeo form because they are friendlier . . . and so I think people will move them around. Here we can see how Vimeo had solved some problems for the Rachel Maddow Show. It allowed staffers to deliver video to viewers more rapidly and easily from the road, which was a rather large affordance; in fact at the time there was no standard method for publishing video from the field. Generally field reporters, if they’d shot something themselves, would have to send it back to MSNBC.com—a process that involved both finding a reliable Internet connection and figuring out a way to move a large file. According to Femia, an online dropbox service called SendThisFile had become a popular kludge for phoning home with newly shot video. Vimeo, however, allowed the producers to publish a video quickly after it was taken—in the example Conaway gives earlier, in the fleeting moments between when the crowd was eagerly anticipating the beginning of the show, and when the show began.2 Moreover, posting a Vimeo video made the source file immediately available for download to producers logged in to the same account back in New York. In other words, using Vimeo was a method not only of quickly and easily publishing video online, but also of moving large files. And while the staff didn’t hesitate to use Vimeo clips instead of the MSNBC .com player on the Maddow Blog, Femia would frequently enter clips that originated on Vimeo into MSNBC.com’s CMS anyway so that they would syndicate to the show’s mobile apps. Where desired, he could also toggle the CMS settings to farm the video out to the rest of the Microsoft’s MSN and Bing video destinations, or conversely to prevent relatively informal clips from being syndicated. As we saw, Vimeo was also useful to the staff in that it changed, they hoped, the expectations of viewers regarding what was in a clip. If audiences

The Maddow Blog Is Born  199

associated the MSNBC.com player with professionally produced content, Vimeo signaled more of a peer relationship that aligned well with the technical capacity, aesthetic purpose, and aura of informality surrounding clips shot on producers’ FlipCams and mobile phones.3 In Conaway’s view, one goal of posting such content was to break down the traditional one-way concept of broadcasting by accompanying professional television content with “a very friendly web presence, which really is mostly what the blog, as a form, is designed to convey. It’s designed to look lo-fi and inviting, as opposed to crisp and clean and elevated. And it’s designed to look kind of broken, a little bit—like there’s some cracks in it and there’s a light shining through.” This is an important aspect of the philosophy behind the original Maddow Blog, whose editorial strategy aimed at making content seem less like broadcast and more like a channel where someone was listening on the other end. In other words, producers’ collection of “home video” with phones and FlipCams (as in Conaway’s examples of the airplane and Anchorage clips) in combination with Vimeo’s player constituted a set of resources through which TRMS accomplished a combination of provincial objectives: improved distribution, to be sure, but also editorial and ­community-building aims—all of which had become deeply intertwined. THE CABLE SITE

It is time now to step back into the vantage point of MSNBC.com, while remembering that such a perspective is a partial fiction given that it, too, is a heterarchic organization made up of many system builders. As we’ve seen, MSNBC.com over the years constructed something of an isometrically stable system predominantly aimed at repurposing and distributing original television content from MSNBC TV and NBC News for the web. The Rachel Maddow Show put pressure on this system by taking hold of the same resources and building toward an online presence optimized for original web content (that has at times even been repurposed for ­television). In a rigid, hierarchical universe of systems, we might expect the nail that sticks up to get hammered down. On paper, after all, MSNBC.com in many ways held the reins to the Maddow program’s web presence. It owned the domain name, the blogging software, the database, and the video rights. But we’ve already established that we don’t live in this sort of universe. In fact, MSNBC.com wrote software for downloading television-quality

Out of One, Many  200

video from FlipCams, so that shows like TRMS that used them to shoot original content for the web could more easily repurpose their content for air. They also directed Newsvine to include native support for Vimeo in the new post-authoring interface that it was releasing on the M3 architecture. In the same way that Maddow’s large Twitter following was attractive to the web venture, “if all of a sudden, an NBC News correspondent’s Vimeo account was seeing a huge amount of video traffic, we would look at that as an opportunity,” said MSNBC.com’s director of multimedia Stokes Young. From any vantage point, the illusion of stability in a system comes from a balance of conflicting pressures. Whether a system is stable or in flux depends in large part on the vantage point one chooses, and it’s when this isometric balance is upset that change happens and organizations evolve. Just as we saw how Sam Go’s team, for example, quickly reframed the Maddow program’s incursion on their resources as the recalcitrant bit of a new resource, rather than looking at it as an unequivocal obstacle, MSNBC.com ultimately saw, in the system TRMS had built, a unique and important opportunity—one that again grew out of a larger tension. As we’ve seen, the cable news channel’s online presence in some ways lived at the margins of MSNBC.com’s general-news experience, taking the form of show-centered micro-sites that rarely surfaced on the MSNBC .com front page beyond a few navigational links. For a long time, this sort of peripheral presence was acceptable, but as online distribution and the conversation economy grew in influence, expectations changed and there was new pressure to raise the profile of the cable channel’s content and personalities. While early into the joint venture cable channel employees had tried their hand at managing some of the website’s content related to their programs, as we saw in previous chapters they very quickly turned these responsibilities over to Microsoft and its MSNBC.com staff, who had to find a way to create an online presence for these programs largely without their direct involvement. Even then, as the TV channel flagged in the ratings prior to 2007, the largely static sections of MSNBC.com dedicated to cable programming drew relatively little in the way of traffic or revenue. As Randy Stearns, MSNBC.com’s East Coast deputy editor, put it, “It was hard to make a good case for moving them closer to the center of the news experience” at MSNBC.com during that period. As MSNBC TV finally began drawing respectable ratings in the wake of Keith Olbermann’s success, however, and as demand for its content

The Maddow Blog Is Born  201

began to increase online, MSNBC.com’s system for putting television online began to destabilize—or to put it another way, this system that had once achieved closure reopened (Kline and Pinch 1996). Stearns was one online employee who found himself at the center of this tension when I spoke with him in 2010. “Cable has been a source of endless frustration and disappointment for me,” he said. As the cable channel in general, and Olbermann’s program in particular, began to pick up steam, he recalled, “It became more apparent to some of us that there was a lot more online potential with cable. . . . It really bothered me that the [editors of the MSNBC .com cover] would not touch cable primetime . . . because it had such a political bias, even though [the content] was often incredibly strong, and even though, particularly in 2007 and 2008, what Olbermann was saying every night actually set the agenda for the next twelve to twenty-four hours in some ways, in terms of what people were talking about.” Stearns argued that the left-wing commentary delivered by Olbermann and other cable hosts could be clearly labeled as opinion, and—with this disclaimer attached—appear in the more central pages of the MSNBC .com flagship site. He recalled the case he and his staff made to the other editors at MSNBC.com: “Nobody else has it. People want it. We should make it easy for people to get it. . . . We shouldn’t let other sites take advantage of it while we marginalize it.” He continued, From where I sit, in this building with these people at NBC, and looking at it from the vantage of contemporary media consumers, it’s shortsighted to think that users have trouble distinguishing between content types and points of view. It doesn’t take into account the way people consume information today. It’s bad business. It’s not fair to our colleagues at cable. I know them, they work incredibly hard, they’re incredibly lean, understaffed. And they deserve to have a home online. And we should be the ones doing it. . . . There’s a certain amount of consternation here about how that’s going to play out. And in fact, around this time in late 2010 the cable channel began its own heterogeneous engineering project aimed at expanding and tightening control over its web presence. According to some reports, as part of this effort MSNBC TV even threatened to buy its own website, independent of MSNBC.com (Stelter 2010b). Whether or not acquiring a liberal commentary site like the Huffington Post was actually under serious consideration at the time, this sort of outspoken resistance put new pressure on the web

Out of One, Many  202

venture to provide a fresh solution to multiple, and potentially conflicting, agendas. Individual TV programs beyond just the Rachel Maddow Show had also taken measures to gain more control over their individual web footprints. When MSNBC’s newest primetime anchor, Lawrence O’Donnell, launched his show the Last Word, his staff included a web producer on NBC Universal’s payroll in the same mold as Laura Conaway. And a number of shows, including the Last Word and Ed Schultz’s the Ed Show, created homepages that defaulted to Newsvine blogs built around the same “Mike Blog” template that Newsvine had designed for the Maddow program, as opposed to the traditional video-centric template that was standard on MSNBC.com.4 Once again, however, MSNBC.com found a way to translate the demands of a seemingly conflicting system into its own terms. You’ll recall that MSNBC.com’s expansion strategy had been to leverage different segments of the online market for news. And as it turns out, while it owned a large mainstream news site (MSNBC.com), a news-themed social network (Newsvine.com), a breaking news site (BreakingNews.com), a hyper­ local news venture (EveryBlock.com), and a lifestyle website in the form of Today.com, it did not yet own a site dedicated to opinion and commentary. Building an online answer to the cable channel’s “op-ed television” brand, however, was an ideal way to corner another portion of the news market, and very much in keeping with this new market strategy MSNBC.com had developed in the wake of its ownership split from MSNBC TV in 2007. MSNBC.com chief Charles Tillinghast explained, “The whole commentary segment of news has grown up over the last five years or so, particularly with the Huffington Post. As general news becomes increasingly commoditized, you need a way to differentiate yourself.” In this case the answer to the need to offer a unique interpretation of the news was an op-ed-themed cable site. By late 2010, then, the cable channel appeared likely to receive its own website. One issue in creating such a site, however, was that MSNBC TV’s various television shows were putting drastically different volumes of content online. It would have been relatively simple, from an editorial perspective, to build a website around a parade of shows that all produced nearly a hundred original articles a week, as the Rachel Maddow Show did. But some shows rarely posted to their blogs and others had no blogs or web-original content to distribute at all. Thus, in planning for a potential cable site MSNBC.com requested that every MSNBC TV program hire a

The Maddow Blog Is Born  203

web producer on the NBCUniversal payroll, or as Sam Go put it, “a Laura [Conaway] for every cable show.” This strategy left some large questions unresolved, however. Part of Conaway’s success as a member of the Maddow team came from the extra­ ordinary level of trust that had developed between her and the rest of the program’s producers, something that would be hard to replicate simply by copying the program’s staffing model. Moreover, some of the sources I talked to were unsure whether the Rachel Maddow Show, after working long and hard to establish an online presence that felt independent of MSNBC .com, would welcome the idea of folding its site into a new cable destination, where the show would again be one content provider among many. In either case, by late 2010, Charles Tillinghast and others were indicating that an editorially independent cable site was probably on the way, and such a site was indeed realized even after Comcast ultimately bought out Microsoft’s shares in MSNBC.com, dissolving the joint venture. The shift to a standalone cable site, which began in 2010, provides an object lesson in the relative nature of systems. From the vantage point of the executives and employees at MSNBC TV who had voiced dissatisfaction with their representation online, and had even threatened to buy an alternative website, the move toward the cable site might be seen as a victory, wherein MSNBC.com was forced to build a more robust web presence for MSNBC TV and relinquish a great deal of editorial control to the TV side of the joint venture. By contrast, from the vantage point of MSNBC.com, the same site would turn MSNBC TV into one more content provider in its collection, one that promised to reach an as yet underutilized segment of the market. As an added bonus for MSNBC.com, the newly proposed staff structure would mean that employees on NBC’s payroll would contribute more content to the MSNBC Digital Network than they ever had before. I’ve put this contrast in stark terms to make a point—as Stokes Young highlighted, there was caution at MSNBC.com not to view NBC News content, including that from cable, as a simple commodity. Likewise, those on the NBCUniversal side of the payroll were aware of the substantial boost in web traffic they enjoyed because of their inclusion in the Blue Site and the MSN network. It’s easy to see how the outcome of the conflict I’ve described—in which both parties believed they’d won the day—could also be viewed as a mutually agreeable solution. The larger point, however, is that the cable site was a highly particular distribution solution that served as an actor in multiple overlapping systems.5

Out of One, Many  204

This observation is further underscored by the fact that, at the time of my fieldwork, the Newsvine development team and its platform were among the leading contenders within MSNBC.com to construct the new site. “One of their functions is to power all of the community features on our site,” said Tillinghast, noting that the Newsvine platform contained much of MSNBC.com’s user-participation functionality. “A lot of the features the cable site needs are driven by that functionality. So it may make sense just to build the whole site on Newsvine.” And indeed, even after MSNBC and Microsoft parted ways, Newsvine’s software would ultimately become one of the primary ingredients of the new cable site.6 INTERACTING SYSTEMS

At this point we’ve traced out the progression of Newsvine from an independent startup to an “arms-length community” site owned by MSNBC .com to one of the primary content management systems helping to power the MSNBC Digital Network and MSNBC TV’s web presence, as well as a potential platform for the construction of a new cable site. We also saw how the Rachel Maddow Show went from having a templated web presence on MSNBC.com, to faking blogware and using third-party web tools, to launching a blog robust enough to serve as a standalone site and hub for the show’s social media efforts, and finally to serving as a new template for the web efforts of other MSNBC TV shows and the exemplar around which the MSNBC Digital Network planned its new cable site. Looked at from the outside, these are major transformations in the digital distribution strategy for MSNBC.com and MSNBC TV. From an analytical standpoint, however, it makes less sense to look at these shifts as the product of a monolithic organization than to examine them as changes wrought by numerous system architects with provincial needs and norms, building toward local teleologies that employed overlapping sets of resources. This is the insight identified by both sociologists of systems and organizational scholars when they point to the numerous stakeholders who influence the design of artifacts and the social practices surrounding them. The aspects of a particular actor that make it a resource for one system builder frequently make it recalcitrant for another. These can be seemingly small—the automated profanity filter that replaced written cursing with grawlix was a boon to Newsvine moderators policing their social network’s comment threads, but an annoyance to authors at MSNBC.com when they found that their blog posts on the flagship site had been sur-

The Maddow Blog Is Born  205

reptitiously edited by the same algorithm. They can also be substantially larger, as when the micro-site template that worked well for cable shows with limited initial interest in their web footprint proved confining for the TRMS team that wanted to capitalize on the web, or when the norms of community policing and editing developed for Newsvine’s independent citizen-­journalism community began to break down when applied to a mainstream media site. These are also examples of the pressure that’s applied to local teleologies when a new actor is lashed into an isometrically stable system. Selling to MSNBC.com shored up Newsvine’s finances, but put dramatic pressure on the previously stable social components of its platform. Acquiring Maddow as a content provider gave the Blue Site access to a large and highly engaged Internet audience, but put pressure on its previously streamlined system for repurposing television content. These observations are not dissimilar from those in previous chapters concerning how MSNBC.com engineered a volta for online video. From whatever vantage point we choose, we see how various actors simultaneously provided resources and exerted pressures that influenced provincial distribution strategies and their implementation. What’s different is that we’ve come to understand how MSNBC.com, like any large media organization, is not a monolith. It is not a single actor, but a construct—a line drawn around the activity of numerous system builders whose goals are sometimes in concert, but other times in conflict. In one of my interviews, Will Femia nicely described the provincial and heterarchic nature of work at MSNBC, highlighting the numerous local teleologies in play at any given time: The marketing team might have an idea [for a campaign]—“Lean Forward.” Dot-com might have its own idea—“Fuller Spectrum of News.” PR might be pitching one way or another. . . . But, ultimately because I’m working with the stuff that the [Maddow] show is producing, to me it feels more tangible and more important than the bigger-picture brand pushes. . . . I’m explaining why I don’t feel a real conflict with them. Because I feel if what they’re doing serves the show, and can work with the content that I’m handling, then great. And if not, then I don’t feel any guilt not playing with them on whatever it is they’re trying to get done. . . . It sounds dicky, but it comes also from the awareness that every team has its own individual politics and goals. I know that the PR lady has to show this many media mentions, or she’s

Out of One, Many  206

trying to impress this particular boss. Those goals don’t necessarily align with my goals or the show’s goals. Being aware of that frees me from thinking that I have to use everything they’re doing, or to participate in whatever deal they just made. This is not a description of rigid organizational hierarchy, but of heterarchy and of heterogeneous engineering—of the lashing together of recalcitrant resources, and the trickiness inherent in navigating an environment full of distinct, but overlapping, local teleologies. In other words, it is not unlike the universe of system builders outside the organization. In making this observation, we might tentatively join the ranks of scholars challenging the very construct of organizational boundaries (Child and McGrath 2001). But what would it mean to do so? To call into focus the constructed nature of organizational boundaries is not to suggest that they have no meaning, but rather an invitation to consider the work that they do. As Robert Chia (1995) notes, in an environment full of flattening hierarchies and project-based economies, the role of contemporary sociologists is, in part, to rediscover what it means to have an “organization” in the first place. Why draw this line? There seem to be objective reasons for doing so. We’ve already identified how media organizations tend to contain overlapping systems and resources. Newsvine’s system and the Rachel Maddow Show’s, for example, intersect in crucial ways. But stepping outside the organizational boundary, consider the number of individuals, groups, and organizations that have a stake in the LAMP solution stack, a publishing platform like WordPress, a social network like Twitter, or a video site like Vimeo or YouTube. These artifacts, platforms, and social environments are all both themselves the product of local teleologies and crucially enrolled in many systems simultaneously, bearing the impressions of numerous heterogeneous engineers. Another reason why it might make sense to think of organizations as singular, despite their complexity, is that issues of ownership and contract require them to be so, at least in the eyes of the law. After all, when it comes to a company like General Electric, which until recently owned NBC Universal, “What are the synergies of film and TV production with jet engines, nuclear reactors, refrigerators, and financial services?” (Noam 2009, pp. 437–438). Ownership and contracts are clearly the connecting thread between all these things—but this puts us right back at the question of what work organizational boundaries do. And one answer is that they

The Maddow Blog Is Born  207

are framing devices, defining what is internal to the enterprise and what is an externality to those seeking to make money off the arrangement (Callon 1998). Put another way, an organizational boundary is an actor in someone’s system, a resource developed to do work—in this case financial work. But as we’ve seen, an actor, once on the scene, rarely goes unenrolled by other system builders. An organization, once it exists, becomes not just a means for its owners to make money or for its employees to draw a paycheck. It also does cultural work as the starting point for a brand, an identity—“progressive,” “innovative,” “environmentally friendly,” and so forth—that can be used to speak to particular audiences and consumers. This identity can be put into financial terms, too—such as “leveraging different areas of the market.” A brand is also not merely a resource for those inside the organizational boundary; it serves as a resource for those who would criticize the organization, too. Recall Femia’s observation that when it came time to moderate comments, NBC News, MSNBC, and Microsoft all meant something beyond what individual producers could control. As Nick Couldry (2008) might point out, there were great asymmetries of resources and power between the editor at MSNBC.com, with access to a host of moderation tools, and a commenter leaving flames across the site, but even so it doesn’t take an “insider” to wield a brand as a resource. What’s more, we need look no further than the divergent editorial strategies of MSNBC TV and MSNBC.com to see that corporate boundaries and brands can prove as recalcitrant for insiders as for anyone else. Among the work such boundaries do is to shield from the public eye the messy, negotiated—that is, heterogeneous—nature of systems assembled by the actors inside the firm. As an example, viewed from the outside, the stories I’ve told thus far might together look something like this: MSNBC.com was one of the first mainstream media organizations to invest heavily in social networking, acquiring Newsvine in 2007. In 2010, further recognizing the commercial importance of the social web, MSNBC.com fully integrated Newsvine’s community features into its primary domain, demonstrating its commitment to the enterprise by pilot-testing its new community blogging software with one of MSNBC TV’s flagship primetime programs and subsequently releasing the platform across the whole of the Blue Site. This narrative, while not untrue, is also not terribly nuanced. As we saw, the Rachel Maddow Show applied substantial resources to obtaining

Out of One, Many  208

early access to the new blogging platform. And while Newsvine certainly became a central system within MSNBC.com, its integration into the Blue Site caused a good deal of turbulence for the former startup and the original community it had started. As we’ll see in the next chapters, from the vantage point of some system builders at MSNBC.com, Newsvine’s new blogging platform proved recalcitrant to varying degrees as well. Identifying organizational boundaries and their associated brands as resources enrolled in multiple systems demonstrates how important they are as actor categories. At the same time, from an empirical perspective we would do well to steer clear of employing them as a priori analytical categories. It turns out that they are instead just some of the useful tools to be grasped at by a universe of system builders. Online distribution paths, in turn, are the result of the pressures exerted by all of these actors, and institutional affiliation matters only insomuch as it serves as one more resource and/or barrier shaping these systems. Having thus deconstructed the monolithic guise of media organizations with which we began, and dispensed with the corresponding division of the world into internal and external forces, we’ve reached bedrock: an unbroken fabric of overlapping, heterogeneous systems across which content is distributed. We’re now prepared to think more directly about the influence of this fabric of systems on our information environment.

11//one size does not fit all heterogeneous audiences and proliferating brands

Gaye Tuchman’s 1978 newsroom ethnography Making News has become a landmark text in journalism studies and media sociology more generally. One of its great contributions is to re-envision the study of journalism as an exercise in the sociology of knowledge. She flags journalism as just one among numerous “organizationally and professionally produced” systems of knowledge, including science and sociology, as well as creative industries like film production and the television business (p. 217). Moreover, since Tuchman’s initial writing, the number of creative industries has exploded, with many traditional sectors of the economy refashioning themselves as producers of cultural products and intellectual property (Deuze 2007). Correspondingly, authors like Karin Knorr Cetina (1997, 1999, 2001) argue that we are gradually becoming a knowledge society, replete with an expanding number of epistemic cultures—intersecting systems of knowledge that are no longer bounded by organizational cultures, but have become a part of everyday life. Knorr Cetina (1997) takes this “discharge of knowledge relations into society” as one of the new key concerns for contemporary social theory (p. 8). To grasp the value to be taken from the notion of epistemic cultures, as well as which parts of the concept we might have to leave behind in this discussion, I’ll take a moment to better explain the construct. The phrase originally belongs to Knorr Cetina’s (1999) comparative study of two scientific fields, high-energy physics and molecular biology—a study that revolves not around the social construction of knowledge itself, but in her words, “the construction of the machineries of knowledge construction” 211

Mass Reach and Mass Media  212

(p. 3). By this she means the social and technical arrangements arising from the “affinity, necessity, and historical coincidence” that surround different communities of investigators and give rise to scientific knowledge (p. 1). In short, she coins the term in an effort to describe how different scientific disciplines are not simply pursuing different lines of inquiry, but constitute different cultures of inquiry, whose objects of study become the focus of an “object-centered sociality,” giving rise to distinct practices, social structures, and shared ontologies, all of which in turn discursively shape the knowledge they produce. In doing so, she carries forward in a compelling fashion Tuchman’s original charge of identifying and explicating different systems of knowledge and cultural inquiry. But Knorr Cetina (1997, 2001) adds an extra element to this project: she proposes that as we become a knowledge society, the notion of epistemic cultures will increasingly apply beyond institutions to informal systems of knowledge production and ultimately to our everyday experiences. In this vein, many scholars point out that a great deal of contemporary creative work online, from open source software to blogs to fan fiction to encyclopedias, is produced not by professionals or creative industries, but through commons-based peer production undertaken by loosely confederated groups of individuals collaborating for free or as part of novel economic arrangements (Benkler 2006; Bruns 2008; Hippel 2005; Lessig 2004). Axel Bruns refers to the networks of collaborators that grow up around these projects as “produsage communities,” and they bear a great deal of resemblance to Knorr Cetina’s (1997, 2001) epistemic cultures in the sense that their sociality is organized around common objects of inquiry or production. Moreover, in Bruns’s (2008) interpretation, an online community can even be its own object of communal production, with users evaluating and assigning merit based on one another’s contributions to the ongoing discussion, irrespective of any intent to create any artifact beyond the forum itself. At the same time, some online interaction is so informal and atomized that it hardly looks social at all in the traditional sense of the term. For example, while there are forms of peer production that require a great deal of cooperation aimed at building a common artifact, such as a software release or an encyclopedia article, many communally produced resources result from far less coordinated and task-oriented user activity, such as when users individually tag photos on Flickr resulting in a searchable image archive, or improve search results when their linking activity is aggregated

One Size Does Not Fit All  213

by a search engine (Bruns 2008; Shirky 2008). While the least coordinated of these actions are often taken with an awareness of and an eye toward a larger unseen audience or user base and its informational wants and needs —therefore involving epistemic concerns—it’s hard to think of them as being rich enough in and of themselves to merit the “culture” label. Rather we might think of activities like tagging or linking more as epistemic encounters: invitations to sociality, or at least to the recirculation of ideas and content. But whether we’re discussing the highly coordinated activities of a tightly knit online community, or far more informal epistemic encounters—as when a Twitter user adds a series of hashtags to her post in hopes of piquing the interest of different imagined audiences—the shared links and embeddable media objects that drive so much traffic in the conversation economy are propagated by users who share them only when doing so fits their personal or group objectives. As Today.com director Jennifer Brown put it, it’s important that the links the site feeds out to social networks be either “something you want to discuss and comment on, or something you want to share because it will make you look cool among your friends.” This isn’t necessarily a new idea. As Jenkins (quoted in Usher 2010) puts it, “Journalists have long embraced a central idea . . . that content represents a resource which community [members] use to talk amongst themselves.” People were sending newspaper clippings through the mail and taping TV segments to share with friends long before “social media” became a buzzword. At least three contemporary trends, however, have dramatically increased the importance to media practitioners of the conversation economy. Chuck Tryon (2013), drawing on a notion of “flexible labor” articulated by David Harvey (1989), outlines the first of these shifts: the rise of contemporary labor routines in which employees work variable hours and often mix work, personal commitments, and leisure throughout the day. As a result of this shift, many aspects of our media diet are no longer regularly scheduled and shared by a large synchronous audience (so-called “appointment viewing,” for example), but instead involve “snatching brief moments of leisure” (p. 10) during which we seek out the media of our choice—a phenomenon Max Dawson and Lynn Spigel (2008) call “flexible leisure.” A second, related factor making the conversation economy especially salient to media practitioners is a dramatic increase in media choice, a trend that has been commented on not just by Tryon (2013), but by many

Mass Reach and Mass Media  214

others such as Markus Prior (2007), Cass Sunstein (2007), and Philip Napoli (2011). Not only are we consuming different content at different times, in the age of hundreds of cable channels, hundreds of millions of websites, and a growing menagerie of personal electronics, our collective attention is now dispersed across more media on more platforms than ever before. In short, since the mid-twentieth century, audiences for media have become dramatically fragmented. Finally, as Napoli (2011) and other scholars (C. Anderson 2011, 2013; Braun 2014; Turow 2004) have collectively argued, contemporary audience metrics are making this fragmentation of audiences highly visible to media producers. Like an artist’s study of figure and ground, this revelation has come to practitioners both through (1) the provision of new and highly detailed analytics for various forms of digital media, which give positive proof of the dispersion of audiences and users across different venues, and (2) drops in program ratings and other traditional audience exposure metrics, which provide the negative proof of the same trend. As Napoli (2011) has illustrated, this tailing off in traditional ratings has resulted in a desire among many in the media industries to find alternative metrics for valuing their content in discussions with advertisers and other patrons. Such alternatives include factors like consumers’ attentiveness, loyalty, appreciation, emotional response, recall, attitude, and behavior—all ideas that get collectively tossed around in competing definitions of the highly freighted term “engagement” so often used by media practitioners. According to Napoli (2011) one of the questions that remains as yet unsettled from an industry perspective appears to be exactly how best to operationalize various aspects of audience engagement in a way that can be monetized as reliably as, say, traditional Nielsen or Arbitron ratings have been. As he highlights, this is ultimately a slow process, wherein many entrenched stakeholders—heterogeneous engineers, all—are still heavily invested in traditional audience metrics and their continuation. What is quite clear in the meantime is that the sharing and circulation of content by users through blogs, websites, and social media have become leading proxies for engagement, and are seen as a means of bolstering more traditionally minded metrics of audience exposure. In addition to generating clickthroughs to online stories, practitioners also hope to take advantage of the ostensible power of synchronous social media like Twitter to draw viewers back “into the rhythms of the broadcast schedule” (Tryon 2013, p. 9). Napoli (2011) refers to media industries’ conception of their audience

One Size Does Not Fit All  215

as the institutionalized audience. C. W. Anderson (2013) in particular has examined how digital distribution has influenced institutionalized news audiences and in the process documented a “fragmenting of the image of the local public” (p. 7): the way that journalists’ notion of their audience as a unified public living within the traditional circulation boundaries of their newspaper or broadcast affiliate has been upset by consumers’ shift to online news consumption. Among those who work in the news, the public has come to be seen instead as a collection of interconnected, but comparatively discrete audiences and communities whose consumption and sharing are based around particular topics, interests, and points of view. In short, in a fragmented, high-choice media environment where exposure depends in part on engaging a user base to share and discuss an outlet’s content—and where engagement is gradually coming to be seen as valuable in its own right—news work is increasingly a series of epistemic encounters. Journalists and others within news organizations are becoming ever more sensitive to the manner in which their output must “fit into the circuits” (Jenkins quoted in Usher 2010) of different online networks, communities, and conversations. EPISTEMIC CULTURES AND DISTRIBUTION

How do audience fragmentation and high media choice affect distribution decisions? As organizational scholars have noted, in the course of their daily activities, parties inside organizations frequently circulate documents and resources that are valued differently and for different reasons across the various groups that utilize them and the contexts in which they appear. This capacity to collaborate in the absence of shared understandings, or even in some cases shared goals, has for many scholars evoked anthropological notions of trade between cultures—the term “trading zone” coined by Galison (1997) is an oft-cited example. Also popular in describing such phenomena has been the idea of “boundary objects” (Star and Griesemer 1989), which refers to objects and ideas that circulate and hold currency across different social groups. Fred Turner (2006) has more recently introduced the idea of the “network forum” to describe the potential of shared physical and media spaces to forge new epistemic cultures from the participation and interactions of existing ones. But as Turner (2006) in particular highlights, if media texts are heterogeneous products, produced and circulated by groups pursuing distinct local teleologies inside organizations, this is often just as true when it comes

Mass Reach and Mass Media  216

to the manner in which contemporary media circulate outside of organizational contexts. Online, for example, the production and consumption of texts are intimately linked (Bruns 2008). As discussed in Chapter 5, the need for journalistic output to fit into the circuits of many different provincial systems of knowledge can have significant implications for distribution. The success of MSNBC.com’s embeddable video player, for instance, was predicated on its ability to work in many contexts—to let users rapidly locate the clip they wanted, precisely cut out only the part that fit their provincial needs, and embed it on (nearly) any site or service they chose. An enormous amount of heterogeneous engineering went into creating this flexibility. What’s more, the embeddable player is only one example of how different groups within MSNBC.com approached the challenge of distributing texts that worked across multiple systems and contexts. MSNBC.com undertook a number of major projects aimed at dealing with the challenges of moving content in the conversation economy, which we now understand to involve the selective uptake and movement of material by users, that is, members of different online audiences and subcultures that operate with widely varying degrees of formal coordination. Such audiences’ needs may involve anything from the maintenance of friendships (perpetuating an inside joke, say, by sharing a link that’s funny only in a very local context), to authoring a topical blog, to editing a Wikipedia article, to building a social network following.1 THE PROLIFERATION OF MSNBC BRANDS

Amit Nizan, MSNBC.com’s director of brand marketing, defined a brand as “a personality attached to a product,” explaining, “It’s not just the actual [product] that is important to people. It’s what it’s connected to, what it feels like. Those sort of intangible qualities that make something unique. . . . [Individuals likewise] have unique personalities. The way that we talk, the way that we dress, everything we do contributes to how we are all different as people. The same thing applies to brands.” As I explained in Chapter 4, brands proliferated at MSNBC.com as the company expanded itself into the MSNBC Digital Network. The goal frequently espoused in press releases, as well as in my interviews with MSNBC.com executives, was to use different brands—alter egos of a sort, in the manner described by Nizan—to extend the company’s reach into as many online contexts as possible. This strategy was clearly motivated

One Size Does Not Fit All  217

by the fragmentation of audiences in a high-choice environment. CEO Charles Tillinghast explained: “We want to have multiple brands that are attracting different segments of the news market. We don’t view the news market as a one-size-fits-all. We’re trying to tailor our products to fit the different tastes of different demographics. And by building sites that align with the show brands as well as ones that don’t have shows, we’re able to do that.” Between 2007 and its ultimate dissolution in 2012, the MSNBC Digital Network purchased Newsvine to target engaged news readers it dubbed “news explorers,” pursued the local news market through the acquisition of hyperlocal news company EveryBlock, launched an internal startup called Breaking News targeted at social media users looking for the latest headlines, and spun off the web presence it managed for NBC News’s Today Show into a new site, Today.com, aimed at producing life and style content for women ages twenty-five to thirty-four. Within many of these “master brands” were “sub-brands” aimed at leveraging even more niche audiences. For example, the Today brand online was spun off over the course of 2011 into multiple columns, blogs, and sub-sites aimed at different groups, including Today Moms for mothers; The Look for fashionistas; The Clicker for “TV junkies”; Life, Inc. for frugal living and budgeting tips; Scoop for celebrity gossip, and numerous others.2 To give an idea of the level of specificity with which brands were being deployed, the Today Food sub-site, which featured recipes, even had its own spinoff, Bites, that showcased culinary trends targeted at foodies who didn’t necessarily enjoy cooking. Similarly, the Blue Site, MSNBC.com’s flagship property, rapidly added sub-brands. To give just one example, in 2010 the site launched a new tech blog, Technolog, that was quickly spun off into two additional blogs, including In-Game, which focused on gaming culture, and later Gadgetbox for hardware news and reviews. Many of MSNBC.com’s technology writers also contributed to Digital Life, Today.com’s life-and-style-themed tech blog. The language of branding was also pervasive in the dot-com office in New York, and not just among individuals who worked on the web presence for television, but also among other Blue Site editorial staff. Shows, blogs, anchors, columnists, sites, and television networks were all discussed in terms of their brand by MSNBC.com staff. Moreover, not all of this brand-speak emanated from managerial circles carving up the news market. The decision to launch many of the themed blogs, for instance, originated within editorial circles, and a number of columnists with whom

Mass Reach and Mass Media  218

I spoke discussed their columns, or even themselves, as being “personal brands” aimed at consumers and communities with particular interests. At the same time, talk of branding occurred comparatively rarely among the television staff. The Rachel Maddow Show’s MSNBC.com web producer, Will Femia, opined that the comparative ubiquity of this language among online editorial staff had a good deal to do with the far greater extent to which dot-com workers were responsible for the distribution of content: It literally is a difference in jobs. . . . The [television producers for the] Maddow Show—they’ll say outright, “It’s not going to rate, but I want to do this story.” And they’re doing it. Then as the producer your job is to produce that segment, not pitch that segment, not market it somewhere, not write the cleverest headline you can so that it’ll get more clicks, not send it to Huffington Post. The stuff that we do online[, by contrast,] has a strong marketing component that requires thinking about [these things]. Even if you’re not thinking about traffic, and even if you’re not thinking about your brand, by virtue of having to pitch stories and figure out their placement and by watching your audience’s reaction to what you’re putting up, you can’t avoid it. [Emphasis in the original.] In other words, online staff were constantly thinking about epistemic concerns: the environments in which the piece they were working on or the content they were handling would play, who would take it up as a resource, where it might go as a result, and what the users in that target environment might do with or to the piece.3 THE RECALCITRANCE OF BRANDING

This proliferation of niche brands and editorial products aimed at every interest and corner of the news market was one of the dominant strategies at MSNBC.com for responding to the diverse and provincial demands of users participating in the conversation economy.4 Whether a user was looking for headlines, game reviews, parenting tips, political commentary, breaking news, celebrity gossip, or information on a product recall, the idea was to offer a corresponding product. But the demands of the web and the logic of branding, whereby—as Nizan put it—different sub-brands must cohere with the image and personality of a parent brand, could also be recalcitrant from the perspective of the provincial system builders at MSNBC.

One Size Does Not Fit All  219

A strong television brand, for instance, did not always warrant a major web presence from the perspective of MSNBC.com’s management. One large reason for this is that the amount or type of content necessary to forge a strong brand identity could be far less than the amount of content needed to sustain a major website. We encountered this difficulty before, when considering MSNBC.com’s request that the cable channel hire additional web producers to develop content for a standalone MSNBC TV site. But it was a concern that extended beyond the fate of the cable channel’s web footprint. An interesting contrast to consider with regard to branding and web presence is that between the NBC Nightly News and NBC News’s Today Show, both of whose online distribution was managed by MSNBC.com. As at all the broadcast news networks, the evening news program, Nightly News, was NBC News’s flagship program. It was—and is—considered the centerpiece of the network’s news division and television news lineup, and in many ways the embodiment of the NBC News brand on the air. Online, however, Nightly’s web presence in 2010 was centered around a micro-site on MSNBC.com, while the NBC News morning show was spun off into a large independent site, Today.com. In the branding terms employed by Tillinghast and others at MSNBC .com, Today had become a “master brand” online, while Nightly remained a “sub-brand” on the web, housed as a micro-site within the larger Blue Site. I asked Tillinghast why the hierarchical relationship between the Nightly and Today brands was inverted between the broadcast network and the web. He responded that the variation was due, first and foremost, to the difference in the amount of video content produced each day that could be repurposed for the web—the Today Show ran for four hours, while Nightly News was a half-hour broadcast. Similarly, the cable channel, for which a dedicated website was planned, produced a full day’s worth of original content each day for television, yielding a wealth of video that could be repurposed online. “Those factors right away make you either a master brand or a sub-brand,” he remarked, giving another example: “Meet the Press is a fantastic brand, but it isn’t a site. You can’t sustain a site with one hour of television a week.”5 A second factor that prevented MSNBC.com from positioning Nightly News as a master brand online (and one that I alluded to in Chapter 4) was that the content it produced was out of sync temporally with the hours when MSNBC.com garnered most of its traffic. Visitors looking for the

Mass Reach and Mass Media  220

latest news would, from the website editors’ perspective, be ill served by a warming over of the previous night’s broadcast—something especially true in a media environment whose logic was increasingly dictated by “flexible leisure” and users’ corresponding need to call up the latest news on demand (Dawson and Spigel 2008; Tryon 2013). Daytime news consumption led to other changes as well—the fact that the Nightly broadcast was now consumed by audiences who had been keeping up with news throughout the day was reflected in the program’s editorial voice in ways that further affected its fit with the website. “The stories it runs are more analysis and interpretation of the day’s events, as opposed to an attempt to tell people what happened for the very first time,” said Tillinghast. “It’s almost become like a newsmagazine.” Of course, both televised and printed newsmagazines do have websites, as do other news brands dedicated to summary and analysis. Both the Economist and CBS News’s 60 Minutes websites were doing quite well around this time, for example. But adapting the NBC Nightly News into an online brand presented another, more particular problem. As Tillinghast explained, “Calling it ‘Nightly’ doesn’t help. In some respects it’s a bit like Newsweek’s problem, being called Newsweek. That’s an oxymoron. If you get it once a week it’s not new. But, in a sense, NBC News is the brand more than Nightly. Nightly and NBC News are sort of one and the same, but the name NBC News would give you a broader platform to expand into lots of different subject areas that are beyond the scope of what Nightly does [emphasis in the original].” In other words, if Blue Site cover editors had previously worried that MSNBC TV content would pollute their objective brand image with partisan commentary, Nightly News in effect created a different branding issue when it came to exposure on the web: in some ways there was little to distinguish it from what the rest of the network was already providing to the Blue Site cover.6 The Today Show brand, by contrast, became a hub for life and style fare —cooking, fashion, parenting, and other content geared toward the show’s target audience of women ages twenty-five to thirty-four. This audience— or, rather, this array of online audiences, and therefore these types of ­content—was extremely valuable to advertisers, but had little home within the Blue Site’s more “hard news”–focused brand.7 In addition to the large volume of video produced for the Today Show’s daily four-hour broadcast, Today.com became an immediate home for all the brand-consonant con-

One Size Does Not Fit All  221

tent that the online staff could produce. Jennifer Brown, the director of Today.com, explained, You have the five million people who watch the show a day. Can you imagine? Converting 10 percent of that would be a huge website. But we’re also not going into it saying we want to duplicate that audience. And we’ve done two, three years of extensive research to date just looking at who our audience is, who our audience is not, but could be. And, basically saying, “Where are the areas where we’re allowed to play?” . . . The show is always at its core, but what we’re looking to do is build around that. It’s a weird thing to say that the show is “only four hours.” Because there [are] a lot of stories you can tell in four hours. But there [are] also a lot of stories we know fit the audience, but may not have the best visuals. Or it may be cost-prohibitive to tell [them through video]. Or maybe . . . [they don’t] fit on-air. And that’s really what we’re cultivating. Moreover, the web venture’s market research showed that the young female target demographic pursued by the show also consumed an inordinately high amount of online video. So even without all the web-original content planned for the site, showcasing Today Show video online would lead to greater revenues than the equivalent amount of video for shows with different demographics. That the show produced more video on a weekly basis than any other at NBC News only made the proposition more attractive, as did the fact that Today’s content was broadcast and flipped to the web during those daytime hours so crucial to news websites’ traffic numbers (Boczkowski 2010). In the MSNBC Digital Network’s terms, Today.com was an attempt to leverage valuable portions of the news market in which the Blue Site had not previously been playing, or in which it had been playing only to a limited extent. Putting this all back into our own terms, the expansion of the Today brand online was an attempt to create a distribution channel that appealed as a resource to an economically desirable set of users. And importantly, one way it went about this in an online context was by acknowledging the fragmentation of this user base into groups and communities that were pursuing different interests and tastes, and responding to this fragmentation by producing online sections dedicated to parenting, dining out, cooking, technology, frugal living, and so forth. Branding in this case was a resource in the ongoing pursuit of new and larger (in aggregate) online audiences.

12//one size does fit all engineering flexible architectures

One of the emerging trends at MSNBC.com during my fieldwork was a resurgence in the use of blogs across the site. Whereas the launch of Newsvine’s new blogging tool—and hence of new blogs—had been stalled for months by the Blue Site redesign, by the time I arrived in Seattle in September 2010, Newsvine CEO Mike Davidson was reporting that the site’s editors were “launching new blogs every week now.” Part of this had to do simply with the tool’s availability. Once the software and basic blog templates were finally ready, it became relatively simple to set up a blog for whomever requested one. But the increasing prevalence of blogs across MSNBC.com was also a reflection of the company’s market strategy of offering news products for diverse interests and uses. Blogs, as we’ve seen, were in effect small, easy-tocreate brands that could be targeted to different interests in a fragmented, high-choice market, placing MSNBC.com content within reach of diverse audiences and thus opening the door for content to be further propagated across the web. Whereas previously the publishing paradigm had been to post myriad individual news stories that people might read on a one-off basis, the aim of blogs was to create loyal followings and communities of users who would provide MSNBC.com’s constellation of sites with return traffic. To accomplish this aim meant creating and foregrounding topical blogs, as well as promoting individual columnists. Under this expansion strategy, the site’s longest running blogs, such as Alan Boyle’s Cosmic Log on space and astronomy news and Bob Sullivan’s consumer issues blog Red 222

One Size Does Fit All  223

Tape Chronicles, were held up by MSNBC.com staff as exemplars to be replicated and expanded on. Blogs and the columnist personalities behind them provided, in Tillinghast’s words, “franchises you can build upon that become a natural pillar or center for a community to form.” In other words, beneath the large-scale branding pushes for massive new sites like Today.com, various producers and section editors at MSNBC .com were using Newsvine’s tools to create numerous products and distribution channels targeted at niche audiences, who would “share them out” into the different groups of friends, online communities, and epistemic cultures in which they participated. At the same time that MSNBC.com was engaged in this epistemic engineering, launching more brands aimed at more niches, these projects were being supported by increasingly generic software tools. And neither blogging nor blogging software was unique in this regard. In fact, as we’ll see, system builders at MSNBC.com responded to the need for increasingly divergent and niche editorial products by creating a series of ever more generic tools for online publishing. NEWSVINE’S M3 PLATFORM

In Chapter 10, I briefly mentioned Newsvine’s third major architectural overhaul, the M3 platform. M3 was designed to fully transition Newsvine’s platform from the software underpinnings of a specific site, Newsvine .com, to a web application that—while hosted on Newsvine’s servers— could power a wide variety of websites spread across numerous web domains, each bearing little cosmetic resemblance to one another. This was accomplished through the developers’ use of successive class extensions, dubbed “chromes.” “Class,” in this context, is a term from object-oriented programming. Much as in real life, objects in programming are tools with a number of uses defined by their designers. Analogous to the way a “real world” tool like a hammer is designed with discrete uses in mind—driving nails and pulling them out—classes, too, perform particular tasks, manipulating data in useful ways determined in advance by the programmer. A simple example of a class object would be a bit of code that, when given a set of data points, can present them variously as a list, a data table, or an outline, depending on the request of the programmer or designer. Classes are one of the key tools that allow developers and designers to create webpages that display data in a broad variety of forms and contexts.

Mass Reach and Mass Media  224

A “class extension” in programming is a method of modifying a class for use in a specific context—analogous perhaps to swapping the head on a socket wrench, or adding snow tires to a car to deal with icy conditions. In the less corporeal world of scripting and software, however, the original, unmodified class remains available for use in other contexts and far more extensive modifications can be made without as much penalty of effort— it’s more like transforming a car into a boat while keeping the engine and passenger compartment. The structure of Newsvine-powered webpages was controlled by classes. “Chromes” in the context of Newsvine were a series of sequential class extensions that could be added to a website to successively refine its appearance and functionality. Returning to our automotive example, say we were talking about building a car instead of a website using the concept of chromes. We would start by building the basic, generic components needed for every car—an engine, steering wheel, driver’s seat, wheels, brakes, and so on. Once these were in place, we could add an initial chrome specifying the type of car— an SUV, crossover, sedan, or subcompact, say. If we’ve chosen the last option, the class extension(s) included in this chrome will make the necessary additions and modifications to our previous design to give us a generic subcompact car. We could then add another chrome on top of this one to specify the sort of subcompact we wanted—a Toyota Yaris or Hyundai Accent, for example. Another chrome layer might then be added to give us a color, and another to install the sports package, and so on. Moreover, because chromes are layered, if we decide we want an Accent instead of a Yaris, or an SUV instead of a subcompact, we can always strip away a few layers back to the chrome where this decision was made and add other chrome layers to build a different car without starting from scratch. Finally, because this process is nondestructive—the various combinations of objects and modifications we create remain available to us—we can simultaneously build all sorts of “cars” off the same codebase. Once the “generic subcompact” chrome layer is in place, we can use it as a starting point to design a bunch of different subcompacts. Or we can start with the Yaris layer and crank out versions of the car in dozens of colors. The same principle goes for SUVs, crossovers, and sedans in our example. And if we decide one day that we want a totally new kind of vehicle—a semi truck, say—we can return to the generic design with which we began and create a new series of chromes for semi trucks without disrupting the workings of the vehicles we’ve already assembled.

One Size Does Fit All  225

The various blog templates used at MSNBC.com, named for their creators—Ashley Blogs, Mike Blogs, and Martin Blogs—looked quite different because they were built on different chromes (Figure 12.1). At the same time, there were many different Martin Blogs across MSNBC.com that bore far greater resemblance to one another than to a Mike blog, but were still different from each other. These various Martin Blogs were also built on different class extensions layered on top of the “generic Martin Blog” chrome. While other technologies, especially CSS, were also used to alter the cosmetic look of the different blogs, chromes gave the different blogs a whole other level of flexibility, so much so that they allowed Newsvine’s basic codebase to be used as the basis for sites that weren’t blogs or Newsvine columns at all. As Mark Budos, Newsvine’s chief technical officer recalled, chromes were introduced to the site’s previous “M2” architecture to accommodate various development requests from MSNBC.com. As the scope and frequency of these requests increased, chromes ultimately became the basis for the new M3 architecture. M2 was all based on Newsvine requirements. But the chrome settings that we put into M2 were obviously for MSNBC because they had all of these different brands they wanted to support. They wanted a “Nightly News Discussion Club.” They wanted an “American Express ‘Your Business’” video site. They wanted an “Elkhart Project” blog, which was a completely different design than anything we’d done before. So they started to have these different looks, . . . all these completely different brands that were using the same architecture. To give an example, in 2010 MSNBC.com purchased control of a popular Twitter account, @BreakingNews, which tweeted links to (what else?) breaking news items. MSNBC.com treated the development of Breaking News as an internal startup project and assigned a small team of three developers to turn it into a brand aimed at capturing the market for what the company referred to as “the moment of discovery” of news. Because MSNBC.com was taking over and expanding an existing Twitter feed, rather than building an editorial product that could launch on a date of their choosing, it needed to spin up a corresponding website in a hurry while the team assigned to Breaking News decided on a more permanent web presence. This task fell to Newsvine, which was able to use the M3 architecture to launch an initial site within days. M3, then, represented a dramatic reworking of Newsvine’s architec-

Figure 12.1. Blog designs. The Last Word and the Ed Show blogs were built on top of the “Mike Blog” chrome, while Today Moms and First Read both used the “Martin Blog” chrome as a starting point. The Maddow Blog (Figure 10.2) was the first blog to be built on the “Mike Blog” chrome. NBCUniversal.

One Size Does Fit All  227

ture in response to the demands placed on the company by MSNBC.com. Through it, Newsvine’s platform went from being highly specific to the situated social and technical demands of a news-themed social network (described in Chapter 7) to a launchpad for a wide variety of online properties, ranging from blogs, to standalone sites built by MSNBC.com for marquee advertisers, to websites for entirely new brands and companies within the MSNBC Digital Network. SKYPAD

Another example of an effort by MSNBC.com to increase its architectural flexibility was its development of SkyPad. From Chapter 9, you’ll recall that the launch of the Maddow Blog was delayed by an extensive redesign of the Blue Site’s story pages, which were not only receiving an updated look, but also being built on top of a new software architecture. Ironically, this architecture upgrade, which involved a new system called SkyPad, was in part intended to help speed up the site’s years-long development cycles. Newsvine’s director of technology, Josh Yockey, said of MSNBC.com, [Around 2005] they had a lot of problems with stability and the server just not being up. And so they made an organizational focus on ensuring that there was no chance that when you went to MSNBC.com there wasn’t a fast page showing you news. And to that monomaniacal goal they ended up sacrificing a lot of agility in their development.1 So they were very careful about rolling out features and very careful about developing them and very careful about testing everything through. The upshot of which is that everything was really stable, but if somebody went in and said, “Hey, we want to change our blog platform,” it would have been two and a half years to get something turned around. To make development more flexible, and hopefully much faster, MSNBC.com’s developers added what they referred to as a “compositional layer” to MSNBC.com’s publishing platform in the form of SkyPad. CEO Charles Tillinghast explained SkyPad as a virtual “punchdown block,” after the old routing systems used with phone equipment. A punchdown block allowed its operator to plug a collection of wires into either end and make connections at will between any sender and any receiver. In the same way, where the Blue Site and other sites in the MSNBC Digital Network had once consisted of single web interfaces connected to single databases, SkyPad was intended to ultimately allow any site in the MSNBC Digital Net-

Mass Reach and Mass Media  228

work to access data from any other, and conversely to allow the company to spin up databases for information not tied exclusively to any one site. “The whole point of all this is that you can replace the front end and back end components independently,” said Tillinghast. “So it’s not a monolithic system. These are all components. And that way you could, say, upgrade the editorial UI to keep it all fresh and webby, even though you don’t change anything on the back. Or you could add a new database on the back so that you could do more queries for stories.” As an example, he cited an interactive feature that MSNBC.com had assembled after the collapse of a major bridge in Minneapolis in 2007. Following the disaster, developers at MSNBC.com had fed bridge repair data from across the whole country into an online database and created an inter­active map that allowed users to explore the repair status of bridges in their area. At the time, this feature story had required a targeted effort on the part of developers. By contrast, the new MSNBC.com architecture was intended to allow other similarly unorthodox data sets to be assembled into standalone databases and “plugged in” to SkyPad, where they could be queried by editorial staff and used to generate data-driven stories or interactive features with little or no further assistance from the site’s developers. SkyPad was thus intended to have several important effects. First, it would speed up development by increasing the modularity of the sites employing it. User interfaces and databases would become isolated components to be updated and swapped out like so many interchangeable parts, thus limiting the scope of the work that had to be done to refresh any one piece of the site.2 Second, it would increase the diversity of data sources that developers and, ideally, editorial staff could draw from—as in the bridge collapse example, rather than trying to fit some interesting new data set into a Procrustean database tied to a particular site’s content management software, a new database could be created for that content and “plugged in” to SkyPad. Third, and finally, SkyPad would effectively serve as a massive application programming interface, or API, allowing developers and designers to easily call up and flexibly present data from any source within the MSNBC Digital Network, which in turn meant they could ideally assemble new page templates and whole new websites on the fly. As mentioned in Chapter 4, one of the first products to come out of SkyPad was the new MSNBC.com story page template, which allowed editors to mash up, arrange, and continually update any combination of text, video, photos, illustrations, user comments—and ultimately other media

One Size Does Fit All  229

as well—from across the MSNBC Digital Network. An important advantage was that the page for any news story on the site could be continually updated over the course of a breaking news event to allow for the frequent addition of new information in a wide variety of media formats. Being able to update a single story page with a broad variety of different media, which could be continually linked to and revisited at the same URL, was seen as a dramatic improvement over the previous CMS’s workflow, which would generally involve publishing text, videos, and story updates all as their own individual items on the site, or combining them only through inflexible templates that did not allow editors to place different media in ways that best told the story they intended. One MSNBC.com staffer in New York described the benefits of the new page templates as “being able to tell the story as it should be told versus sticking to a regulated template. . . . Stories [that involve multiple types of media] can evolve naturally in the same space versus having to Frankenstein it all together.” Much as the M3 architecture did for Newsvine’s platform, SkyPad was intended to turn the MSNBC.com’s own CMS into a generic platform, one that could present a wider array of content in a greater diversity of forms to serve the growing number of editorial strategies and interests in the MSNBC Digital Network. GENERALITY IN THE SERVICE OF SPECIFICITY

Each of these changes made to the architectures of Newsvine and MSNBC .com were substantial, labor intensive, and—most interestingly—aimed at making the platforms useful in a broad array of highly specific, even unforeseeable, editorial circumstances. If the MSNBC Digital Network had seen a brief window of opportunity to acquire or launch a new brand, it would, with these new architectures, have been able to rapidly spin up a corresponding site, whatever its unique editorial requirements. If the über-specific editorial need to mash up local exit polls with bridge repair data ever arose, that too would have been possible in theory under the new system. These architectural changes, as we might expect, dovetailed well with MSNBC.com’s publicly espoused “one size does not fit all” market strategy of creating editorial products to suit a broad variety of needs. More surprising, in some ways, is that this proliferation of products for niche audiences and their highly specific needs was being underwritten by the construction of ever more generic and multi-purpose publishing tools. This was true of M3 and SkyPad, as we’ve seen, and also of some of

Mass Reach and Mass Media  230

the other MSNBC.com technical artifacts that they supported and interacted with, including the Blue Site story pages and the company’s remarkably multipurposed video player. This flexibility was accomplished through a combination of strategies, including the construction of artifacts built to be inherently flexible—for example, story pages that would accept nearly any kind of content—and the assembly of artifacts designed to be rapidly repurposed, such as chrome-powered sites built by Newsvine that could quickly be stripped of one style and given another.3

13//mass reach with and without mass media

It might be tempting to slot the various epistemic concerns surrounding branding into one distinct analytical category, while putting the architectural interoperability of MSNBC.com’s software systems into another. But understanding the role of heterogeneous engineering means being wary of such hard distinctions. Rather, the proliferation of niche brands and editorial products, combined with increased architectural flexibility, should be seen for what it is, or at the very least what it attempts: mass reach without mass mediation. In other words, these two developments were part of a combined strategy for enrolling a diverse array of otherwise recalcitrant audiences, who themselves make strategic use of heterogeneous resources, including media, in the “‘going concern’ of their daily lives” (Law 1987, p. 133). The architectural components of this strategy were often constructed to include affordances for the rapid repurposing not just of architecture, but also of content. MSNBC.com’s video player, for instance, grouped clips by show, playlist, topic, and search query to meet the needs both of web producers and of users with highly distinct provincial requirements. Similarly, Newsvine’s blogging tool allowed editors publishing an article to simultaneously cross-post it to any MSNBC.com blog to which it might be relevant. A writer for Technolog, the site’s tech blog, for instance, could simultaneously publish a story about “parental controls for Facebook” to Technolog and the Today Moms blog, or a post about a new XBox product to Technolog and the site’s gaming blog, In-Game, simply by checking off destinations on a list. This meant that a columnist whose primary focus was maintaining a “personal brand” or contributing to a particular section 231

Mass Reach and Mass Media  232

of the site could nonetheless easily send content anywhere within the larger set of Newsvine-powered properties managed by MSNBC.com. But these architectures were simply a technical piece of a heterogeneous system that lashed together myriad components, only some of which were what we’d traditionally think of as technology or software. And while architecture could be made more flexible, other components wanted to hold on to their shape. As Newsvine’s director of technology Josh Yockey put it, “The community’s not flexible. You can’t just plug it into new content and say, ‘Go commune!’” Just as we discovered some of the ways in which niche branding created recalcitrance for system builders, then, it is instructive to look at the recalcitrant aspects of flexible architectures. THE COMMENT PROBLEM

Both researchers and practitioners have noted that one of the greatest challenges facing legacy news organizations online is providing forums for user comments and interaction that stay civil and productive (Davenport 2010; Hermida and Thurman 2008; Seelye 2007; Singer 2004). As Jane Singer (2010) put it, “In general, new relationships between users and journalists seemed valuable as an abstract concept but often proved difficult in real life. The open discourse invited by a ‘comment is free’ philosophy sounds great in theory—all that good democracy-in-action stuff. But the reality was rougher, and many journalists expressed dismay over the disturbingly confrontational nature of user contributions to the conversation” (p. 280). ABC News, for instance, had sufficient trouble with rowdy comments that it at one time found it necessary to maintain a full-time moderation staff to review every comment posted to its site. Meanwhile, as mentioned in Chapter 8, the developers at CBS Interactive went through the immense trouble of building a sophisticated commenting system for CBSNews.com, only to temporarily shut it down during the 2008 election when users became too confrontational and their commentary too offensive. Until the integration of Newsvine’s blogging tool into the Blue Site in 2010, all comments on MSNBC.com blogs were subject to moderation before appearing anywhere on the site. Staff members at Newsvine were universally opposed to this “pre-moderation” strategy, however, calling it “the quickest way to kill a conversation.” In their view, pre-moderation and the comment systems on mainstream news sites more generally were designed around a misguided idea: that users were there to leave feedback on a story, rather than engage with one another. In the Newsvine offices, this design

Mass Reach with and without Mass Media  233

strategy was derogatorily referred to as the “urinal style” of comment system design. Newsvine’s director of technology Josh Yockey explained, “The idea that there can be these running conversations moving around the site and users engaging with each other on topics that the editors aren’t controlling is difficult for them to design around. And so they favor what we call the ‘urinal style’ of commenting, where everybody goes in and leaves their contribution and then leaves. And then we flush it out and then the next guy comes in. As opposed to a conversation.” For philosophical reasons—and reasons of convenience—the Newsvine developers preferred to use their existing system of comment threads on MSNBC.com blogs. This was not a universally popular idea among the editors and web producers at MSNBC.com, however, and in 2009 when I conducted my first interviews, Newsvine was preparing to build a pre-­ moderation option into its blogging tool for the Blue Site. At the launch of the blogs in 2010, however, this hadn’t been done. Newsvine CEO Mike Davidson explained that while some editorial groups within MSNBC.com had initially made the ability to pre-moderate user comments a precondition for adopting the Newsvine software, the developers ultimately decided to try releasing their tools without making these changes. “We just didn’t build it, and people moved onto the new platform,” said Davidson. In his telling, editors were ultimately appeased by the fact that they didn’t “have huge backlogs of unapproved comments that somebody’s got to get to. It seems like a much lower-maintenance system for everybody.” This account, of course, is from the vantage point of Newsvine and paints the benefits of the new moderation system as ultimately overcoming the recalcitrance thrown up by resistant editorial groups. In practice, resistance among editors to the abandonment of pre-moderation did not come strictly from the conservative editorial impulse suggested by Yockey. This is not to say that MSNBC.com’s editorial staff was devoid of the conservative “gatekeeping mentality” toward comments that Alfred Hermida and Neil Thurman (2008) have described, but rather to point out that there were other reasons for preserving pre-moderation. Some of the pushback, for instance, was very much in line with Christian Davenport’s (2010) observations in the Washington Post that vitriolic web comments could at times scare off sources. In any event, in 2010 Newsvine’s community moderation system went live on MSNBC.com and subsequently spread across the site with the launch and conversion of numerous blogs to the Newsvine platform, as well as

Mass Reach and Mass Media  234

the inclusion of featured Newsvine comments on the Blue Site’s new story pages. When this happened, things were not uniformly neat and tidy from anyone’s perspective. Web producer Ian Sager participated in an MSNBC .com task force that explored issues surrounding user participation on the site. He recalled that “a lot of people had concerns that interactions on Newsvine were getting really rough and a little dark.” In particular, stories from the site’s health editors were drawing worrisome comments: [Sources] who were sharing very personal stories about weight loss or trials and tribulations with drugs, or really intimate stories about their life—where they were risking a lot to speak out—were being trashed by people. And for no reason.1 . . . [MSNBC.com staff] didn’t know if this was a problem with the Newsvine community. They didn’t know if we were moderating comments correctly. Maybe an editor needed to jump in when things got out of hand and say, “Hey, I’m going to ban this comment. I’m going to delete it because this is not cool. And anyone else who tries behavior like this, from now on we’re deleting [it].” Basically showing the rules of the game. The conversation at MSNBC.com that Sager relates here, from the vantage point of the site’s editorial staff, is fascinating given what we know about how the original Newsvine community was cultivated and the changes it underwent after the social network’s acquisition. Newsvine had built a technology that relied as much or more on carefully fostered social norms as on its technical platform, but when it was integrated into the Blue Site, that technical platform was effectively plugged into an entirely different social environment—the raucous, vitriolic melee so frequently described by journalists and scholars alike as occurring on legacy media sites. People who came to leave thoughtful comments or interact with one another were mixed in with others who, as Femia put it, “are screaming at their TV and don’t realize there’s a human on the other end.” And while Newsvine had a full-time human moderator on duty, his capacity to successfully carry out his job was highly dependent on the ability of the community to accurately locate and report abusive comments. Unfortunately, as developer Josh Yockey recounted in Chapter 8, the community that could provide this support, the one that understood and abided by the norms on which the software was predicated, increasingly avoided the Blue Site threads altogether. “Our existing community members tend to pull out of those conversations, because they’re overwhelming. They don’t report bad comments or anything. They just stay out.” Absent its important

Mass Reach with and without Mass Media  235

social components, the Newsvine comment platform continued to operate from a technical perspective, but ceased to function at the high level that was originally intended. Put another way, it malfunctioned because it was, in fact, a sociotechnical system that had been stripped of the social, as opposed to the technical, components that allowed it to function correctly. It’s important to note here that, according to the individuals with whom I spoke, user behavior differed widely across the various Blue Site blogs and pages. Several properties, including the Maddow Blog, Alan Boyle’s long-running Cosmic Log astronomy blog, and the First Read politics blog, benefited from extensive moderation and community-building efforts by their respective editorial staffs. And this is precisely what’s most interesting about Sager’s earlier comment: a social component—editors stepping in to moderate comment threads, give slaps on the wrist, and coach users on proper behavior—was just the component that Blue Site staff were considering adding back in at the time of my fieldwork. Meanwhile, however, the “firehose” of new Newsvine registrants from the Blue Site continued to increase as Newsvine blogs and comment threads began to permeate MSNBC.com. This intensified many of the pressures on Newsvine.com’s social structure. The “Green Site” still functioned reasonably well as a social news site, but it was becoming gradually harder and harder to moderate. In response, Newsvine continued to lash together resources to preserve its existing “community implementation,” as Yockey termed the social and technical structure of the Newsvine social network. In mid-2010, Yockey himself was working to design a Bayesian spam filter aimed not only at helping to flag inappropriate replies in forums where Newsvine community members feared to tread, but also at highlighting interesting and useful comments—a way of both steering readers to helpful content and turning a spotlight on users who offered helpful, appropriate contributions. This was a creative fix, to be sure, but also somewhat narrow in scope. Around the same time, Newsvine’s CEO, Mike Davidson, was proposing an even larger, long-term change intended to alter the way Newsvine comments and communities operated in far more fundamental ways, as well as the manner in which users approached and shared content from MSNBC TV and NBC News properties. NEWSVINE 3.0

One of the fundamental features of Newsvine’s software was the ability to create groups—columns to which multiple users could contribute. Groups, which could be created around any subject, were not only a way

Mass Reach and Mass Media  236

for users with similar interests to congregate on the Green Site; they also became the starting point for building the company’s new blogging tool, since MSNBC.com blogs were effectively columns shared by multiple authors. But user groups—at least on the social network Green Site—also had another function. When an article was published or shared to a group column, an additional comment thread, housed in a tab alongside the main public comment thread, was created for group members to converse solely with one another about it. Davidson’s proposed major change in Newsvine’s software architecture revolved around the way such groups and their discussions were structured. As he summarized the situation, Newsvine’s existing infrastructure was built to address a different set of challenges than those currently faced by the social news platform. Prior to the company’s acquisition by the MSNBC Digital Network, the problem to be solved had been how to build a productive community from scratch. And Newsvine had arguably been successful at this. “We had this nice, big budding community of very thoughtful people,” said Davidson. But, in his words, the primary challenge MSNBC.com invited was quite the reverse. “What we’ve done at MSNBC has been the opposite. We’ve taken this gigantic number of people—40 million people—and tried to inject community into that.” While successful from a metrics standpoint, as MSNBC.com users flocked to the soapbox provided by Newsvine, the resulting conversations could often be thoughtless or inappropriate. Davidson argued that “the next revolution in online news” would come in elevating the discourse on large media sites, which were rich in traffic, but notoriously poor in civility: “I’ve come to the conclusion that you just can’t have communities above a certain amount of people. If you’re going to take that as a fact, then you can either say ‘Okay, we’re not going to have any community on our site,’ and go back to the way things worked ten, twenty years ago, or you can say ‘We’re going to have a lot of small communities.’ That’s something nobody has done yet: . . . taken a large-scale news site and created thousands and thousands of communities of likeminded people out of their audience.” The idea, which was still in early development in September 2010, was outlined to me by Davidson in the following terms. Commenters would be divided into groups of roughly 150 users each by virtue of their similar interests or other mutual affinities (a coterie of college friends, for example).2 Just how this carving up of the Newsvine-registered user base would take place, or how it might be done gracefully, was as yet undecided. In

Mass Reach with and without Mass Media  237

any case, rather than large public comment threads, users browsing the Blue Site would instead converse about articles only within their member groups, though group discussions could be browsed and comments voted on by unaffiliated visitors. New users on MSNBC.com would be initially unaffiliated with any group, and their comments would be invisible to the public—or at the very least kept in a quarantined Greenhouse-like area of the site, visible only to visitors who dug for them. Meanwhile, Newsvine’s existing karma system would be modified to rate the quality not just of individual users’ contributions, but also those of entire groups. Groups that made especially good contributions and received excellent ratings would be elevated in status on the site, and groups that did particularly well would be afforded special privileges, foremost among which would be placement alongside highly trafficked MSNBC.com content. This selective access to public exposure would result in friendly competition among groups for higher rankings and better placement on the site, all of which could, in principle, only be attained through continued high-quality, civil discussions. What’s more, new members would only be admitted to groups by existing members. Because a person’s comments would only receive attention if they appeared as part of a highly ranked group’s discussion, a fierce competition would also develop among unaffiliated users for membership in well-rated groups. Existing groups in turn could sift through the quarantined comments of unaffiliated users looking for particularly high-quality contributions, hoping to thereby recruit promising members whose commentary would help them rise to the top of the rankings. According to the developers, the new quarantine area, in this sense, would resemble a sort of conversational sporting draft. “If you get accepted into a group that has a lot of privileges, that’s kind of like having waterfront property on MSNBC,” said Davidson. “If five of the greatest groups on the site have carte blanche privileges to comment on the Maddow Blog, on Hardball, on Countdown, on just about every property, then getting into one of those groups is a privilege. And it may take a while for you to do. If we put artificial constraints on the number of people that can be in a group, for instance, then that makes it even harder.” As with any game or set of structured incentives, there were questions about how much effort users would expend on participation in this new system and how difficult it should be to attain various objectives, such as membership in selective groups or privileged placement on the website. Erring on the side of caution would mean starting with a low barrier to

Mass Reach and Mass Media  238

entry, though the staff mused that ratcheting up the difficulty of membership or the seriousness with which users regarded the system could be done via a variety of means, including the use of a private beta period, such as that on which Newsvine’s original community was founded; a nominal fee for access to the discussion feature; or caps on the number of members allowed in a group, so that low-quality contributions could get one ejected in favor of a more promising user. An increase in the quality of comments would also ideally create more opportunities for featuring user-­ contributed material on MSNBC.com.3 Much could be said here about this proposed commenting system—on the potential for users to self-select into “echo chambers” (Sunstein 2007) devoid of opposing views; on the “gamification” of discussion through the creation of a system where groups would compete on the basis of interesting conversation (Bogost 2011; Zichermann 2011); or on the commodification of discourse in the form of “waterfront property” on MSNBC.com. Of course, similar activities were happening across the web at the time. Concerns over the potentially polarizing nature of online media consumption began not long after the Internet became widely available (Alstyne and Brynjolfsson 1996). Users on Twitter and many social media sites compare follower counts like trophies. And several sites—most notably Gawker’s media network—had already begun introducing tiered commenting systems by 2010, in which only trusted users’ comments appeared on public story pages.4 In the present context, what’s unique and fascinating about Newsvine’s proposal, a form of which was eventually enacted in 2013 by MSNBC.com’s successor (and Newsvine’s new parent), NBC News Digital, is that it heterogeneously engineers a solution to the problem of a recalcitrant user base by carving users up into myriad niche editorial cultures that can be selectively placed and promoted anywhere within the family of websites related to MSNBC and NBC News. I don’t doubt the sincerity of Newsvine’s goal to create civil discourse on a mass media site, nor do I wish to play the role of armchair strategist by evaluating its potential for success. What’s interesting is that they aimed to achieve these goals by effectively engineering the user community into groups that could be managed as a collection of brands compatible with MSNBC.com’s larger branding scheme. It’s easy to imagine editors and executives arranging Newsvine groups in much the same way they arranged sub-brands—selecting out ongoing political discussions, life and style conversations, and debates over game systems, and

Mass Reach with and without Mass Media  239

deciding on their placement and prominence on the basis of their consonance with the image of the site’s various properties. FRIENDS OF THINE, FRIENDS OF THEIRS

The new group commenting architecture and the corresponding challenge that Davidson laid out—of designing a sociotechnical system to socialize and positively direct the energies of the overwhelming numbers of commenters who frequent large legacy media sites—in some ways signaled a major shift in Newsvine’s system building and a measure of how the local teleologies of system builders can change as they enroll new and recalcitrant resources. The social media company that began by developing a prominent publishing platform for citizen journalists and everyday users was now developing products to suit the needs of corporate media giants like Microsoft and NBCUniversal, subsequently becoming one of the principal architects of the new MSNBC TV website. Within the MSNBC Digital Network, Newsvine came to be described as a “design shop” or a “white-label brand,” creating products that bore other companies’ imprimaturs. Some described it as an “ingredient brand,” akin to the manner in which small GORE-TEX labels are affixed to prominently branded North Face jackets. The Newsvine developers were becoming accustomed to these descriptors during my fieldwork. At the same time, if Newsvine’s role as an ingredient brand took some getting used to, and the focus embodied in the revised comment architecture was new, these changes nonetheless shared a surprising amount of DNA with the objectives of Newsvine’s initial system. Newsvine’s original mission of designing spaces where mainstream media and citizen media could co-exist and benefit one another was arguably still intact. After all, if the best contributions of Newsvine users rose to the top of an NBC or MSN site that garnered tens of millions of unique visitors each month, it would give previously anonymous publishers remarkable exposure. An element of this mission could certainly be said to have been accomplished. But the “new” and “old” Newsvine systems would share DNA in another important way as well—they would be built from the same codebase. Due to the poor economy and other factors, Newsvine, as explained earlier, never received great increases in staff or resources. Yockey talked about the effect this had had on the company’s products—here he is discussing the origin of MSNBC.com’s polling feature, “question boards,” but he could equally have been talking about many other features and products the team

Mass Reach and Mass Media  240

has built: “We didn’t want to get in a situation where we were maintaining a bunch of code for [Newsvine.com] and a bunch of code for MSNBC and it ended up being 2x, without changing the size of the staff. I found ways to mold [the requested] features into our existing capabilities. Then it was a lot of front-end code to make it look like they wanted it to look—a lot of Javascript and a lot of templating.” Nearly every product that Newsvine built for MSNBC.com contained code from its original social network. This worked well for Newsvine as a system builder. Improvements made for MSNBC.com could in principle be turned around to benefit users on Newsvine.com, which the team still viewed as their core product, and which still carried the citizen-media branding flag for the MSNBC Digital Network. In practice, such improvements often got rolled out slowly. The site looked much the same in 2010 as it did after its acquisition. Davidson remarked: If you’re a user of Newsvine, you feel like Newsvine employees work for you. When I use Twitter, for instance, . . . I feel like everybody who is working for Twitter . . . should be working on making the user experience better for me, the user. They should be at my service. I feel that way with every company. . . . And so if I was a Newsvine user who used the site every day, I would feel the same thing. “Okay, well I know the staff is small, but shouldn’t they be adding more features to Newsvine? Shouldn’t they be speeding it up?” . . . And we think to ourselves, “Yeah, we should be.” But we have all these other things that we also have to attend to. From the vantage point of Blue Site staff, incorporating Newsvine technologies could occasionally have unpredictable consequences. During my fieldwork, for instance, Newsvine’s new blogging tool would occasionally become a recalcitrant object for editorial staff, say when one of their stories would encounter a profanity filter intended for Newsvine.com users. And we’ve already seen how the initial commenting features of the Green Site yielded mixed results when applied to other MSNBC.com properties. The dual-use platform was nonetheless essential, from Newsvine’s perspective, if it was to keep maintaining its social media service while developing new tools for the rest of the MSNBC Digital Network. All of this meant that any new groups architecture that Newsvine developed for MSNBC.com—or, ultimately, NBC News Digital—would eventually be applied back to its social network, if not released there in-

Mass Reach with and without Mass Media  241

itially. Tyler Adams, Newsvine’s full-time moderator, as well as the other Newsvine staff, anticipated that both the Green and Blue Sites would benefit from the new system. Whatever the case may ultimately have been, News vine.com made a fascinating transition from a small community built on and around norms developed by “friends of mine and friends of theirs” to one recreated in part to help make governable the behavior and (frequent lack of) norms exhibited on one of the most heavily visited media sites in the country. OWNERSHIP AND INNOVATION: MANAGING HETERARCHY AT MSNBC.COM

Given the amount of coordination that this proliferation of brands and increased flexibility of architectures required, and the cross-cutting nature of the strategy at MSNBC.com, the developments described here might seem to defy the notion that heterogeneous distribution systems are forged provincially.5 But there are significant ways in which they do, in fact, fit this scheme. First, Nick Couldry (2008) reminds us that power asymmetries, and thus top-down management, are easily explained in actor-network friendly terms by considering all of the related, corresponding asymmetries in the resources available to management versus individual project teams. Such differences in access are historically situated outgrowths of the more traditional bureaucratic organizational styles and structures that now exist alongside newer heterarchic arrangements. The basis of this argument is one of the points on which John Law (1992) is most frequently quoted: If we want to understand the mechanics of power and organization it is important not to start out assuming whatever we wish to explain. For instance, it is a good idea not to take it for granted that there is a macrosocial system on the one hand, and bits and pieces of derivative microsocial detail on the other. If we do this we close off most of the interesting questions about the origins of power and organization. Instead, we should start with a clean slate. For instance, we might start with interaction and assume that interaction is all that there is. Then we might ask how some kinds of interactions more or less succeed in stabilizing and reproducing themselves: how it is that they overcome resistance and seem to become “macrosocial”; how is it that they seem to generate the effects such as power, fame, size, scope, or organization with which we are all familiar. (p. 380)

Mass Reach and Mass Media  242

In other words, we can recognize management at MSNBC.com as a provincial system builder in its own right, one that wielded a uniquely large set of resources assembled through corporate partnerships, contractual agreements, and so forth. A history of the firm is not warranted here, but even without it we can appreciate the historically contingent, situated, and heterogeneous resources that gave MSNBC.com management access to a good deal of power as a system builder, without resorting to thinking of systems in terms of firmly structured hierarchies. Nor was management monolithic in any sense—I’ve given examples throughout of executives making different arguments and pursuing different agendas. We should not discount managerial activities as leaning exclusively on traditional notions of hierarchy or involving less in the way of heterogeneous engineering than other system building activities. Jennifer Brown, the director of Today.com, for instance, recounted her role in terms highly consonant with the notion of heterogeneous engineering: “I’m very, very aware of what all the different departments are doing, and how to connect the dots between ‘This person is asking for this,’ ‘This group is working on this,’ ‘This sales team is selling this.’ It’s hard to do. Which is why that skill has evolved into this outwardly facing title, ‘director of strategy and development.’” Brown went on to explain how keeping continual tabs on different people, departments, and provinces within MSNBC.com and NBC News meant initiating and keeping an eye on many different forms of communication, from instant messaging to email to Facebook to in-person coffee breaks. Her account, in other words, puts on display the heterogeneity of tools, motives, and concerns that have to be aligned to unfold the local teleology with which she was concerned. A second reason why large-scale coordination is not necessarily at odds with provincialism is that provincialism itself was a managerial strategy of the MSNBC.com venture. Insofar as top-down authority existed, it was often used to empower provincial system builders. Companies that MSNBC.com acquired, like EveryBlock and Newsvine, maintained their own internal staff structures. With Breaking News, the social media company that MSNBC.com launched internally, the small team behind it was given independent authority over its product outside the normal reporting structure of MSNBC.com. And major initiatives within the company, like the creation of the video player or the spinoff of Today.com, were often handled not by individual units within the company’s official hierarchy,

Mass Reach with and without Mass Media  243

but by the sort of diversely composed project teams described by Mark Deuze (2007), Monique Girard and David Stark (2002), Katherine Kellogg et al. (2006), and other scholars of “heterarchic” and “postbureaucratic” firms. Tillinghast explained the rise of project-based management alongside MSNBC.com’s more traditional managerial structure in terms that tie together the proliferation of brands and the flexibility of platforms: You have to do a hybrid. Because what you’re trying to achieve is agility, but you need scale. . . . [Development] teams are very agile, but they have a hard time scaling. In a couple ways. One, they don’t have the personnel to build out deep systems. The other one is that they’re not necessarily thinking about the long term. They’re really more tactical in their orientation. So you have to have other people who worry about that stuff. And so part of what we’re looking at is trying to create certain parameters for them to follow in the early stages [of developing new projects] so they don’t create something that’s just completely incompatible with the rest of our company. So, there’s certain tools or certain languages we don’t necessarily want them to build on. Because it would be hard to reintegrate them. But then as they mature, you want to be able to cascade [these projects] down to the core organization and free up [these teams of developers] to keep doing more innovative stuff. Basically, at some point the cost of ownership overwhelms the ability to innovate. And so you want to have your core organization be excellent at ownership. And another piece of it be good at innovation. The other way we’re getting at this is by breaking out our brands. We’re going to create different brand teams or product teams. And they’ll have a leader and be more self-contained as well. So we will be creating more teams to improve speed and agility. And then they’ll be supported by certain [operation and maintenance] functions that remain core. We’ve built a platform [SkyPad] over the last four years that scales very well and is very flexible to build new sites on top of. The idea is that you build up sites that look completely different, but they all sit on top of the common platform. They’re all served by the same servers. You can keep your operations cost very low, but you’re capable of handling any volume of traffic that comes in. In other words, the coordination involved in the rise of niche brands and flexible platforms was fueling provincialism, not undermining it. And

Mass Reach and Mass Media  244

as Couldry (2008) suggests, even the groups that were “good at ownership” and fell within the more traditional hierarchical structure of the company could be viewed as heterogeneous engineers; if they had less authority, this could be seen as an asymmetry of resources. Even the growth of flexible platforms, as we’ve seen, was driven by provincial motives aimed at enrolling more users while saving labor and resources. And as contemporary organizational scholars have noted, this proliferation of provincial groups and project teams is a strategy that many large media companies have ­pursued. Finally, not all the groups engaged in system building were defined from the top down. New distribution channels like blogs and Twitter accounts could be initiated by editorial groups rather than by managerial decree, and so be understood as a provincial act of system building, aimed at enrolling and taking advantage of available resources. Indeed, if we describe sub-brands as “niche” editorial products, we’ve already suggested that there were niches to be filled, which opens the door to a wide range of opportunistic and provincial motives. Throughout this portion of the book I have foregrounded a notion that, at one point, I called “mass reach without mass mediation” to describe the increasingly common relationship between niche brands and editorial cultures on the one hand, and developers building highly flexible software architectures on the other—two aspects of an arrangement aimed at enrolling as many users, system builders, and users-as-system-builders as possible. But it’s important to qualify this observation. After all, this is a book ultimately bent on examining the systems involved in putting television news online. The framework of heterogeneous engineering has taught us that, to do so, we have to step outside a priori categories to look at actors and overlapping systems that at first glance may appear to have little to do with television or with news. But while we may well one day receive nearly all of our television content over the Internet, for the immediate future we are still fundamentally talking about the online presence of a mass medium. Web producer Laura Conaway, for instance, was quick to point out that while getting Maddow Blog content to circulate still involved carefully building relationships with users and other bloggers, the television program by itself gave the blog a great deal of exposure, not just by pointing to the URL, but by actually putting blog content on the air. Especially now that MSNBC TV and NBC News have moved to their own websites, we can expect to see them—in

Mass Reach with and without Mass Media  245

true heterogeneous engineering fashion—press the advantage given them by broadcast and its ability to aggregate attention in promoting their web presence, from featuring blog content on the air as in Conaway’s example, to suggesting Twitter hashtags in the chyron graphics that appear at the bottom of the TV screen during network programming (Joyella 2011a).6 The division of television content into brands aimed at different audiences long predates the Internet, and even cable (Turow 2004). Before there was an MSNBC TV channel for news and commentary with a progressive political bent, there were targeted editorial products on the air, from network children’s programming to the Black Entertainment Television cable network. What’s different now is that online media companies, including those that distribute TV news, depend on their audiences—or as Jay Rosen (2006) famously put it, “the people formerly known as the audience”— in fundamentally new ways. Media companies can no longer conceive of audiences merely as consumers. Rather, as we’ve seen, they’ve become a fundamental part of the distribution path forged by system builders at MSNBC.com, MSNBC TV, and elsewhere. We’ve come to view users as heterogeneous engineers whose decisions about when to consume, when to pass along, and when to incorporate media texts into the “ongoing concern of their daily lives” (Law 1987, p. 133) are based on whether or how they can serve as resources in unfolding local teleologies—whether that means the maintenance of friendships through forwarding links of mutual interest or writing scathing reviews of television programming on one’s personal blog with the help of an embeddable video player. The realities of this distribution environment put pressure on system builders at MSNBC that were reflected in their media products, which were increasingly divvied up to fit online brands to suit every interest and need, while technological platforms were made increasingly flexible to under­ write this trend. And once again, organizational boundaries don’t align neatly with the systems we’ve encountered. Multipurpose third-party tools like Twitter and Vimeo were in play, and the public-facing interfaces of the company’s own products, like the online video player, were similarly made ever more flexible, to allow users to further slice, dice, and adapt media products to suit their provincial projects. From a societal standpoint we might worry, as did postwar critics of American mass media like Theodor Adorno and Max Horkheimer (1991), that these products are still not diverse enough. Conversely we might fret

Mass Reach and Mass Media  246

that the niche editorial strategies surrounding market segmentation will disrupt our political culture (Sunstein 2007) or that the logic of market segmentation is fundamentally flawed or civically damaging in other ways (Prior 2007; Turow 2004). We might also take a critical stance toward the manner in which system builders who host our public discourse by developing online social media and commenting platforms go about their business (Gillespie 2010, 2014a, 2014b). The point is that we need to continue to explore the connections between production and consumption that occur through the development of distribution strategies, and fill in the gaps that currently exist in the study of media sociology and particularly the sociology of news, fields that have tended to skip from media production directly to audience effects, too often ignoring the ways in which information reaches audiences. Doing so will begin to illustrate some of the common mechanisms through which heterogeneous engineers online seek mass influence and mass reach with and without mass media, and will advance our debates about the effects of media distribution on public discourse and civic engagement.

14//conclusion

Much scholarship on distribution and sociotechnical systems has focused on what Thomas Hughes called “modern” systems, like power grids and railways, which were essentially built to last and to remain stable over time. In Rescuing Prometheus, one of Hughes’s later works, he pointed to a future in which “postmodern” systems would become the norm—in which project-based economies would replace aspects of traditional hierarchical management in the creation and maintenance of sociotechnical systems, and “discontinuous change [would] become the expectation of professionals.” Organizational scholars have done much to document the emergence of these trends, calling attention to the evolving organizational landscape with adjectives like “postbureaucratic” and “heterarchic.” The MSNBC cases detailed in this volume are excellent examples of these developments. But while the concepts and observations I have offered in response have been arrived at with an eye toward the sorts of new and emerging organizational forms and distribution networks that are often said to be hallmarks of the Internet, many of these observations hold true for historical media distribution systems as well (Turner 2006). Richard John (1995), for example, in documenting the origins of the U.S. Postal Service, began his inquiry by asking not how news content reached the screens of mobile devices, but how it came to the post offices frequented by pioneers of the Michigan back woods in the 1830s. He similarly uncovered a system that was imagined quite differently by its various p ­ articipants— indeed one forged by and at times helping to sustain very different cultures, both internally and on the part of its users and patrons. 247

Mass Reach and Mass Media  248

As Alisa Perren (2013) has said, the study of distribution, while not entirely ignored by scholars of media (as the postal example illustrates), has been spread conceptually across a diverse array of fields and disciplines, so that deriving a coherent framework for understanding and critiquing systems of distribution requires a good deal of synthesis work. In this concluding chapter, I will attempt to summarize the set of conceptual tools developed throughout this manuscript and to highlight their connection and utility to other literatures and cases that might collectively be thought of as an emerging field of “distribution studies.” This task will carry through to the Epilogue, wherein I utilize these ideas to describe the developments at MSNBC since my 2010 fieldwork. TRACING DISTRIBUTION PATHWAYS

I began the book with the supposition that many significant relationships shaping both our access to content and the nature of the systems delivering it can be discovered by tracing the path of media as it makes its way to us. Depending on the analysis, one can trace the path of information through the physical infrastructures that deliver it to the end user or— as has primarily been the case in my analysis—trace the various people, groups, companies, and software applications that manipulate, repackage, and recontextualize media products as they move along this path from producer to user.1 Following and questioning the path of MSNBC television content from the time it aired to the time it reached users online uncovered a range of historically situated relationships, intra- and inter-organizational concerns, and economic pressures that ultimately shaped that path. For example, investigating the time delay in placing televised news content online revealed tensions between the online venture and cable providers over how to monetize audiences, as well as the economic pressures created by users whose sharing—and occasionally bootlegging—activities could dramatically impact the online revenue generated by a particular clip. As we traced the process whereby video was changed from its original televised format into a series of objects available online, we saw how historically situated arrangements of resources at Microsoft and NBC Universal led to a far-flung arrangement—a path in which video was broadcast on one coast and re-recorded on the other. In considering how those videos were placed in different locations across the MSN and MSNBC Digital networks, we saw how editorial differences between on-air and online staffs

Conclusion 249

became inscribed in the various pages and URLs through which content was made visible to audiences—an MSNBC TV op-ed segment would be welcome within the bounds of its own “micro-site,” for example, but not on the cover of MSNBC.com. The principle of tracing can likewise be applied to other distribution cases. For example, asking why in 2009 and 2010 network television programming suddenly disappeared from Internet-connected set-top boxes like Boxee and Google TV reveals the substantial tensions between mainstream content providers and device manufacturers (Braun 2013). Curiosity about why Hollywood DVDs were for a time unplayable on Linux computers (Coleman 2013; Postigo 2012) or why certain products from iTunes are only playable on five devices at a time (Gillespie 2007) can be an entry point into monumental questions surrounding the nature of copyright law and the limits of free expression. And questions about how nearly everyone in some university dormitories ends up with access to Netflix programming via a small handful of shared passwords (Tryon and Dawson 2013) or why unauthorized access to Hollywood fare online differs significantly across schools and nationalities (Strover and Moner 2013) can lead to revealing conversations concerning how college students from different national and socioeconomic backgrounds value commercial media ­products. Often what we encounter in existing studies are “moments of tracing,” wherein the (un)availability of content in a particular context—often a controversial one—is interrogated to reveal significant and sociologically interesting relationships. In tracing the path of content from end to end in this book, I have attempted to examine not just controversial moments, as revealing as these often are (Braun 2013), but also those aspects of distribution infrastructure and strategy that have been settled, which might have been otherwise but have become seemingly natural. These stable moments in the journey of content are what Bruno Latour (1988) refers to as bits of the chreod, the apparently “necessary path.” Tracing out the “voltas” of media distribution allows us to reevaluate these moments, an exercise that, like all good provocations in sociology, invites us to plumb and pry at the balance of structure and agency enacted in these technologically mediated, inherently social systems. As Lisa Parks (2012b) puts it in highlighting the need to examine the infrastructures of media distribution, it is a call for a form of criticism that is organised not upon the spectacle of the monumental or the new, but upon the materiality of the thing—a

Mass Reach and Mass Media  250

form of criticism that leads to investigations of the raw materials, resources, labour, affect and energy required to sustain a system. (p. 81) THE IMPORTANCE OF TRANSPARENT INTERMEDIARIES

Among the taken-for-granted moments in the passage of media from producer to user is the invocation of transparent intermediaries—­“business-tobusiness” distribution firms. The role of these third parties is to provide the technical infrastructures for distributing a provider’s content while remaining invisible to end users. We have seen how the technical affordances provided by these actors can influence the flow and availability of content and trigger interesting questions about the intersection of structure and agency. For instance, YuMe’s algorithms enabled a technical compromise between the needs of MSNBC and its advertisers, pairing commercials with embedded video only in “brand-safe environments.” This ultimately made it financially possible for MSNBC.com to offer up an embeddable player that gave great freedom to users, while at the same time reproducing and reifying in a new context what was still ultimately a traditional ad-supported content model. More broadly, as I argue in Chapter 2, by enabling millions upon millions of video views, and simultaneously collecting enormous volumes of information about audiences and their use of content, companies like Brightcove, YuMe, Tremor Video (which subsumed Transpera), and their numerous competitors are helping to reshape content providers’ understanding of their viewership. Nearly every online firm involved in distribution today provides its clients with extensive analytics, taking advantage of content providers’ growing appetite for this information to secure their own position and influence in an evolving media ecosystem. The experience of the firm Ooyala is typical. The company is a Brightcove competitor and was just entering the market around the time MSNBC .com was weighing whether to build its own video player or buy one off the shelf. Ooyala’s “Backlot” software provides both infrastructure and analytics. It delivers clients regular performance breakdowns for their online content, detailing the number of times a video appeared in users’ browsers and devices (“displays”), the number of times it was actually played (“plays”), and the total number of hours of video playback delivered over the requested interval. These performance metrics can be broken down by numerous categories, including the type of device, operating system, and browser on which the video was played; the countries, cities, and cli-

Conclusion 251

ent-designated market areas in which it was played; as well as the web domains and specific webpages on which it was viewed. The popularity of videos in different geographic locations can be displayed via tables, graphs, and “heat maps”—actual geographic maps on which target markets, states, or countries are shaded according to their viewing habits. Clients can also see how many times a video was played by unique users, how many users watched the whole thing, and a how many played through at least 25, 50, or 75 percent of the clip. The immediacy of these viewing metrics can be striking at times—for example, clients can pull up a video and, as it plays, watch the software sketch a graph that plots the number of viewers for each second of the clip in real time. The application also keeps a track of more longitudinal measures, such as the average number of videos watched per viewer per day, the “conversion rate” for each video (that is, the number of times it appeared and how often it was viewed), the average amount of time viewers spent watching it, and the amount of data in gigabytes delivered by the service. Sharing on social networks, via e-mail, and through video embeds is also recorded, and can similarly be broken down by region, device, and web address. As C. W. Anderson (2011, p. 51) put it, “The manner in which [media producers] imagine their audience has public consequences.” In short, as digital distributors increasingly become ratings and analytics firms, the sorts of audience surveillance they provide will have important effects on the our media culture and the content available to us. As I argued in Chapter 2, not everything that we value or find consequential about our cultural identities or public discourse can be easily quantified, measured, binned, graphed, or regressed, and because those things that are measured may be condensed, weighted, and acted on in different ways by content providers, each of which has its own commercial interests, media scholars have a duty to pay close attention to what transparent intermediaries and new modes of distribution mean for the “configuration” (Woolgar 1991), “institutionalization” (Napoli 2011) or “industrial construction” (Turow 2004, 2011) of audiences, as well as the public consequences of those constructions. In the same way that YuMe’s infrastructure reproduces traditional economic roles of advertiser and publisher in new media spaces, Ooyala’s Backlot software is replete with tabs and buttons with labels like “Ad Sets” and “Paywalls” that reify and replicate existing business models and industry categories. The powerful analytics generated by transparent intermediaries may likewise alternately reinforce and unsettle the industry’s under-

Mass Reach and Mass Media  252

standing of its audiences in culturally significant ways. And while it would be foolish to be needlessly or naively critical of such features—they are, after all, logical and intelligent responses to economic realities—media researchers and critical scholars going forward would do well to explore the balance and locus of structure and agency enacted in the technical architectures of online distribution, as well as the work they do as pieces within heterogeneous systems that prop up some cultural and economic categories, while helping to disrupt or reimagine others. RELATIVE SYSTEMS, VANTAGE POINTS, AND LOCAL TELEOLOGIES

Much of the literature on organizations has continued to evolve to reflect a shift in many economic sectors from a situation where firms operate largely via top-down hierarchies optimized for the production of scarce commodities to one in which industries focus far more on generating informational goods by way of “flatter” (that is, less hierarchical) and more chaotic organizational structures. In these newer structures, traditional hierarchies are supplemented or even supplanted by a wide range of divisions and project teams, inside and outside the traditional boundaries of the firm, constantly monitoring and reacting to one another’s movements in real time. These insights can be applied back to scholarship on sociotechnical systems, where many (though importantly not all) case studies have hitherto centered around projects wrought by more traditional organizational forms, and—despite an appreciation for heterogeneity—have examined even non­ traditional systems from the privileged position of a single builder. This approach, however, leads to a top-down-esque picture of one person or group subordinating other actors into a system that works for the designers, when in fact, as systems scholars have long acknowledged, the appearance of a system largely depends on where, and from whose point of view, one begins to trace out the networks of relationships involved. Indeed, in a “postmodern” or “heterarchic” environment, this observation becomes far more than a truism or methodological caveat. It is rather the key to understanding large contemporary media systems in an era when conglomerates not only support the creation of traditional evening newscasts, but also have a hand in operating cruise lines and producing jet engines. Making sense of such apparently disparate operations only works when we are willing to hold in abeyance our tendency to view large systems as singular—if contested—objects of inquiry. Rather, in tracing out a distri-

Conclusion 253

bution system that somehow knits together social media developers, TV anchors, and defense contractors, we must admit that we are dealing with multiple overlapping systems, all being constructed simultaneously by many different system builders within and across organizations—each of whom has in mind a system that accords with her own unique goals and understandings and each of whom is attempting to promulgate her own provincial teleology by rearranging shared resources. From the vantage point of producers at the Rachel Maddow Show, MSNBC.com was a system for reaching audiences. For Newsvine staffers, it was a financier whose deep pockets could pay back investors and bankroll further development of its social media site. For Nexidia’s leadership, MSNBC.com was a media client that could help diversify its portfolio of largely defense contracts. And for MSNBC.com’s web video team, Nexidia was a provider of tools that could increase the visibility and spread of its clips online and improve the monetization of its player. Contemporary online distribution systems are a product of the heterarchic institutional forms highlighted by organizational scholars, constructed by constellations of actors who often assess purpose and value in very different ways. Consequently when we describe such arrangements we find the history and sociology of sociotechnical systems blending with the sociology of knowledge. Media distribution systems elicit the language of “trading zones,” “boundary objects,” “network forums,” and “epistemic cultures” as readily as they do notions of “recalcitrance” or “reverse s­ alients.” ISOMETRIC PRESSURE

If we suspend the notion that systems are built—and change—­primarily through the work and direction of top-down management, we’re left with the problem of explaining how such systems evolve over time. What we previously thought of as a single unified system, we now understand to be a collection of overlapping ones, the product of multiple system builders working toward different visions by manipulating a common set of resources. Sometimes the objectives and actions of these different system builders dovetail, while at other times they conflict. Viewed in this context, a system reaches a stable state not because no one is applying pressure to it, but because many actors are applying pressures that ultimately offset one another. Stability comes when a formerly dynamic assemblage of actors and resources comes to rest in a formation that balances the competing pressures applied by numerous system build-

Mass Reach and Mass Media  254

ers. Whether a system appears stable or not also depends largely on where one draws the boundaries of the analysis (as demonstrated in Figure 6.1). Likewise, when this balance of pressures changes—because of a shift in the objectives of a particular system builder in the mix, because a new actor is introduced or an old one disappears, or because analytical boundaries have been redrawn—the system becomes dynamic again. To give a few examples, Newsvine’s social network was largely stable from the startup’s perspective until MSNBC.com was enrolled in its system, whereupon the disparate objectives of the larger company and its users ultimately compelled a rethinking and re-engineering of the Newsvine software, staff, and community. Likewise, the system of templates that MSNBC.com used for managing the web presence of MSNBC TV shows was largely stable from the online staff ’s perspective until the Rachel Maddow Show was born and began applying pressure to reimagine the site’s possibilities.2 EPISTEMIC CULTURES, EPISTEMIC ENCOUNTERS

Distribution systems are inherently tied up with the synthesis and dissemination of knowledge. As such, it is only possible to parse systems aimed at media distribution by knitting together scholarship on sociotechnical systems with the sociology of knowledge. This is not necessarily a tall order, given the close relationship of these fields within the larger discipline of science and technology studies. Gaye Tuchman (1978) and Karin Knorr Cetina (1999) have argued compellingly that different knowledge-based professions employ different methods of inquiry closely related to their subject matter and the institutions in which they operate. In short, they seek out, construct, package, and ultimately distribute knowledge and factual claims in accordance with provincial social and professional practices. As we’ve seen, the rise of heterarchic institutional forms means that many different epistemic cultures can be housed within a single firm. Additionally, as more and more of daily life and contemporary culture involves “objectual practice” (Knorr Cetina 1997, 1999, 2001) in the sense that aspects of our sociality are increasingly organized around common objects of inquiry or production, Knorr Cetina makes a compelling claim that the sorts of epistemic practices once relatively confined to knowledge-based professions like journalism or high-energy physics are now to be found everywhere in our culture. What’s more, the Internet has increasingly allowed for the formation of interest-based communities whose reputation economies center around

Conclusion 255

the interrogation of shared topics of interest (Bruns 2008; Wellman and Gulia 1999). Participation in such communities can range from highly informal (for instance, ending up part of a network of Twitter users who tend to value particular topics and sorts of information) to deliberate (such as joining an online forum for fans of a particular TV show) to quite formal (for example, going through the regimented process of joining the Debian developer community [Coleman 2013]). I contend that it is useful to think of such networks and communities as constituting, if not distinct epistemic cultures in the formal sense in which Knorr Cetina originally used the term, then at least individuals engaged in a series of epistemic encounters, sharing news and other content with an awareness and imagination of others’ standards—formal or ­informal—for valuing, interpreting, and distributing information via various forms of online sharing. As institutionalized audiences similarly shift from being seen by news workers as a unified public to being viewed as participants in different online communities and spaces, media organizations are also thinking in terms of epistemic encounters, increasingly packaging content strategically with an eye toward the online communities and locales that will give it the best social distribution. This is a productive way of considering the inversion of the traditional question of media sociology. Namely, if we formerly spoke of the mass media in general and the news media in particular as a bottleneck that selectively passed along messages to the public, to understand online distribution we must now begin to better appreciate the role played by a public that selectively passes along messages from the mass media. This framework offers us a way of productively extending more traditional lenses on the sociology of media; it also helps us to understand how distribution issues are increasingly being felt as editorial concerns within contemporary media organizations. Here, MSNBC.com’s proliferation of niche brands offers a prime example. As former CEO Charles Tillinghast put it, the idea was to attract users from the niche communities related to the site’s myriad interest-based columns and brands—astronomy blogs, hyperlocal news sites, and so on—in the hope that they would return regularly to find content relevant to their own social networks and “share it out from there” as a means of gaining social capital. This editorial strategy—along with the attempt to take advantage of economies of scale by hosting diverse editorial products on a common technical platform—has distribution concerns right at its heart and has become an increasingly common approach among a wide range

Mass Reach and Mass Media  256

of online media companies, from AOL, Maker Studios, and Yahoo! to Gawker, Vox Media, and Cheezburger, Inc. In short, an attention to epistemic concerns surrounding the “spreadability” (Jenkins et al. 2013) of content helps us to bridge our portrait of production and distribution, of content creators and end users. The sociology of knowledge, together with the sociology of sociotechnical systems, gives us a set of tools for examining attempts at mass reach, with and without the complement of mass media. DISTRIBUTION STUDIES: AN INVITATION

In keeping with a major theme throughout the book, I’d like to point to John Law’s (1987) notion of heterogeneous engineering as a framework into which many of the concerns and concepts discussed here can be productively nested. Law (2007), as with many scholars subsequently inspired by the idea of actor networks, often blurs the line between theory and method in his approach, which involves tracing out the associations between actors in a system. According to Law, the scope and means of analysis are not bounded by their mode of interaction (such as the individual, dyad, group, organization, institution, or society), but instead are determined by the effect to be explained—a distribution route, a publishing platform, an organization. He draws into the assay any agent whose influence is inscribed in that system, regardless of the (heterogeneous) forms these may take. Tracing the path of content—the figurative volta—as it moves from the producer through a network of different channels and technical platforms to end users is one such method of tracing associations, one that exposes the influence of many different actors, big and small and in myriad forms. These actors include entities and influences that are often invisible to us as end users of these systems (transparent intermediaries), or that, due to the creative fiction of branding, may appear deceptively unified when viewed from the outside. This approach also reveals that distribution systems stabilize and change in response to a sensitive balance of pressures applied by many system builders pursuing provincial agendas, only a few of whom occupy traditional managerial roles. And finally, it shows the intimate relationships between distribution and editorial concerns at a time when cultural practices once seen as the purview of information-based professions like journalism or scientific research have become a part of the daily life of the masses. I am all for heterodoxy in theories and methods. I have used the con-

Conclusion 257

cept of heterogeneous engineering and borrowed liberally from actor-­ network theory (ANT) because I find them incredibly useful. At the same time, I have unapologetically left behind terms and jargon from ANT that I found less useful, and appropriated numerous ideas from other, at times competing, theories of technology and systems. As Law (2014) himself says, “ANT isn’t the only way of thinking about heterogeneity. Different approaches are needed” (para. 3). It is my sincere hope that more and more scholars will bring their lenses to the study of distribution. Indeed there are many who are beginning to do so. As I highlight early on in this book, distribution touches on many of the most important concerns in media sociology, only some of which are addressed at length here, but all of which deserve sustained critical and scholarly attention. Media distribution affects the conditions under which we congregate and imagine our communities; the way we work and the way industries are structured in an increasingly knowledge-based economy; which social groups become visible and/or scrutinized within our larger society; the limits of our freedom of speech and assembly; our means of conversation and political participation; as well as our literary and artistic culture. What’s more, as our systems of distribution become increasingly responsive in the digital era, many of these conditions have the potential to change in real time. Now more than ever, we need good scholarship and an array of analytical frameworks devoted to the study of distribution. To close with the aspiration with which I opened this book, I hope that I have provided one such a framework that will ultimately contribute to our understanding of how it is that a great diversity of media products make their way to the screens sitting atop our desks, gracing our living rooms, and tucked in our pockets.

epilogue

While the bare beginning of Comcast’s takeover of NBCUniversal was in the works during my 2010 fieldwork, it was not until mid-2012 that Microsoft sold to NBC its remaining stake in the MSNBC Digital Network, thereby dissolving the joint venture and putting an end to the epoch described in this book. In the subsequent restructuring, control of the websites in the original MSNBC Digital Network was transferred to NBC News Digital, the entity within NBCUniversal that had previously managed a small handful of NBC News–affiliated sites not under the purview of MSNBC .com, including the Grio and Education Nation. The Blue Site—the general interest news site originally known as MSNBC.com—was rebranded as NBCNews.com, while the MSNBC.com domain was given over to a new website dedicated expressly to the cable television network. The SkyPad platform described in part 3 of the book was jettisoned and the new MSNBC.com was built on the open source content management system, Drupal, with much of the work contracted to high-end web design firm, Lullabot. And while Newsvine did not end up becoming the basis for the entire cable site’s publishing platform, the Newsvine team was a central player in the redesign and its “Newsvine 3.0” commenting system is currently the basis for all the community features on the new site. Its new group-based commenting system was also rolled out across Newsvine.com prior to the launch of the redesigned cable site. For a time, the biggest visible change to the Blue Site was its logo, which shifted quickly from “msnbc.com” to “NBC News.” But this site, too, was eventually redesigned in its entirety, as was the Blue-Site spinoff, Today. 258

Epilogue 259

com. Of these two redesigns—NBCNews.com and Today.com—the Today site initially hewed closer to the look developed under the MSNBC joint venture, while NBCNews.com’s makeover was more dramatic and prioritized mobile audiences, nearly tripling mobile page views of the site (Kirkland 2014). Some even argued that the site prioritized mobile to a fault, with a tiled layout and “hamburger menu” derided by critical users and by web designers unhappy with the site’s appearance on desktop browsers. NBC News Digital eventually responded to these critiques with a facelift for the site that returned at least some of the look of the old Blue Site. The original MSNBC.com video player, too, was phased out in favor of another existing player developed by Comcast’s transparent-­intermediary subsidiary, thePlatform. As mentioned previously, live streams of NBC News and MSNBC TV also became available to users who authenticated themselves as cable subscribers. Under NBC News Digital, management of much of the social media presence of MSNBC and NBC News was transferred to a dedicated social media team within the company, though the social media presence of individual shows continued to be managed by the programs and their affiliated web staffers. Meanwhile, the Maddow Blog, now the show’s “true” homepage and part of the larger cable site, ceased to look like a blog for a time, with a layout more evocative of a traditional news site’s headline stack. Its layout eventually returned to a (more) blog-like look, however, partly at the request of the Maddow staff. Even so, the Maddow Blog took on a more formal sheen, becoming less of a backstage look at the show’s thought process and more of a high-output liberal blog reminiscent of the Huffington Post. Some of these changes, like the rollout of a standalone cable site, Newsvine’s concomitant redesign, and the introduction of cable subscriber authentication, appear to be logical continuations of the story told across the pages of this book. Others require a bit more explanation. For example, in 2012, when it became clear that there was an opportunity to hire the prolific and influential political blogger Steve Benen, who’d previously written for the Huffington Post and Washington Monthly, the Rachel Maddow Show staffers decided they couldn’t pass it up. As with Laura Conaway before him, Benen was hired largely to keep on doing for Maddow’s web presence what he had already been doing elsewhere. But if Conaway’s modus operandi had been to develop content that opened a window onto the broadcast property she worked on, Benen’s liberal blogging style—while highly compatible with the show and its editorial outlook—was more outwardly

Epilogue 260

focused. The addition of a new blogger allowed Conaway, with her broadcast background, to pitch in more with the preparation of the nightly show. Benen’s output, meanwhile, continued to be prodigious, and subsequently became the primary content of the Maddow Blog. The increased output and change in tone led to a greater uptake of the blog’s content by columnists and writers for mainstream media organizations, and has yielded even greater traffic and social media exposure for the Maddow Blog around the web. Still, the blog’s content rarely appears on the front page of MSNBC .com, even though it is now a dedicated cable site. This is likely because even after restructuring, the cover of the cable site is still run by another editorial team with its own writers and editors, a team largely distinct from the editorial staffs of the cable channel’s individual shows. Even after restructuring and spinning off the cable site, in other words, differences in editorial outlook and editorial responsibility seem to still be influencing the placement of content on the site. The Maddow staff continues to run the Maddow Blog as their own website. Moreover, as indicated by the refigured layout of the Maddow show’s MSNBC.com pages, which were tweaked to be more blog-like than those of other shows, the Maddow Blog continues to affect design decisions in other parts of the company. Additionally, the Rachel Maddow Show, with the ongoing help of Will Femia, has continued extending its social media presence, with new accounts on Tumblr, Flipboard, and Google+, along with a rapidly growing email newsletter. Even in the few years since my fieldwork, social distribution has continued to increase dramatically in importance, to the extent that even when posts to the Maddow Blog don’t make it to the homepage of MSNBC.com, some of these social media accounts can, by themselves, bring as much traffic to Maddow’s pages as stories placed on the cable site’s cover. We can see here how heterarchy is alive and well within the constellation of companies and teams surrounding MSNBC, as well as how the addition of new actors, from a behemoth like Comcast to an individual like Steve Benen, can result in previously isometrically stable arrangements becoming dynamic. In this same vein, we can see how, with the departure of Microsoft, decisions were ultimately made to move important properties off of the platforms developed with Microsoft’s resources. While NBC News Digital’s own design teams and those of major subsidiaries like Newsvine and Comcast’s thePlatform are obviously still prominent decision makers behind the evolving web presence of MSNBC and NBC News, more work

Epilogue 261

is also being outsourced to outside contractors, increasing the number of actors—including the number of transparent intermediaries—involved in distributing their content to users, as well as the number of vendor relationships needing to be factored into many decisions. At the same time, many of the old technical infrastructures of the joint venture, while largely removed from the public-facing aspects of MSNBC and NBC News sites, have continued to trundle along behind the scenes, accessible to staff wishing or needing to use them as part of their workflow. During the transition to Comcast’s video solution, for example, video was still published in MSNBC.com’s old video player—even though the links were not made public—as part of the process of subsequently publishing it in the new video player. This odd workaround—a nice example of heterogeneous engineering—permitted videos to be readied for the launch of the new cable site while they were also still being published to and available on the legacy site. The kludge also continued to operate for a time after the launch of the new site. In fact, behind the scenes a number of joint-­venture-era editorial tools were simply left running, and occasionally used by staffers, until someone got around to switching off the physical servers on which they resided. So even amid some changes that seem to show the influence of a centralized management and hierarchy—the reining in and professionalization of social media, the possible standardization of NBC News Digital’s various publishing platforms, or the integration of Comcast’s thePlatform video player, for example—we can see that complex assemblages of system builders persist both inside and outside the organization. Indeed, as more teams of heterogeneous engineers are created inside the company and the number of outside vendor relationships that have to be managed increases, the complexity of overlapping interests surrounding MSNBC and NBC News’ online presence may well grow over time, just as it appears to have done over the history of the joint venture. As a staffer explained, “Even as corporate initiatives bring some things into alignment, there’s an overall tendency towards chaos.” Call it “chaotic” or call it “postbureaucratic,” the original MSNBC, after all, evolved and became successful largely in response to the mundane inter- and intra-organizational tensions among its system builders, whereas the platonic top-down vision of its corporate executive founders ultimately failed to materialize. At MSNBC and other media companies studied by organizational scholars, the systems that appear hierarchical and uniform on paper are likely to be far more heterarchic and heterogeneous in practice.

This page intentionally left blank

NOTES

CHAPTER 1  Introduction

 1 A notable, historically oriented exception is the aforementioned work of Richard John (1995).  2 See also Connected Viewing, an excellent new anthology edited by Jennifer Holt and Kevin Sanson (2013).  3 A few media scholars, including Turow (1992) and Silverstone (1994), have taken a systems perspective on media as well. Their contributions will be explicated in the coming chapters.  4 For an excellent review of such critiques, see Ryfe (2012).  5 In the wake of restructuring that occurred after the primary fieldwork for this book, Newsvine’s ownership has shifted to NBCUniversal.  6 For the most part, I have temporally bounded the narrative of this book around the life of MSNBC as a joint venture, beginning with its inception in 1995 as a cooperative project between Microsoft and NBC and ending prior to its ultimate acquisition by Comcast in 2012. This decision has both analytical and highly pragmatic components. Analytically, the time between the joint venture’s creation and its dissolution seems a fitting and logical epoch to chronicle in the history of the organization and the landscape that shaped it. From a pragmatic standpoint, the dissolution of the joint venture also led to the migration of many of my sources to other firms and professional opportunities, as well as the end or modification of a number of my original access agreements. In writing this volume, I have tracked the relevant industry news and taken care to follow up with a number of my key sources and other relevant contacts since the time of my fieldwork. In the Epilogue I take up the Comcast acquisition, including its larger meaning for the theories, infrastructures, and trends documented in this volume. CHAPTER 2  Why Media Distribution Matters

 1 As John (1995) and Gillespie (2014a) highlight, government regulations surrounding many forms of communication that predate the Internet, such as postal mail or the telephone, proscribe many forms of explicit surveillance. Still, it’s worth noting physical analogues to the sorts of digital surveillance referenced here. Lisa Parks (2012b), for instance, uses the example of a speech by Dutch parliamentarian Geert Wilders about the “Islamicization of Europe” in which he argued that the very fact that many Muslim immigrants’ satellite dishes were pointed toward stations in their countries

263

NOTES TO PAGES 27–43  264

of origin was evidence that these families did not wish to consume local media or, by extension, to integrate into European society. In other words, while Internet services may have piqued interest in, and concern about, the potential use of distribution infrastructures as a politicized means of user surveillance, such issues are not unique to digital technologies.  2 The “PROTECT PIPA Act” is itself an acronym for Preventing Real Online Threats to Economic Creativity and Theft of Intellectual Property Act. CHAPTER 3  Mapping Sociotechnical Systems That Deliver Media

 1 In his various works, Hughes uses the notion of “system” in at least three different, if related, ways. One is an actor’s category from a very specific historical moment—the development of formal disciplines of systems engineering, operations research, and systems analysis. Another use is to suggest a management strategy strongly influenced by that historical moment—“systems thinking” in a broad variety of organizational contexts not limited to those taking a formal systems-engineering approach. And finally, he also uses “system” as an academic construct that can refer to not only contemporary system builders, but also individuals like Bell, Sperry, or Edison, who historically predated the disciplinary notion of systems engineering, but engaged in similar enterprises. It is in this last capacity that I will tend to use the notion of system, though I accept Hughes’s (1998, 2004) contention that formal notions of systems engineering have had an influence on the organizational forms adopted by Internet companies.  2 This view of his work, which Law puts forward, comes from a specific moment in Hughes’s scholarship. Hughes (1998) went on to chronicle shifts in the nature of technological systems, concluding that in the last half of the twentieth century, system building took on a distinctly postmodern style, exemplified by an embrace of discontinuity and a shifting “commitment by industry to change-generating projects rather than to long-lived processes” (p. 5). Moreover, he suggests that this movement has been lead and most strongly exemplified by professionals in the computer and Internet industries (Hughes 1998, 2004), calling it a shift to an “information-revolution organizational style” (Hughes 2004, pp. 101–102). Many passages in his later work are in fact very consonant with Law’s notion of precariousness and continual change. We’ll revisit Hughes and his notion of postmodern systems more extensively later.  3 As mentioned, readers familiar with actor-network theory will no doubt think here of Latour (1988) and his notion of “delegation” of human tasks to technological artifacts. While Latour’s and Hutchins’s analyses are quite consonant in this regard, like many scholars of technology, I find the strong form of material agency in many of Latour’s accounts—in which gadgets are seemingly imbued with volition—to be problematic at times. As such, while I acknowledge Latour’s significant contributions to the sociology of technology throughout this text, my own conceptual frameworks are often grounded elsewhere.

NOTES TO PAGES 45–53  265

 4 Network news divisions have a similarly tense relationship with their broadcast affiliates when it comes to distributing content online that might cannibalize local ratings.  5 When MSNBC.com did provide a live stream of television news coverage in 2010, it almost invariably came from the broadcast news division, rather than from the MSNBC cable channel, to avoid cutting too deeply into the MSOs’ bread and butter. This continues to be true for its successor, NBC News Digital.  6 Web producers had access to a show’s script, though the utility of this access varied. For example, a show like Countdown with Keith Olbermann placed stories into predictable numbered slots, which could easily be assigned headlines and other metadata in advance. In the case of other programs, scripts could prove less predictable and prescriptive, and evolve more over the course of a day. In such cases it was often simpler for the web producer to focus on creating metadata for the finished broadcast.  7 Redmond editors had access to a live East Coast feed of MSNBC TV, as opposed to the one that cable customers in Seattle saw, which was delayed by three hours to accommodate the difference between time zones. For shows like the Today Show or NBC Nightly News, which could change slightly between their live and taped versions, the Redmond offices would record both the live version and the West Coast edition three hours later, updating the online version of the clips accordingly. There were ­actually a number of procedures at MSNBC for sending a video to the web. Those that are relevant I will describe in more detail later. At the moment, I’m simply illustrating the method by which the majority of television content was processed and made available online.  8 While the MSN brand still exists and is identified with a number of webpages, as well as a collection of apps and services run by Microsoft, MSN no longer includes MSNBC or NBC News content. At the time of this writing, MSN’s web presence is packaged largely as a single website with different sections pertaining to various topics like money and entertainment. This is in contrast to its earlier incarnation as a family of distinctive standalone websites sharing some common links and branding.  9 In this vein, Microsoft invested in the 1996 startup WebTV, and went on to acquire the company the following year, eventually turning it into a full division of the corporation (Croal 2000; Markoff 1997). 10 Under the terms of the deal, Microsoft paid $220 million for a 50 percent stake in NBC’s America’s Talking cable channel, which was then converted into MSNBC TV. Moreover, Microsoft and NBC agreed to each invest a minimum $200 million in the creation of MSNBC.com (Collins 2004). 11 More specifically, NBC acquired an 82 percent (i.e., majority) stake in the cable channel in 2005, agreeing at the time to purchase the remaining shares from Microsoft in two years (Carter 2005). 12 After NBC took a controlling interest in the cable channel, the offices were eventually consolidated into 30 Rockefeller Center in Manhattan, where NBC News is headquartered, as a cost-cutting measure.

NOTES TO PAGES 54–62  266

13 This multiplicity is something that will be fleshed out and explored further in the following chapters as we dig further into the heterarchic (Stark, 1999, 2009) nature of MSNBC and contemporary media firms. 14 A GUID is a unique identifier, generally a URL, that accompanies the clip when it is published in RSS feeds. CHAPTER 4  To Turn It On and Have It Play

 1 Whether and how such cross-linking might have been mandated by Microsoft beyond the standard header and footer links included in page designs was a relevant, but obviously touchy, subject. Consequently I have neither firsthand knowledge, nor firm documentary evidence, of it. Without passing entirely into the realm of conjecture, there’s limited visual evidence that different MSN sites may have been obligated to include one another’s content. MSNBC.com itself, for instance, housed links to MSN’s Powerwall and Wonderwall gossip sites, which were run on very different editorial strategies than MSNBC.com. It’s also worth noting that these links were carefully tucked away in inconspicuous places on MSNBC.com’s front page. This sort of mandated cross-linking, if or where it occurred, would be a resource in distribution terms, in the sense that it would drive additional traffic to MSNBC.com, but it’s also easy to imagine MSNBC.com editors threading a needle in the way they reciprocated—including links to other sites as required, but finding ways to make them inconspicuous so as not to dilute the editorial image of their own site.  2 Microsoft Internet Explorer has since dropped substantially in market share, but was still the most-used web browser at the time.  3 Less technical readers may want some introduction to the basic concept of objectoriented programming. A nice analogy would be that of MUDs and other computer games or environments that simulate the real world. In the physical environment, it makes sense to us that everyday objects lend themselves to a particular class of uses. We understand that you can write on a sheet of paper, fold it, crumple it, tear it, and so on. We also understand that paper is less useful for other purposes—we wouldn’t try to put out a fire with it, or use it to hammer a nail. When an object like a piece of paper or a pistol is recreated in a game, the designers have to explicitly define all the things it can be used for and how it should behave under specific circumstances, so that the user can interact with it in familiar or predictable ways. Similarly, in object-oriented programming, developers create data “objects” that are context-aware, allowing them to be used in multiple ways. For instance a series of data points—a bunch of headlines, say—could be programmed as an object that could be presented variously as a list, a series of slideshow captions, a feed of RSS items, and so on. In this same way, clips in MSNBC.com’s video player are able to appear in a number of forms across the site.  4 MSNBC.com was in a transitional period at the time of my fieldwork, and some shows like the Ed Show or the Last Word had made their homepages a blog rather than a video-centric show front. This trend will be discussed in Chapter 10.

NOTES TO PAGES 66–91  267

 5 At the time, this initiative by MSNBC management was not a decision to become a liberal counterweight to Fox News, but rather one to begin offering and highlighting opinion programming, irrespective of its political leaning. As former NBC Universal CEO Jeff Zucker later described it at a gathering of Harvard Crimson alumni, “We didn’t have a secret meeting where we said, ‘maybe we should just go left’” (Stoll 2011). Over the years, MSNBC has employed a number of prominent conservative commentators in primetime, including former Republican Congressman Joe Scarborough and Policy Review alum Tucker Carlson.  6 Similarly—and not unrelated—the twenty-four-hour news cycle is frequently pointed to by industry observers as an explanation for declines in television ratings among evening news broadcasts.  7 I had a very similar conversation with a CBS Interactive producer concerning the use of automation to lay out portions of the CBSNews.com homepage: “Obviously you want to incorporate as much automation as possible because of limited resources. But [you need to be able to] overtake that automation as easily as possible. . . . On any given day you want to customize. You don’t want just the last three videos that went into your system, you want . . . the best three. And do you populate that by most popular? Sometimes most popular isn’t exactly the way you want to present your site. If it’s ‘Man Eats Own Foot,’ and you’ve got three sub-stories on that, you probably don’t want that to be the image of CBSNews.com, for your average user to come and say, ‘This is what CBSNews.com is all about.’” CHAPTER 5  Clip & Share

 1 This observation highlights the “monitorial” role (Schudson 1998, 2000) of social distribution as conceived by Prior (2007), wherein politically motivated actors identify personal and civic harms and bring them to the attention of the public. See Chapter 2 for a fuller description of this concept.  2 The resizing feature was on the development roadmap, but not yet available when the player was first released (Young 2008).  3 When clips were first posted, in lieu of an official transcript of the clip, the system used the more immediately available closed caption text that accompanied each show. This would be updated and replaced once a better transcript was entered into the system, but also allowed users access to the functionality as quickly as possible.  4 These technological restrictions by Hulu include, but are not limited to, extensive geofencing that in many cases denies access to all countries outside the United States; domain-blocking that shuts down embedding of content on explicit sites and un­ authorized mirrors; and user-agent banning that prevents devices other than PCs from loading Hulu videos unless the user logs in as a premium subscriber.  5 At the time of this writing, Iran has no official copyright relationship with the United States and does not recognize World Trade Organization copyright laws.  6 This philosophy was similarly reflected in the web venture’s expansion into the MSNBC Digital Network, which encompassed not just the Blue Site, but also multiple

NOTES TO PAGES 93–97  268

online destinations that were subsequently launched, acquired, or spun off from the original domain. See Chapter 11.  7 Much the same rhetoric has accompanied other new technological platforms. A particularly good example is that of the American broadcast news networks when they turned to blogging in 2005 (see, for example, Lafayette 2005; Squires 2005; Steinberg 2005).  8 Note that I do not classify MobiTV as a “transparent” intermediary because, as a branded subscription service, its role in distribution was readily apparent to mobile audiences.  9 In particular, Apple had already begun to publicly refuse to support Adobe’s Flash video standard on iOS devices like the iPhone, iPad, and iPod Touch (Jobs 2010). 10 It’s worth noting Transpera’s (since acquired by Tremor Media) long-term business strategy here. Mobile devices were likely to eventually converge on a video standard —or grow support for more sorts of video. But Transpera—or rather, Tremor—in controlling this extensive ad network, hoped to maintain a valuable point of leverage in this market over the longer term (Roizen 2008). 11 For less technical readers, WebKit is a layout engine—a component of a web browser that controls how webpages are displayed. Different browsers use different layout engines. For instance, on your desktop computer, Google Chrome and Apple Safari browsers use WebKit, Mozilla Firefox uses a layout engine called Gecko, Microsoft Internet Explorer employs another called Trident, and the Opera browser until recently used one called Presto. Phone browsers similarly employ one or another layout engine to render web pages. Differences between layout engines and the way they interpret web standards—and therefore display pages—have lead to much wailing and gnashing of teeth among web developers, who must write pages that will display correctly in all of them. The choice to develop MSNBC.com’s mobile sites for a single layout engine thus reduced the resources necessary to construct a working mobile site, while ensuring that the site would still work on the most popular phones and mobile devices (since these generally use WebKit). In that regard, it was a way of threading a needle between opposing pressures. It did, however, have the side effect that some mobile browsers and phones would not work with some MSNBC.com mobile sites. An example is that MSNBC.com show fronts did not detect when they were being loaded in an Opera mobile browser. As a result, the device would attempt to load the desktop video player. On a phone that did not support Flash, this resulted in a largely nonfunctional page. 12 As critical observers, we might well ask what it means to use a data point like choice of device as a proxy for other attributes about audiences and point to how similar distribution decisions, when made by a large cross-section of media providers, have the potential to systematically exclude or differentiate particular groups. As Lotz (2007) notes, the cost of technologies like the iPhone “introduces significant socio-economic distinctions in who can afford the technologies that enable . . . convenient and mobile

NOTES TO PAGES 97–112  269

television, and such economic division further fractures norms of television viewing within the culture” (p. 249). Such fractures arguably become even more significant to take note of when the content in question is something of particular democratic importance, such as news or political commentary. 13 In addition to an Android version of the MSNBC.com app, work was at one point begun on an Android port of the Rachel Maddow iPad app. An Android version of an app featuring NBC News coverage of the 2011 British royal wedding was also announced, but apparently not delivered in time for the wedding itself (see Choney 2011, as well as the comments that follow the post). 14 For less technical readers, CSS stands for “cascading stylesheet.” CSS stylesheets are sets of instructions directed at web browsers that change the look and feel of a webpage, in principle without altering any of its content. Generally, a standard stylesheet will be developed and applied to many or all of the pages of a large website to give the whole site a unified visual identity. A CSS override is the selective replacement of a few instructions in a standard style to distinguish one page from another. In other words a bit of strategically altered CSS code on top of a highly modular page design allowed the designers to create mobile show fronts for different programs without too much heavy lifting. 15 This marketing research was described in my interviews with MSNBC.com president Charles Tillinghast, Today.com editorial supervisor Sarika Dani, and Today.com director Jennifer Brown. 16 Viacom simultaneously sought a declaratory judgment on the same matter (Lawler 2011a). CHAPTER 6  Breaking Apart the Monolith

 1 I am not claiming here that the professional identity of journalists fails to cut across intra-organizational boundaries—on the contrary, Deuze’s (2007) own review of changing work conditions within the news media suggests that professional identity is still a dominant force in news production. At the same time, as he puts it, “professional ideology” is “interpreted, used and applied differently across media” and “largely determined by the culture of the newsroom or publication one works for” (p. 168).  2 There are extant debates among the scholars who subscribe to a number of the theoretical frameworks I point to here. For example, social construction of technology (SCOT) theorists often criticize actor network theory for it’s notion of “general symmetry”—for treating human and non-human actors as equivalent for analy­ tical purposes. Meanwhile, SCOT tends to focus on negotiations and controversies surrounding the stabilization of technological forms. Pinch (2010a) refers to this as a conceptual “methodology for making certain non-humans visible. If the non-humans are relevant to social groups then they are relevant to the analysis” (p. 83). This was a response to the charge by theorists sympathetic to ANT, who have often critiqued SCOT as an attempt to reduce explanations of socio-technical systems to social

NOTES TO PAGES 113–127  270

effects. In my view it is unnecessary to take sides in such a debate—rather, I wish to acknowledge and draw from the theoretical insights of multiple traditions here and in Chapter 11, where I introduce additional frameworks for exploring knowledge systems, such as Knorr Cetina’s notion of epistemic cultures.  3 In light of this, there have now been several articles and manuscripts that make the case for applying ANT to media and/or news work. See, for example, Turner 2005; Hemmingway 2005, 2008; Couldry 2008; Domingo 2008; C. Anderson 2010, 2013; Mould 2009; Plesner 2009; and Schmitz Weiss and Domingo 2010.  4 In Chapter 5 I pointed out how many technological systems that appear stable to the end user may actually be in flux.  5 For example, Mould (2009) uses ANT to examine independent film production in Australia, Hemmingway takes an ANT approach to the rise of “personal digital production” within the BBC, and Plesner (2009) examines the use of email as an information gathering device in newswork using ANT as a framework.  6 The historical reasons for this shift are complex and a detailed account of them is beyond the scope of this manuscript. Hughes (1998) attributes the transition to the spread of countercultural values during the 1960s, the demonization of the militaryindustrial complex—which pioneered many aspects of twentieth-century system building—during the Vietnam War, as well as the prominent failure of several Great Society projects aimed at bringing a rather too hierarchical systems approach to public works and welfare programs.  7 The program, titled Hubris: Selling the Iraq War, has been bootlegged to YouTube multiple times by different users. The most popular copy alone has garnered 180,872 views at the time of this writing.  8 A similar authentication system was pioneered by, and proved lucrative for, video sites like MegaVideo that are (or, in the specific case of MegaVideo, were) notorious for providing streams of pirated content. It would be quite interesting to see cable and network executives once again take a page from “unauthorized” distributors.  9 In fact, Hughes’s (1987) writing recognizes that different systems exist simultaneously, compete, and even spin off from one another. The difference between the portrayal of systems by Hughes and by ANT scholars is less about how each understands or renders the universe of systems and more about what the different analytical frameworks emphasize about this universe and which elements of it they choose to grapple with most directly. CHAPTER 7  Friends of Mine, Friends of Theirs

 1 The version launched in December 2005 was initially referred to as the first private beta test (Newsvine 2005), but as the private beta was expanded to additional waves of users, the initial testing came to be referred to as an alpha test (Newsvine 2006d).  2 Some proprietary solutions require software from the same vendor at other levels of the solution stack. For instance, Microsoft’s IIS server software is generally intended to be run on a Windows machine.

NOTES TO PAGES 127–136  271

 3 While this description of a solution stack for building a web application is accurate, since 2010 the increase in popularity of cloud hosting services, like Amazon Web Services and Google App Engine, has led to additional and alternative methods for deploying websites and web applications.  4 Python is used for writing many types of software, including desktop applications. Since Python is not specific to the web, programmers who use it for web development often have to add extra instructions to their programs, instructions that would be assumed in a largely web-specific programming language like PHP. Python is still a popular scripting language for web development, however, and it’s often used in conjunction with “frameworks”—extensive pre-written libraries of code that developers can borrow, and so obviate the need for any extra web-specific instruction. The upshot is that while you will often hear of sites written in PHP, when Python is used on a site developers will more often refer to the name of the framework, such as Django or Pylons. Frameworks for other scripting languages, such as Rails (a Ruby framework), have also become shorthand terms. In fact, time-saving frameworks are popular for many scripting languages, including PHP, as a means of speeding up development and organizing code more efficiently.  5 On Newsvine, users could only vote an item up, not down, although they could also report an item as abuse.  6 As with other features, like voting on content, I don’t wish to portray this functionality as alien or unique. I feel it’s possible, and important, to be analytical about the inclusion of features and functionality without pedantically describing common items as though readers had never encountered them before. At the time of this writing, for instance, many of the most popular forum services and software packages including ProBoards, Simple Machines Forum, and phpBB, include similar profanity filters, as do major news providers like CBSNews.com (in their comment threads), and public-facing portions of major social networks like Facebook.  7 I will revisit the sixth clause and its fate later.  8 Yockey also aptly compared the social and administrative role of the Code of Honor to the list of basic policies and guidelines on Wikipedia that has been examined by scholars like Geiger (2007), Black et al. (2008), Slattery (2009), and Reagle (2010).  9 I should add some context to the notion of a single moderator. While Newsvine had a dedicated moderator, Tyler Adams, COO Calvin Tang also played an active role in interacting with the user community until his departure from the company in 2010. Newsvine hired a second full-time moderator shortly after my fieldwork was complete. An avid user was also hired some time beforehand as a part-time employee, and while it was not her primary responsibility, she occasionally shared in some of the moderation duties as well. The general point, however, is that moderation duties on Newsvine.com had always been, and continued to be, carried out by a relatively tiny group of individuals. 10 The Watchlist was one of Newsvine’s following mechanisms. On the site you could friend users and join groups, as you’d expect from a social network. But it was

NOTES TO PAGES 140–150  272

also possible to subscribe to content from an author or group on the site without designating the author as a friend or joining the group, an act called “Watchlisting.” Facebook would introduce a similar distinction between “friends” and “subscribers” in 2011. 11 While the Newsvine site has changed substantially since my fieldwork was conducted, much of its underlying infrastructure remains the same. There are a small number of moderation tools at Newsvine that work only if they are not made public. Per the request of the Newsvine staff I will refrain from describing these here, since they are not necessary to the discussion and tend to follow the more general pattern of observations in this chapter—that is, the salient points here can be made without their inclusion. 12 Today sites across the web have numerous content-sharing buttons. At the time, however, such buttons were far less common, particularly on mainstream media sites. 13 The following year, the New York Times would close its bulletin-board system, one of the few community features it supported at the time. Many of the users from that bulletin-board system emigrated en masse to Newsvine, creating a large group on the site self-titled the “NYTimes Forums Refugees.” This virtual diaspora and the ensuing culture clash on Newsvine is worth its own treatment, however it is of limited relevance to the present discussion. 14 It’s interesting to consider the ramifications for Newsvine’s public discourse of privileging the comments of a particular publication’s readership (which as I discuss in Chapter 2, may already be somewhat homophilic). Indeed, when the New York Times closed its own reader forums and a large group of Times readers migrated en masse to Newsvine, the staff recounted that there were, for a time, significant culture clashes between the group calling themselves the “NYTimes Forums Refugees” and Newsvine’s organic user base. CHAPTER 8  Different Things to Different People

 1 The designation “M2” comes from a new table that was added to Newsvine’s database to facilitate this upgrade. The widgets that users added to their pages were to be called “modules,” but a “modules” table already existed in the database. So the new table was instead called “M2” as shorthand for “modules 2.” The “M2” moniker eventually came to refer to the entire update of the site architecture that was developed in tandem with the new modules. Ultimately, the term ceased to refer to the database at all and became instead a numbering system for major updates to the site’s architectural framework. At the time of my fieldwork the site was being readied for a transition to the M3 framework, and the original codebase had been given the retronym M1.  2 As an example, here was one upset user’s reaction to the announcement on the Newsvine blog: “You know, Mike, my WHOLE thing all along is the idea of ‘baiting’ people with the promise of a pay-off for their work while you and yours were the ones really cashing in. I used the phrase ‘bait and switch’ and I have been proven true. If you had just dropped the pay for seeding thing like I suggested from the outset, I and others

NOTES TO PAGES 151–177  273

wouldn’t be pissed off at your ‘success.’ Your tactics smack of ethical failure. You promised community and then cashed in!”  3 Another, more recent example, is that of the Huffington Post, which was criticized for not paying writers at all and subsequently cashing in on their efforts through its sale to AOL.  4 RAID stands for “Redundant Array of Independent Disks.” It is a hardware configuration intended to make data storage and provision more reliable. When a RAID system fails, data become inaccessible; a RAID failure in this context would in turn make a website inaccessible.  5 For less technical readers, Linux—the operating system on which Newsvine built its site—is a variant of a larger family of operating systems called “Unix systems” descended from and/or modeled after the original UNIX operating system developed at Bell Labs in the 1960s and 1970s.  6 One other difference was that Newsvine was able to include the full text of MSNBC .com articles as seeds, rather than just links to them, as was the case for seeded articles from unaffiliated media outlets.  7 While boyd and Ellison (2008) do an excellent job of reviewing the rise of general interest social networks, there’s been little scholarly work on the effects that the ascension of these services had on smaller, niche social networks. Boyd’s review, for instance, appears to assume that because long-tail social networks were pursuing niche audiences, their trajectory has been independent of, or in response to, the rise of sites like MySpace and Facebook. CHAPTER 9  If We’re Not Online, We Don’t Exist

 1 Aspects of this arrangement may be somewhat unique to U.S. cable news. For an excellent overview of cable news production practices in non-U.S. contexts, see Bivens (2014).  2 I have no firsthand knowledge of this, but it was the case at many networks. A representative quote comes from ABC News’s Judy Muller, who wrote in 2005 about her reticence to blog, “As if we didn’t have enough to do? Most reporters already have to file stories for a variety of platforms, as we call them: TV, radio, Internet, digital programming. Imagine if we had to keep running diaries of what we’re doing as we’re doing it. Imagine a surgeon trying that. . . . [It] would be a malpractice lawyer’s dream.”  3 As a few examples, a number of editors of the tech blog Engadget were rumored to have defected from AOL over their dislike of its slow-to-evolve content management system (Winer 2011), while Google has acknowledged that its release of Wave cannibalized the efforts of, and market for, its other product teams, including those working on Gmail and Buzz (Gray 2010). As an example from another legacy news company outside of television, when the Washington Post Company consolidated its online publishing tools onto a single CMS called Methode, the paper’s ombudsman, Patrick Pexton, opined on the new system as “a bafflement to most of us trying to

NOTES TO PAGES 181–216  274

figure it out. Every newsroom employee is struggling with it, not just the tech people. And it has a horrible name” (Pexton 2011).  4 While my sources referred to Femia as being the only MSNBC.com TV-web producer dedicated to a single cable program, owing to the limited resources of the TV-web staff, he continued to pitch in and help out with other primetime programs. So although it’s unequivocally true that he dedicated more time by far to a single show than any other member of the team, it’s not entirely accurate to say that he no longer worked on other programs. CHAPTER 10  The Maddow Blog Is Born

 1 Femia, of course, was in the middle of this colloquy, with the Maddow program as his “client” and MSNBC.com as his employer. While this might easily have been a difficult position for him, he recounted: “At the same time it was a great relief that I was essentially powerless in the whole exchange. . . . I didn’t really wield any influence, other than that I wasn’t afraid to be frank even with Charlie [Tillinghast] and say, ‘Look, this is what’s going to happen. We don’t have an upper hand here. We don’t have any cards to play. If she doesn’t like what we’re giving her, then she goes.’”  2 This level of instantaneity was only available to paying Vimeo users.  3 An interesting side note is that Rachel Maddow gave everyone on her staff a FlipCam (a pocket video camera roughly the size of a phone) as a way of encouraging them to acquire more content that could be used for the blog, and even for the show itself.  4 NBC News more broadly also took measures to exert greater control over its web presence, creating a new “director of social media” position in 2010 and hiring former NPR chief and NYTimes.com head Vivian Schiller into another new position in 2011, as “chief digital officer.”  5 Readers familiar with the work of Star and Griesemer (1989) will note the resonance here with their concept of “boundary objects”—material objects that circulate among different cultures or constituencies that may value them for different reasons. Kellogg et al. (2006) have made similar observations about the nature of other information technologies that play a mediating role within contemporary organizations.  6 Under the new Comcast ownership, Newsvine did not ultimately build the new cable site in its entirety, but did build the community participation mechanisms that became a prominently featured component of the new site. CHAPTER 11  One Size Does Not Fit All

 1 Until now, when using the terms “system builder” and “heterogeneous engineer,” I have generally been referring to groups that work together to unfold a shared local teleology. So it may seem something of a shift to refer to individuals or loosely confederated groups who forward, share, and blog about media products as heterogeneous engineers. But this fits quite well with the relative nature of systems, where the identity of the builder and the shape of the system are different depending on the chosen vantage point. As Law (1987) states, “We are all heterogeneous engineers,

NOTES TO PAGES 217–230  275

combining as we do, disparate elements into the ‘going concern’ of our daily lives” (p. 133).  2 The Scoop column was actually folded into Today.com from its previous home on the Blue Site.  3 The idea that television editorial staff don’t have to consider the placement of their content—that they have a reliable outlet for whatever they produce, whether groundbreaking or mundane—is likely a bit overstated. Correspondents compete for air-time and frequently have to think carefully about which news shows might be interested in the pieces they’re assembling. And while cable producers attached to a specific program have a comparatively reliable path to air, television journalists as a group increasingly must look for outlets, whether on air or online, wherever they can find them.  4 This strategy was not by any means unique to MSNBC.com, but has been a major component of the gameplan for many large media conglomerates and online companies. Boczkowski (2010) has similarly argued that the future of news may lie largely in niche editorial products.  5 In another example of provincial heterogeneous engineering with respect to online distribution, David Gregory, the former moderator of Meet the Press, for a time created his own site, DavidGregoryTV.com, dedicated partially to the show. It was phased out even before Gregory’s departure from Meet the Press, and much of the online activity it hosted was moved to the show’s Newsvine-powered blog.  6 Notably, the staff at NBC Nightly News had not tended to chafe at the show’s placement online. During my fieldwork, Cynthia Joyce, the MSNBC.com web producer assigned to the program, said, “I do think that the Nightly staffers initially saw the website as little more than a billboard with some extra bells and whistles—as opposed to a part of a larger distribution platform.” As we saw in detail in previous chapters, however, this was not so for the Rachel Maddow Show, which used various elements of heterogeneous engineering to aggressively challenge its placement within MSNBC .com’s branding scheme.  7 The journalism studies literature, including Tuchman (1978) and Hartley (1982), has long pointed out that “hard news” is a gendered actor’s category. CHAPTER 12  One Size Does Fit All

 1 These sorts of tradeoffs of stability for speed are a common engineering problem, often encapsulated in the “project triangle,” a diagram that prompts designers with the phrase “Fast, good, cheap; you may pick any two.”  2 Importantly, modularity of components is a common development strategy, not one unique to MSNBC.com.  3 Again, stepping back from organizational boundaries, these strategies aren’t unique to MSNBC. Much has been made of how the popular software tool WordPress, for instance, transitioned from a single-use blogging tool to an extraordinarily flexible content management system capable of powering sites that look nothing like blogs

NOTES TO PAGES 234–248  276

(Mullenweg 2009). Similarly, much of the initial success of popular platforms like Twitter and YouTube has been attributed to the fact that they were generic enough in their functionality to support uses that their designers had not initially envisioned (Ma 2009; Rao 2011). CHAPTER 13  Mass Reach with and without Mass Media

 1 In my interviews at Newsvine, which took place before this conversation, Davidson acknowledged this same issue and provided some of the same examples from MSNBC.com health stories. On the one hand, he said, there were some stories where the comments were “heartless” to the extent that “we feel bad about that as a company.” On the other hand, he also pointed to examples where the value of audacious user comments could arguably be seen as a matter of editorial outlook, such as a story in which the sales rep for a Botox provider apparently was used as a source and offered herself as an example of the procedure’s benefits. As Davidson recounted, “People in the comment threads were like, ‘Oh, my God. You look terrible. Women do not need to get plastic surgery to be beautiful. This is wrong.’ To me, I read that stuff and I’m proud of it. I think most women who read most of these comments would be proud of what our community was saying, which is, ‘Plastic surgery is something that the media has made you think that you really want. You would have looked better without it.’”  2 The number 150 is familiar to both anthropologists and those in management circles as an oft-given value of “Dunbar’s number,” a hypothetical limit to the number of individuals who can productively work together in a group. The term is named for anthropologist Robin Dunbar who first suggested the idea.  3 The developers anticipated that not all groups would be competitive—circles of college friends or co-workers, for instance, might form a group for its own sake without the intention of competing for public exposure.  4 A tiered commenting system modeled after Gawker’s discussion threads was another option being bandied about at MSNBC.com in 2010 as an alternative to Newsvine’s proposal (though either option would ultimately have been built on Newsvine’s technology).  5 Though Newsvine’s new architecture was ultimately rolled out after MSNBC.com had been acquired by NBCUniversal and transformed into the revamped NBC News Digital, much of Newsvine’s redesign was drafted prior to these changes and in discussion with the original MSNBC.com staff.  6 Conversely, I was told that some television programs that attracted older, less Internet-savvy audiences were more difficult to maintain an online presence for, given that audiences were less likely to look up the show’s content online, let alone engage with it in helpful or lucrative ways. CHAPTER 14  Conclusion

 1 For more on tracing a path through the physical infrastructures see, for example, Blum (2012), Parks (2012a), and Starosielski (2012). Also, it’s worth noting that the two forms of tracing I describe are not mutually exclusive within a single analysis.

NOTES TO PAGE 254  277

 2 While the metaphor of isometric pressure is novel in this context, it is consonant with previous analytical lenses from field theory and actor network theory, as well as with the notion of a “provisional settlement” employed by Boczkowski (2004) and by Girard and Stark (2002).

This page intentionally left blank

BIBLIOGRAPHY

Acland, C. R. 2003. Screen Traffic: Movies, Multiplexes, and Global Cultures. Durham, NC: Duke University Press. Action Engine. 2008. July 22. “Action Engine Reports Outstanding Growth in 2008 for Mobile Applications for Major Media Brands.” Press release retrieved from LexisNexis Academic Database. Adorno, T., and M. Horkheimer. 1991. “The Culture Industry: Enlightenment as Mass Deception.” In Horkheimer and Adorno, Dialectic of Enlightenment. New York: Routledge. Albrecht, C. 2009a. June 23. “Everything You Need to Know about TV Everywhere.” NewTeeVee. Retrieved Nov. 29, 2014, from http://gigaom .com/2009/06/23/what-you-need-to-know-about-tv-everywhere. ———. 2009b. Mar. 5. “Media Companies Plan Weapons of Mass Authentication.” NewTeeVee. Retrieved Nov. 28, 2014, from http://gigaom.com/ 2009/03/05/media-companies-plan-weapons-of-mass-authentication. Alstyne, M. van, and E. Brynjolfsson. 1996. “Electronic Communities: Global Village or Cyberbalkans.” In Proceedings of the 17th International Conference on Information Systems. Atlanta: Association for Information Systems. Althusser, L. A. 2011. “Ideology and Ideological State Apparatuses: Notes towards an Investigation.” Pp. 204–222 in I. Szeman and T. Kaposy, eds., Cultural Theory: An Anthology. Malden, MA: Wiley-Blackwell. Ananny, M. 2013. (1971). “Press-Public Collaboration as Infrastructure: Tracing News Organizations and Programming Publics in Application Programming Interfaces.” American Behavioral Scientist 57(5): 623–642. Anderson, B. 1991. Imagined Communities. Rev. and ext. ed. New York: Verso. Anderson, C. W. 2010. “Principles of Journalistic Symmetry: Building News Networks before and after the ‘Publication’ of News.” In International Symposium on Online Journalism. ———. 2011. “Deliberative, Agonistic, and Algorithmic Audiences: Journalism’s Vision of Its Public in an Age of Audience Transparency.” International Journal of Communication 5: 529–547. ———. 2013. Rebuilding the News: Metropolitan Journalism in the Digital Age. Philadelphia: Temple University Press. Arnason, H. H. 2004. A History of Modern Art: Painting, Sculpture, Architecture, Photography. 5th ed. Upper Saddle River, NJ: Prentice Hall. 279

BIBLIOGRAPHY 280

Arrington, M. 2007. Oct. 7. “Breaking: Newsvine Acquired by MSNBC.com.” TechCrunch. Retrieved Nov. 29, 2014, from http://techcrunch.com/2007/10/ 07/breaking-newsvine-acquired-by-msnbccom. Bagdikian, B. H. 2004. The New Media Monopoly. Rev. ed. Boston: Beacon Press. Balsamo, A. 2012. “I Phone, I Learn.” Pp. 251–264 in P. Snickars and P. Vonderau, eds., Moving Data: The iPhone and the Future of Media. New York: Columbia University Press. Bamford, J. 2008. The Shadow Factory. New York: Knopf Doubleday. Battaglio, S. 2008. Oct. 2. “Maddow Makes MSNBC Look Smart.” TV Guide. Retrieved Nov. 29, 2014, from http://www.tvguide.com/biz/Maddow-Makes -MSNBC-19196.aspx. Battelle, J. 2005. The Search: How Google and Its Rivals Rewrote the Rules of Business and Transformed Our Culture. New York: Portfolio. Bauder, D. 2011. Jan. 23. “Olbermann and MSNBC: A Failing Relationship.” Boston.Com. Retrieved Nov. 29, 2014, from http://www.boston.com/ae/tv/ articles/2011/01/23/olbermann_and_msnbc_a_failing_relationship. Bell, A. 1991. The Language of News Media. Cambridge, MA: Blackwell. Benkler, Y. 2006. The Wealth of Networks: How Social Production Transforms Markets and Freedom. New Haven: Yale University Press. Best, J. 1990. Threatened Children: Rhetoric and Concern about Child-Victims. Chicago: University of Chicago Press. Bijker, W. E. 2006. Why and How Technology Matters. Pp. 681–706 in R. E. Goodin and C. Tilly, eds., The Oxford Handbook of Contextual Political Analysis. New York: Oxford University Press. Bivens, R. 2014. Digital Currents: How Technology and the Public Are Shaping TV News. Toronto: University of Toronto Press. Black, L. W., H. T. Welser, J. M. DeGroot, and D. Cosley. 2008. “‘Wikipedia Is Not a Democracy’: Deliberation and Policy-Making in an Online Community.” Paper presented at the International Communication Association Annual Meeting in Montreal. Blum, A. 2012. Tubes: A Journey to the Center of the Internet. New York: Ecco. Boczkowski, P. J. 2004. Digitizing the News: Innovation in Online Newspapers. Cambridge, MA: MIT Press. ———. 2010. News at Work: Imitation in an Age of Information Abundance. Chicago: University of Chicago Press. Bogost, I. 2011, May 3. “Persuasive Games: Exploitationware.” Gamasutra. Retrieved Nov. 29, 2014, from http://www.gamasutra.com/view/feature/6366/ persuasive_games_exploitationware.php. Bourdieu, P. 1983. “The Field of Cultural Production, or: The Economic World Reversed.” Poetics 12:311–356.

BIBLIOGRAPHY 281

Bournique, D. 2006. Dec. 15. “MSNBC’s Hidden Mobile Gems.” WAP Review. Retrieved Nov. 29, 2014, from http://wapreview.com/229. boyd, d. 2007. “Why Youth (Heart. Social Network Sites: The Role of Networked Publics in Teenage Social Life.” Pp. 119-142 in D. Buckingham, ed., Youth, Identity, and Digital Media. Cambridge, MA: MIT Press. boyd, d. m., and N. B. Ellison. 2008. “Social Network Sites: Definition, History, and Scholarship.” Journal of Computer-Mediated Communication 13(1): 210–230. Boyle, J. 2008. The Public Domain: Enclosing the Commons of the Mind. New Haven: Yale University Press. Braun, J. A. 2013. “Going over the Top: Online Television Distribution as Sociotechnical System.” Communication, Culture and Critique 6(3): 432–458. ———. 2014. “Transparent Intermediaries: Building the Infrastructures of Connected Viewing.” In J. Holt and K. Sanson, eds., Connected Viewing: Selling, Streaming and Sharing Media in the Digital Era. New York: Routledge. BrightRoll Research. 2011. BrightRoll Video Advertising Report: Q1 2011. Market research report. Retrieved Nov. 29, 2014, from http://www.brightroll.com/_ research/industry-reports/video-advertising-reports. Brodkin, J. 2011. May 3. “Former MySQL Boss: Code ‘in Better Shape than Ever’ under Oracle.” Network World. Retrieved from LexisNexis Academic Database. Bruns, A. 2008. Blogs, Wikipedia, Second Life, and Beyond: From Production to Produsage. New York: Peter Lang. Bugos, G. 2000. “System Reshapes the Corporation: Joint Ventures in the Bay Area Rapid Transit System, 1962–1972.” Pp. 113–131 in A. C. Hughes and T. P. Hughes, eds., Systems, Experts, and Computers: The Systems Approach in Management and Engineering, World War II and After. Cambridge, MA: MIT Press. Burke, K. 1965. Permanence and Change: An Anatomy of Purpose. 2d rev. ed. New York: Bobbs Merrill. Butcher, D. 2009. Mar. 19. “MSNBC’s Rachel Maddow Launches iPhone App.” Mobile Marketer. Retrieved Nov. 29, 2014, from http://www.mobilemarketer .com/cms/news/television/3280.html. Callon, M. 1986. “Some Elements of a Sociology of Translation: Domestication of the Scallops and the Fishermen of St Brieuc Bay.” Pp. 196–233 in J. Law, ed., Power, Action and Belief. Boston: Routledge and Kegan Paul. ———. 1998. “An Essay on Framing and Overflowing: Economic Externalities Revisited by Sociology.” Pp. 244–269 in M. Callon, ed., The Laws of the Markets. Oxford, Eng.: Blackwell. Carey, J. 1989. “A Cultural Approach to Communication.” Pp. 11–28 in Carey,

BIBLIOGRAPHY 282

Communication as Culture: Essays on Media and Society. Boston: Unwin Hyman. Carter, B. 2002. Aug. 19. “Woes of Donahue Cast Shadow over MSNBC.” New York Times, C1. Retrieved from LexisNexis Academic Database. ———. 2004. Apr. 30. “Some Stations to Block Nightline War Tribute.” New York Times. Retrieved Nov. 29, 2014, from http://www.nytimes.com/2004/04/30/ business/media/30TUBE.html. ———. 2005. Dec. 24. “Microsoft Quits MSNBC TV, But Web Partnership Remains.” New York Times. Retrieved Nov. 29, 2014, from http://www.nytimes .com/2005/12/24/business/media/24msnbc.html. ———. 2011. Jan. 21. “Olbermann Hosts Last Countdown on MSNBC.” New York Times. Retrieved Nov. 29, 2014, from http://mediadecoder.blogs.ny times.com/2011/01/21/olbermann-hosts-last-countdown-on-msnbc. Chaffee, S. H., and M. J. Metzger. 2001. “The End of Mass Communication?” Mass Communication and Society 4(4): 365–379. Chamberlain, D. 2010. “Television Interfaces.” Journal of Popular Film and Television 38(2): 84–88. Chia, R. 1995. “From Modern to Postmodern Organizational Analysis.” Organization Studies 16 (4): 579–604. Child, J., and R. G. McGrath. 2001. “Organizations Unfettered: Organizational Form in an Information-Intensive Economy.” Academy of Management Journal 44(6): 1135–1148. Choney, S. 2011. Mar. 30. “‘Royal Wedding’ App Makes You Feel Like You’re There.” Digital Life. Retrieved Nov. 29, 2014, from http://www.today.com/ tech/royal-wedding-app-makes-you-feel-youre-there-1C8368679. Clayborne, C., ed. 1998. The Autobiography of Martin Luther King, Jr. New York: Warner Books. Cohen, D. 2011a. “Experian Hitwise: Rachel Maddow Bests Bill O’Reilly.” Social Times. Retrieved Nov. 29, 2014, from http://socialtimes.com/experian-hit wise-rachel-maddow-bests-bill-o%E2%80%99reilly_b13031. ———. 2011b. “Experian Hitwise: Rachel Maddow Rebounds.” TVNewser. Retrieved Nov. 29, 2014, from http://www.mediabistro.com/tvnewser/experian -hitwise-rachel-maddow-rebounds_b69663. Coleman, G. 2010. “Ethnographic Approaches to Digital Media.” Annual Review of Anthropology, 39: 487–505. ———. 2013. Coding Freedom: The Ethics and Aesthetics of Hacking. Princeton, NJ: Princeton University Press. Collins, S. 2004. Crazy Like a Fox: The Inside Story of How Fox News Beat CNN. New York: Portfolio. comScore. 2011, Apr. 12. “ComScore Releases Mar. 2011 U.S. Online Video Rankings: Google Sites Nears 2 Billion Content Viewing Sessions.” Press

BIBLIOGRAPHY 283

release, Comscore.com. Retrieved Jan. 28, 2015, from http://www.comscore .com/Insights/Press-Releases/2011/4/comScore-Releases-March-2011-US -Online-Video-Rankings. Cook, J. 2005. Nov. 11. “Seattle Startup to Offer Online News, Etc.” Seattle Post-Intelligencer, D1. Retrieved from LexisNexis Academic Database. ———. 2006. Dec. 11. “New York Times Adds Sharing Tool.” Seattle Post-­ Intelligencer, C1. Retrieved from LexisNexis Academic Database. Cottle, S. 2007. “Ethnography and News Production: New(s) Developments in the Field.” Sociology Compass 1(1): 1–16. Couldry, N. 2008. “Actor Network Theory and Media: Do They Connect and on What Terms?” Pp. 93–109 in A. Hepp, F. Krotz, S. Moores, and C. Winter, eds., Connectivity, Networks and Flows. Cresskill, NJ: Hampton Press. Couric, K. 2011. May 9. “Rachel Maddow Tells Katie Couric, “I’m Not Competing on the Pretty-Girl-on-Cable Front.” Glamour. Retrieved Nov. 29, 2014, from http://www.glamour.com/magazine/2011/05/rachel-maddow-tells -katie-couric-im-not-competing-on-the-pretty-girl-on-cable-front/1. Croal, N. 2000. Dec. 4. “2001: A Tech Odyssey.” Newsweek, 48. Retrieved from LexisNexis Academic Database. Cronin, K. 2009. Dec. 11. “Brightcove Facebook Whitelisting in Action with Thomas Pink and Adjust Your Set. Brightcove Blog. Retrieved Nov. 30, 2014, from http://blog.brightcove.com/en/2009/12/brightcove-facebook-white listing-action-thomas-pink-and-adjust-your-set. Davenport, C. 2010. Mar. 4. “Blowback: Do Comments Scare Off Sources?” Washington Post. Retrieved Nov. 30, 2014, from http://blog.washingtonpost .com/story-lab/2010/03/blowback_commenting_on_debt_co.html. Davies, N. 2008. Flat Earth News. London: Chatto and Windus. Dawson, M. and L. Spigel. 2008. “Television and Digital Media.” Pp. 275–289 in C. Morley and M. Halliwell, eds., American Thought and Culture in the Twenty-First Century. New York: Columbia University Press. Deuze, M. 2007. Media Work. Cambridge, UK: Polity Press. Diakopoulos, N. 2014. “Algorithmic Accountability: Journalistic Investigation of Computational Power Structures.” Digital Journalism. Retrieved Feb. 2, 2015, at http://www.tandfonline.com/doi/abs/10.1080/21670811.2014.9764 11#preview. Dimitrova, D. V., C. Connolly-Ahern, A. P. Williams, L. L. Kaid, and A. Reid. 2003. “Hyperlinking as Gatekeeping: Online Newspaper Coverage of the Execution of an American Terrorist.” Journalism Studies 4(3): 401–414. Domingo, D. 2008. “Interactivity in the Daily Routines of Online Newsrooms: Dealing with an Uncomfortable Myth.” Journal of Computer-Mediated Communication 13(3): 680–704.

BIBLIOGRAPHY 284

Downey, G. 2001. “Virtual Webs, Physical Technologies, and Hidden Workers: The Spaces of Labor in Information Internetworks.” Technology and Culture 42(2): 209–235. Duneier, M. 1999. Sidewalk. New York: Farrar, Straus and Giroux. Duryee, T. 2008. Sept. 7. “Action Engine Lays Off Employees; May Sell or Shut Down.” GigaOm. Retrieved Nov. 30, 2014, from http://gigaom.com/2008/ 09/07/419-action-engine-lays-off-employees-may-sell-or-shut-down. Dwyer, T. 2010. Media Convergence. New York: Open University Press. Epstein, S. 1996. Impure Science: AIDS, Activism, and the Politics of Knowledge. Berkeley: University of California Press. Ernesto. 2009. “Sports Broadcasters Go after Live Streaming Pirates.” TorrentFreak. Retrieved Nov. 30, 2014, from http://torrentfreak.com/sports-broad casters-go-after-live-streaming-pirates-090406. Farhi, P. 2011. Jan. 22. “Keith Olbermann and MSNBC Part Ways.” Washington Post. Retrieved Nov. 30, 2014, from http://www.washingtonpost.com/wp -dyn/content/article/2011/01/21/AR2011012106861.html. Fast-Talk Communications. 2002. June 18. “Fast-Talk Invited to Exhibit on Capitol Hill at Exclusive Small Business Homeland Security Expo.” Press release retrieved from LexisNexis Academic Database. Femia, W. 2014. June 27. “Tonight! Hubris: Selling the Iraq War.” Maddow Blog. Retrieved Nov. 30, 2014, from http://www.msnbc.com/rachel-maddow -show/tonight-hubris-selling-the-iraq-war-0. Fenton, N. 2010. “News in the Digital Age.” Pp. 557–567 in S. Allan, ed., The Routledge Companion to News and Journalism. New York: Routledge. Ferenstein, G. 2011. Mar. 16. “Welcome to CNN’s New Digital Playground: TV Everywhere!” Fast Company. Retrieved Nov. 30, 2014, from http://www .fastcompany.com/1739004/welcome-cnns-new-digital-playground-tv -everywhere. Flanagan, M., D. C. Howe, and H. Nissenbaum. 2008. “Embodying Values in Technology: Theory and Practice.” Pp. 322–353 in J. van den Hoven and J. Weckert, eds., Information Technology and Moral Philosophy. Cambridge, Eng.: Cambridge University Press. Foremski, T. 2010. May 26. “MediaWatch: Journalists Won’t Report News Unless It Can Drive Page Views.” SiliconValleyWatcher. Retrieved Nov. 30, 2014, from http://www.siliconvalleywatcher.com/mt/archives/2010/05/media watch_mond_7.php. Franco, V., R. Piirto, H. Y. Hu, and B. V. Lewenstein. 1995. “Anatomy of a Flame: Conflict and Community Building on the Internet.” IEEE Technology and Society Magazine 14(2): 12–21. Friedl, K. 2006. Nov. 27. “Does ‘Crowdsourcing’ Exploit Net Users? Some Say

BIBLIOGRAPHY 285

Yes.” Newassignment.Net. Retrieved Nov. 30, 2014, from https://web.archive .org/web/20070305211541/http://www.newassignment.net/blog/kevin_ friedl/nov2006/10/a_mob_by_any_oth. Fuchs, C. 2009. “Information and Communication Technologies and Society: A Contribution to the Critique of the Political Economy of the Internet.” European Journal of Communication 24(1): 69–87. Furman, P. 2002. Aug. 14. “Donahue Ratings Grim for MSNBC.” Daily News, 52. Retrieved from LexisNexis Academic Database. Galison, P. 1997. Image and Logic: A Material Culture of Microphysics. Chicago: University of Chicago Press. Galtung, J., and M. H. Ruge. 1965. “The Structure of Foreign News: The Presentation of the Congo, Cuba and Cyprus Crises in Four Norwegian Newspapers.” Journal of Peace Research 2(1): 64–91. Gamson, W. A., and A. Modigliani. 1989. “Media Discourse and Public Opinion on Nuclear Power: A Constructionist Approach.” American Journal of Sociology 95(1): 1–37. Gans, H. 1980. Deciding What’s News: A Study of “CBS Evening News,” “NBC Nightly News,” “Newsweek,” and “Time.” New York: Vintage Books. Garber, M. 2011. Mar. 30. “So, Then . . . if You Jump the New York Times’ Paywall, Are You Stealing? Nieman Journalism Lab. Retrieved Dec. 1, 2014, from http://www.niemanlab.org/2011/03/so-then-if-you-jump-the-new-york -times-paywall-are-you-stealing. Gay, G., and H. Hembrooke. 2004. “Understanding Perspectives: Social Construction of Technology.” Pp. 15–29 in Activity-Centered Design: An Ecological Approach to Designing Smart Tools and Usable Systems. Cambridge, MA: MIT Press. Geiger, S. 2007. “There Is No Cabal: An Investigation into Wikipedia’s Legal Subculture.” In Technically Human. Retrieved Feb. 1, 2015 from http:// www.stuartgeiger.com/wordpress/academic-works/2007/05/31/ there-is-no-cabal-an-investigation-into-wikipedias-legal-subculture. ———. 2011. “The Lives of Bots.” Pp. 78–93 in G. Lovink and N. Tkacz, eds., A Critical Point of View: A Wikipedia Reader. Amsterdam: Institute of Network Cultures. Giddens, A. 1984. The Constitution of Society: Outline of the Theory of Structuration. Berkeley: University of California Press. Gillespie, T. 2007. Wired Shut: Copyright and the Shape of Digital Culture. Cambridge, MA: MIT Press. ———. 2010. “The Politics of ‘Platforms.’” New Media and Society 12(3): 347–364. ———. 2014a. July 4. “Facebook’s Algorithm—Why Our Assumptions Are Wrong, and Our Concerns Are Right. Culture Digitally. Retrieved Dec. 1,

BIBLIOGRAPHY 286

2014, from http://culturedigitally.org/2014/07/facebooks-algorithm-why -our-assumptions-are-wrong-and-our-concerns-are-right. ———. 2014b. “The Relevance of Algorithms.” Pp. 167–193 in K. A. Foot, P. J. Boczkowski, and T. Gillespie, eds., Media Technologies: Essays on Communication, Materiality, and Society. Cambridge, MA: MIT Press. Gillette, F. 2010. May 7. “Rachel Maddow Touts the ‘The Maddow Blog.’” New York Observer. Retrieved Dec. 1, 2014, from http://observer.com/2010/05/ rachel-maddow-touts-the-the-maddow-blog. Girard, M., and D. Stark. 2002. “Distributing Intelligence and Organizing Diversity in New Media Projects.” Environment and Planning A 34(11): 1927–1949. Gitlin, T. 2003. The Whole World Is Watching: Mass Media in the Making and Unmaking of the New Left. 2003 ed. Berkeley: University of California Press. Glaser, M. 2010. Mar. 22. “Why ‘TV Everywhere’ Will Fail.” MediaShift. Retrieved Dec. 1, 2014, from http://www.pbs.org/mediashift/2010/03/why-tv-every where-will-fail081. Goffman, E. 1974. Frame Analysis: An Essay on the Organization of Experience. Cambridge, MA: Harvard University Press. Goggin, G. 2011. “The Intimate Turn of Mobile News.” Pp. 99–114 in G. Meikle and G. Redden, eds., News Online: Transformations and Continuities. New York: Palgrave Macmillan. Gold, M. 2010a. Feb. 24. “ABC News Plans 25% Cutback.” Los Angeles Times. Retrieved Dec. 1, 2014, from http://articles.latimes.com/2010/feb/24/ entertainment/la-et-abc-news24-2010feb24. ———. 2010b. Feb. 26. “ABC News to Cut Half Its Domestic Correspondents, Close Bricks-and-Mortar Bureaus.” Los Angeles Times. Retrieved Dec. 1, 2014, from http://latimesblogs.latimes.com/entertainmentnewsbuzz/2010/ 02/abc-news-to-cut-half-its-domestic-correspondents-shut-down-all -bureaus-except-washington.html. Gordon, J. 2007. The Mobile Phone and the Public Sphere: Mobile Phone Usage in Three Critical Situations. Convergence 13(3): 307–319. Gough, P. J. 2005. May 25. “Podcasts Catch on at NBC, ABC News Divisions.” Hollywood Reporter. Retrieved from LexisNexis Academic Database. Grabher, G. 2002. “The Project Ecology of Advertising: Tasks, Talents, and Teams.” Regional Studies 36(3): 245–262. Gray, L. 2010. Mar. 14. “Gmail Failures, Crazy Ideas and Wave’s Leapfrog.” Louis Gray.com. Retrieved Dec. 1, 2014, from http://blog.louisgray.com/2010/03/ gmail-failures-crazy-ideas-and-waves.html. Halavais, A. 2009. Search Engine Society. Cambridge, Eng.: Polity Press. Hall, S. 1973. “The Determinations of News Photographs.” Pp. 176–190 in

BIBLIOGRAPHY 287

S. Cohen and J. Young, eds., The Manufacture of News: Social Problems, Deviance and the Mass Media. London: Constable. Hall, S., C. Crichter, T. Jefferson, J. Clarke, and B. Roberts. 1978. Policing the Crisis: Mugging, the State, and Law and Order. New York: Macmillan. Harper, C. 1998. And That’s the Way It Will Be: News and Information in a Digital World. New York: NYU Press. Hartley, J. 1982. “Selection and Construction.” Pp. 75–86 in Understanding News. New York: Methuen. Harvey, D. 1989. The Condition of Postmodernity: An Enquiry into the Origins of Cultural Change. London: Blackwell. Haythornthwaite, C. 2007. Social Networks and Online Community. Pp. 1–11 in A. N. Joinson, K. Y. A. McKenna, T. Postmes, and U.-D. Reips, eds., The Oxford Handbook of Internet Psychology. New York: Oxford University Press. Hemmingway, E. 2005. “PDP, the News Production Network and the Transformation of News.” Convergence 11(3): 8–27. ———. 2008. Into the Newsroom: Exploring the Digital Production of Regional Television News. New York: Routledge. Hermida, A., and N. Thurman. 2008. “A Clash of Cultures: The Integration of User-Generated Content within Professional Journalistic Frameworks at British Newspaper Websites.” Journalism Practice 2(3): 343–356. Hess, C., and E. Ostrom. 2003. “Ideas, Artifacts, and Facilities: Information as a Common-Pool Resource.” Law and Contemporary Problems 6(1/2): 111–146. Hess, K. 2013. “Tertius Tactics: ‘Mediated Social Capital’ as a Resource of Power for Traditional Commercial News Media.” Communication Theory 23(2): 112–130. Hetherington, A. 1985. News, Newspapers and Television. London: Macmillan. Heuman, J. 2006. “Configuring the Viewer in Transition: Communication Policy and the Television Viewer between ‘Old’ and ‘New’ Media.” Ph.D. diss. University of Wisconsin-Madison. Hilgartner, S., and C. L. Bosk. 1988. “The Rise and Fall of Social Problems: A Public Arenas Model.” American Journal of Sociology 94(1): 53–78. Hippel, E. von. 2005. Democratizing Innovation. Cambridge, MA: MIT Press. Holt, J. 2011. Empires of Entertainment: Media Industries and the Politics of Deregulation, 1980–1996. New Brunswick, NJ: Rutgers University Press. Holt, J., and K. Sanson, eds. 2013. Connected Viewing: Selling, Streaming, and Sharing Media in the Digital Age. New York: Routledge. Hood, M. 2008. Oct. 13. “Rachel Maddow to Redmond for Web Wisdom.” BlatherWatch. Retrieved Dec. 1, 2014, from http://blatherwatch.blogs.com/ talk_radio/2008/10/rachel-maddow-t.html. Hopkins, H. 2010. Feb. 3. “Facebook Largest News Reader?” Experian Hitwise. Re-

BIBLIOGRAPHY 288

trieved May 30, 2011, from http://weblogs.hitwise.com/us-heather-hopkins/ 2010/02/facebook_largest_news_reader_1.html. Hughes, T. P. 1979. “The Electrification of America: The System Builders.” Technology and Culture 20(1): 124–161. ———. 1983. Networks of Power: Electrification in Western Society, 1880–1930. Baltimore: Johns Hopkins University Press. ———. 1987. “The Evolutions of Large Technological Systems.” Pp. 51–82 in W. E. Bijker, T. P. Hughes, and T. J. Pinch, The Social Construction of Technological Systems: New Directions in the Sociology and History of Technology. Cambridge, MA: MIT Press. ———. 1998. Rescuing Prometheus. New York: Pantheon. ———. 2004. Human-Built World: How to Think about Technology and Culture. Chicago: University of Chicago Press. Hutchins, E. 1995. Cognition in the Wild. Cambridge, MA: MIT Press. ———. 2003. “How a Cockpit Remembers Its Speeds.” Cognitive Science 19(3): 265–288. Hutchins, E., and T. Klausen. 1996. “Distributed Cognition in an Airline Cockpit.” Pp. 15–34 in Y. Engestöm and D. Middleton, eds., Cognition and Communication at Work. New York: Cambridge University Press. Introna, L. D., and H. Nissenbaum. 2000. “Shaping the Web: Why the Politics of Search Engines Matters.” Information Society 16(3): 169–185. Jackson, M. H., M. S. Poole, and T. Kuhn. 2002. “The Social Construction of Technology in Studies of the Workplace.” Pp. 236–253 in L. A. Lievrouw and S. M. Livingstone, eds., Handbook of New Media: Social Shaping and Consequences of ICTs. Thousand Oaks, CA: Sage Publications. James, M. 2011. Apr. 29. “Viacom Profit Soars 53% on Paramount Pictures Revenue.” Los Angeles Times. Retrieved Dec. 2, 2014, from http://articles.latimes .com/2011/apr/29/business/la-fi-ct-viacom-20110429. Jenkins, H., S. Ford, and J. Green. 2013. Spreadable Media: Creating Value and Meaning in a Networked Culture. New York: New York University Press. Jobs, S. 2010. Apr. “Thoughts on Flash.” Apple. Retrieved Dec. 2, 2014, from http://www.apple.com/hotnews/thoughts-on-flash. John, R. 1995. Spreading the News: The American Postal System from Franklin to Morse. Cambridge, MA: Harvard University Press. Johnson, A. 2007. Oct. 8. “Msnbc.com Acquires Newsvine.” MSNBC.Com. Retrieved Dec. 2, 2014, from http://www.nbcnews.com/id/21138371/ns/busi ness-us_business/t/msnbccom-buys-social-news-site-newsvine. Johnson, S. 2005. Everything Bad Is Good for You: How Today’s Pop Culture Is Actually Making Us Smarter. Pp. 17–62. New York: Berkley Publishing Group. Joyella, M. 2011a. Mar. 25. “NBC’s Dateline: Using Subtle Social Media to Turn Stories into Trending Topics.” Mediaite. Retrieved Dec. 3, 2014, from http://

BIBLIOGRAPHY 289

www.mediaite.com/online/nbcs-dateline-using-subtle-social-media-to-turn -stories-into-trending-topics. ———. 2011b. May 10. “Traffic Data: Cable News, Led by CNN, Crushing Newspapers on the Web.” Mediaite. Retrieved Dec. 3, 2014, from http:// www.mediaite.com/online/traffic-data-cable-news-led-by-cnn-crushing -newspapers-on-the-web. Kaplan, D. 2010. Oct. 6. “Tired of Facing ‘Brand Insanity,’ MSNBC.com Considers Name Change.” PaidContent.Org. Retrieved Dec. 3, 2014, from http://gigaom.com/2010/10/06/419-tired-of-facing-brand-insanity -msnbc-com-considers-name-change. Kay, R. 2006, Sept. 21. “LAMP.” ComputerWorld. Retrieved from LexisNexis Academic Database. Kellogg, K. C., W. J. Orlikowski, and J. Yates. 2006. “Life in the Trading Zone: Structuring Coordination across Boundaries in Postbureaucratic Organizations.” Organization Science 17(1): 22–44. Kerner, L. 2010. “Demand Media Will Be the First $1 Billion Tech IPO since Google—Here’s Why.” San Francisco Chronicle. Retrieved Dec. 3, 2014, from http://www.sfgate.com/news/article/Demand-Media-Will-Be-The-First-1 -Billion-Tech-2464047.php. Kim, Y.-C., and S. J. Ball-Rokeach. 2006. “Civic Engagement from a Communication Infrastructure Perspective.” Communication Theory 16(2): 173–197. Kirkland, S. 2014, Mar. 14. “NBC News Engagement Up since Site Relaunch Despite Reader Complaints.” Poynter. Retrieved Dec. 3, 2014, from http://www .poynter.org/latest-news/media-lab/243501/nbc-news-engagement-upsince-site-relaunch-despite-reader-complaints. Kittur, A., B. Suh, B. A. Pendleton, and E. H. Chi. 2007. “He Says, She Says: Conflict and Coordination in Wikipedia.” Paper presented at the annual ACM Conference on Human Factors in Computing Systems, San Jose, CA. Kline, R., and T. J. Pinch. 1996. “Users as Agents of Technological Change: The Social Construction of the Automobile in the Rural United States.” Technology and Culture 37(4): 763–795. Knorr Cetina, K. 1997. “Sociality with Objects: Social Relations in Postsocial Knowledge Societies.” Theory, Culture and Society 14(4): 1–30. ———. 1999. Epistemic Cultures: How the Sciences Make Knowledge. Cambridge, MA: Harvard University Press. ———. 2001. “Objectual Practice.” Pp. 175–188 in T. R. Schatzki, K. Knorr Cetina, and E. von Savigny, eds., The Practice Turn in Contemporary Theory. New York: Routledge. Kokas, A. 2014. Shot in Shanghai: Blockbusters, Social Networks, and Sino-U.S. Media Convergence. Unpub. manuscript.

BIBLIOGRAPHY 290

Kompare, D. 2010. “Reruns 2.0: Revising Repetition for Multiplatform Television Distribution.” Journal of Popular Film and Television 38(2): 79–83. Kopytoff, V. 2006. Apr. 17. “Populist News Sites Give Readers What They Want.” San Francisco Chronicle, E1. Retrieved from LexisNexis Academic Database. Kravets, D. 2011. Feb. 2. “Feds Tackle Sports-Streaming Pirate Sites.” Threat Level. Retrieved Dec. 3, 2014, from http://www.wired.com/2011/02/feds -tackle-sports-piracy. Kurtz, H. 2011. Feb. 27. “Rachel Maddow Seizes Her Moment at Post-Keith Olbermann MSNBC.” Daily Beast. Retrieved Dec. 3, 2014, from http:// www.thedailybeast.com/articles/2011/02/27/rachel-maddow-seizes-her -moment-at-post-keith-olbermann-msnbc.html?cid=hp:mainpromo7. Lafayette, J. 2005. Dec. 12. “ABC’s ‘World News’ Going Multimedia; Net Rolls Out Online Expansion Plans. Television Week, 8. Retrieved from LexisNexis Academic Database. ———. 2011. Apr. 25. “MSNBC Leans on Spike Lee: Celebrated Filmmaker Shoots New Promo Campaign.” Broadcasting and Cable. Retrieved Dec. 3, 2014, from http://www.broadcastingcable.com/news/advertising-and-mar keting/msnbc-leans-spike-lee/52593. Langberg, M. 2004a. Feb. 23. “Oh No! Live TV on Cell Phones.” San Jose Mercury News, 1E. Retrieved from LexisNexis Academic Database. ———. 2004b. Feb. 26. “TV Almost Too Small to See.” San Jose Mercury News. Retrieved from LexisNexis Academic Database. Larkin, B. 2008. Signal and Noise: Media, Infrastructure, and Urban Culture in Nigeria. Durham, NC: Duke University Press. Latour, B. 1987. Science in Action: How to Follow Scientists and Engineers through Society. Cambridge, MA: Harvard University Press. ———. 1988. “Mixing Humans and Nonhumans Together: The Sociology of a Door-Closer.” Social Problems 35(3): 298–310. ———. 1990. “Drawing Things Together.” Pp. 19–68 in M. Lynch and S. Woolgar, eds., Representation in Scientific Practice. Cambridge, MA: MIT Press. ———. 1993. We Have Never Been Modern. Cambridge, MA: Harvard University Press. ———. 1996. Aramis; or, The Love of Technology. Cambridge, MA: Harvard University Press. ———. 2002. “Morality and Technology: The End of the Means.” Theory, Culture and Society 19(5–6): 247–260. Latour, B., and S. Woolgar. 1986. Laboratory Life: The Social Construction of Scientific Facts, 2d ed. Princeton, NJ: Princeton University Press. Law, J. 1986. On the Methods of Long-Distance Control. Pp. 234–263 in J. Law, ed., Power, Action and Belief. Boston: Routledge and Kegan Paul. ———. 1987. “Technology and Heterogeneous Engineering: The Case of Portu-

BIBLIOGRAPHY 291

guese Expansion.” Pp. 111–134 in W. E. Bijker, T. P. Hughes, and T. J. Pinch, eds., The Social Construction of Technological Systems. Cambridge, MA: MIT Press. ———. 1992. “Notes on the Theory of the Actor-Network: Ordering, Strategy, and Heterogeneity.” Systemic Practice and Action Research 5(4): 379–393. ———. 2002. “On Hidden Heterogeneities: Complexity, Formalism, and Aircraft Design.” Pp. 116–141 in J. Law and A. Mol, eds., Complexities: Social Studies of Knowledge Practices. Durham, NC: Duke University Press. ———. 2007. “Making a Mess with Method.” Pp. 595–606 in W. Outhwaite and S. P. Turner, eds., The Sage Handbook of Social Science Methodology. Los Angeles: Sage Publications. ———. 2014. Aug. 24. “Research Approach.” Heterogeneities.net. Retrieved Feb. 1, 2015, from http://heterogeneities.net/approach.htm. Law, J., and M. Callon. 1988. “Engineering and Sociology in a Military Aircraft Project: A Network Analysis of Technological Change.” Social Problems 35(3): 284–297. ———. 1992. “The Life and Death of an Aircraft.” Pp. 21–52 in Shaping Technology/Building Society. Cambridge, MA: MIT Press. Law, J., and A. Mol. 2002. “Local Entanglements or Utopian Moves: An Inquiry into Train Accidents.” Pp. 82–105 in M. Parker, ed., Utopia and Organization. Oxford, Eng.: Blackwell. Lawler, R. 2011a. Apr. 8. “Cable’s Big iPad Fight: It’s All about Control. NewTeeVee. Retrieved Dec. 4, 2014, from http://gigaom.com/2011/04/08/cable -ipad-app-fight. ———. 2011b. Jan. 28. “TV Everywhere Is Hulu’s Best Chance for Survival.” NewTeeVee. Retrieved Dec. 4, 2014, from http://gigaom.com/2011/01/28/ hulu-tv-everywhere. Learmonth, M. 2009. Nov. 9. “Could Microsoft Exit Media Biz? Giant Increasingly Puts Bucks into Ad Systems Rather Than Ad Content.” Advertising Age. Retrieved from LexisNexis Academic Database. Lessig, L. 2004. Free Culture: How Big Media Uses Technology and the Law to Lock Down Culture and Control Creativity. New York: Penguin Press. ———. 2006. Code: Version 2.0. New York: Basic Books. Lewin, K. 1951. Psychological Ecology. Pp. 170–187 in D. Cartwright, ed., Field Theory in Social Science. New York: Harper. Lewis, P. H. 1996. “Microsoft Offering News without Charge on Internet.” New York Times, D4. Retrieved from LexisNexis Academic Database. Lieberman, A. 2001. June 9. “Ballmer Wouldn’t Do MSNBC Again.” New York Post, 19. Retrieved from LexisNexis Academic Database. Lobato, R. 2012. Shadow Economies of Cinema: Mapping Informal Film Distribution. London: British Film Institute.

BIBLIOGRAPHY 292

Lotz, A. D. 2007. The Television Will Be Revolutionized. New York: NYU Press. Lowrey, W. 2006. “Mapping the Journalism-Blogging Relationship.” Journalism 7(4): 477–500. Ma, O. 2009. Dec. 29. “Ordinary Citizens, Extraordinary Videos.” Broadcasting Ourselves: The Official YouTube Blog. Retrieved Dec. 5, 2014, from http:// youtube-global.blogspot.com/2009/12/ordinary-citizens-extraordinary -videos.html. MacGregor, P. 2007. “Tracking the Online Audience.” Journalism Studies 8(2): 280–298. MacManus, R. 2011. Oct. 7. “Newsvine Acquired by MSNBC: Leading Citizen Journalism Site Snapped Up by MSM.” ReadWriteWeb. Retrieved Dec. 5, 2014, from http://readwrite.com/2007/10/07/newsvine_acquired_by_ msnbc. Malaby, T. M. 2009. Making Virtual Worlds: Linden Lab and Second Life. Ithaca, NY: Cornell University Press. Markoff, J. 1997. Apr. 7. “Microsoft to Buy WebTV, Blending PC’s, TV’s, and the Internet.” New York Times, D1. Retrieved from LexisNexis Academic Database. Marlowe, C. 2001. June 4. “Roamable Inks MTV, MSNBC.” Hollywood Reporter. Retrieved from Highbeam Business Database. Marres, N. 2012. Material Participation: Technology, the Environment and Everyday Publics. New York: Palgrave-Macmillan. Mayer, V., M. J. Banks, and J. T. Caldwell, eds. 2009. Production Studies: Cultural Studies of Media Industries. New York: Routledge. McMillian, J. 2011. Smoking Typewriters: The Sixties Underground Press and the Rise of Alternative Media in America. New York: Oxford University Press. Mills, C. W. 1959. The Sociological Imagination. New York: Oxford University Press. Mitchelstein, E., and P. J. Boczkowski. 2009. “Between Tradition and Change: A Review of Recent Research on Online News Production.” Journalism 10(5): 562–586. ———. 2010. “Online News Consumption Research: An Assessment of Past Work and an Agenda for the Future.” New Media and Society 12(7): 1085– 1102. Moraes, L. de. 2002 Apr. 4. “Donahue to Be Opposite O’Reilly, in More Ways Than One.” Washington Post, C7. Retrieved from LexisNexis Academic Database. Mould, O. 2009. “Lights, Camera, But Where’s the Action?” Pp. 203–213 in V. Mayer, M. J. Banks, and J. T. Caldwell, eds., Production Studies. New York: Routledge. MSNBC. 2013. Oct. 29. “MSNBC Launches MSNBC.com.” Press release.

BIBLIOGRAPHY 293

TVNewser. Retrieved Dec. 5, 2014, from http://www.mediabistro.com/ tvnewser/the-new-msnbc-com-is-now-live_b201712. “MSNBC Goes Mobile with News.” 2006. Apr. 5. TechWeb. Retrieved from Lexis­ Nexis Academic Database. “MSNBC Mobile Views Up 3x over 4 Months.” 2009. Aug. 3. Wireless and Mobile News. Retrieved from LexisNexis Academic Database. MSNBC.com. 2001. June 26. “MSNBC.com Launches ‘MSNBC Now’ Delivering Wireless News and Headlines Using Roamable Technology.” Press release. MSNBC.com. Retrieved from LexisNexis Academic Database. ———. 2010. “The MSNBC Digital Network Acquires www.breakingnews.com: Overlooked Breaking News Consumers Have a Place to Turn.” Press release. Retrieved Dec. 5, 2014, from http://www.nbcnews.com/id/34702956/ns/ about-press_releases/#.U86vXKjdlY4. ———. 2011. “Rachel Maddow: Host, ‘The Rachel Maddow Show.’” Corporate biography. Retrieved Dec. 5, 2014, from http://www.msnbc.com/the-rachel -maddow-show/rachel-maddow-biography. Mullenweg, M. 2009. Nov. 18. “WordPress Wins CMS Award.” WordPress.org. Retrieved Dec. 5, 2014, from http://wordpress.org/news/2009/11/word press-wins-cms-award. Muller, J. 2005. Aug. 30. “A Mixed Bag in Williams’ Blog.” Radio commentary. Morning Edition. Washington, D.C.: National Public Radio. Retrieved Dec. 5, 2014, from http://www.npr.org/templates/story/story.php?storyId=4823307. Napoli, P. M. 2011. Audience Evolution: New Technologies and the Transformation of Media Audiences. New York: Columbia University Press. National Telecommunications Information Administration. 2011. Digital Nation: Expanding Internet Usage. Government report. Retrieved Dec. 5, 2014, from http://www.ntia.doc.gov/files/ntia/publications/ntia_internet_use_ report_Feb._2011.pdf. Nayyar, V. 2011. Mar. 16. “How to Set Default Homepage in Windows 7 Internet Explorer?” PCTips3000. Retrieved Dec. 5, 2014, from http://www.pctips 3000.com/how-to-set-default-homepage-in-windows-7-internet-explorer. Neff, G. 2012. Venture Labor: Work and the Burden of Risk in Innovative Industries. Cambridge, MA: MIT Press. Neff, G., and D. Stark. 2004. “Permanently Beta: Responsive Organization in the Internet Era.” Pp. 173–188 in P. N. Howard and S. Jones, eds., Society Online: The Internet in Context. Thousand Oaks, CA: Sage Publications. Negroponte, N. 1995. Being Digital. New York: Knopf. Newman, M. Z. 2012. “Free TV: File-Sharing and the Value of Television.” Television and New Media 13(6): 463–479. Newsvine. 2005. Dec. 2. “Welcome to the Newsvine Private Beta.” Newsvine Blog.

BIBLIOGRAPHY 294

Retrieved Dec. 5, 2014, from http://blog.newsvine.com/_news/2005/12/01 /30417-welcome-to-the-newsvine-private-beta. ———. 2006a. Mar. 10. “Newest Additions to Newsvine.” Newsvine Blog. Retrieved Dec. 5, 2014, from http://blog.newsvine.com/_news/2006/03/10/127685 -newest-additions-to-newsvine?lite. ———. 2006b. Jan. 18. “Newsvine Code of Honor.” Newsvine Blog. Retrieved Dec. 5, 2014, from http://blog.newsvine.com/_news/2006/01/18/63239 -newsvine-code-of-honor. ———. 2006c. Mar. 25. “Newsvine Code of Honor.” Retrieved Dec. 5, 2014, from http://web.archive.org/web/20060325032001/http://www.newsvine .com/_cms/info/codeofhonor. ———. 2006d. Jan. 5. “We’re Opening the Gates.” Newsvine Blog. Retrieved Dec. 5, 2014, from http://blog.newsvine.com/_news/2006/01/05/49085 -were-opening-the-gates. ———. 2007a. Oct. 7. “MSNBC.com Acquires Newsvine.” Newsvine Blog. Retrieved Dec. 5, 2014, from https://web.archive.org/web/20130816171947/ http://blog.newsvine.com/_news/2007/10/07/1008889-msnbccom-acquires -newsvine?. ———. 2007b. Dec. 13. “Newsvine Update—What’s New since MSNBC.” Newsvine Blog. Retrieved Dec. 5, 2014, from http://blog.newsvine.com/_news/ 2007/12/13/1157120-newsvine-update-whats-new-since-msnbc. ———. 2008a. “The Greenhouse.” Retrieved Dec. 5, 2014, from http://www .newsvine.com/_nv/cms/backyard/greenhouse. ———. 2008b. Mar. 28. “New Code of Honor Goes Live.” Newsvine Blog. Retrieved Dec. 5, 2014, from http://blog.newsvine.com/_news/2008 /03/28/1396664-new-code-of-honor-goes-live. ———. 2008c. “Newsvine Code of Honor.” Retrieved Dec. 5, 2014, from http:// www.newsvine.com/_nv/cms/info/codeOfHonor. ———. 2008d. June 24. “Tuesday’s Newsvine Outage.” Newsvine Blog. Retrieved Dec. 5, 2014, from http://blog.newsvine.com/_news/2008/06/24/1606803 -tuesdays-newsvine-outage. ———. 2009. “A Change in How Newsvine Handles MSNBC.com Discussion Threads.” Newsvine Blog. Retrieved Dec. 5, 2014, from http://blog.newsvine .com/_news/2009/01/21/2339642-a-change-in-how-newsvine-handles -msnbccom-discussion-threads. Nexidia. n.d. “Rich Media.” Multimedia marketing presentation. Retrieved May 31, 2011, from http://www.nexidia.com/resources/rich_media__how_it_ works_demo. Nguyen, V. 2010. Apr. 29. “MSNBC.com New Mobile Site: Design for Modularity.” Biofusion Design. Retrieved Dec. 5, 2014, from https://web.archive.org/

BIBLIOGRAPHY 295

web/20100515203237/http://www.biofusiondesign.com/2010/04/29/msnbc -com-new-mobile-site. Noam, E. M. 2009. Media Ownership and Concentration in America. New York: Oxford University Press. Noguchi, Y. 2006. Apr. 6. “MSNBC Brings Free News Clips to Cellphone Audience.” TechNews. Retrieved from LexisNexis Academic Database. Olbermann, K. 2008. Aug. 19. “Countdown Tuesday: Knows Picking.” The News Hole. Retrieved June 11, 2011, from http://thenewshole.msnbc.msn.com/_ news/2008/08/19/4363113-countdown-tuesday-knows-picking. Olmstead, K., A. Mitchell, J. Holcomb, and N. Vogt, N. 2014. Mar. 26. “News Video on the Web.” Survey. Pew Research Center Journalism Project. Retrieved Dec. 5, 2014, from http://www.journalism.org/2014/03/26/news -video-on-the-web. Outing, S. 2005. July 26. “Sophisticated Web Stats Give Editors Better Idea of Reader Interests.” Editor and Publisher. Retrieved from Factiva Database. Parks, L. 2012a. “Satellites, Oil, and Footprints.” Pp. 122–140 in L. Parks and J. Schwock, eds., Down to Earth: Satellite Technologies, Industries, and Cultures. New Brunswick, NJ: Rutgers University Press. ———. 2012b. “Technostruggles and the Satellite Dish: A Populist Approach to Infrastructure.” Pp. 64–84 in G. Bolin, ed., Cultural Technologies: The Shaping of Culture in Media and Society. New York: Routledge. Paterno, J. 2011. Mar. 2. “With So Many Huge Competitors, It’s Tough Being Google.” Business Insider. Retrieved Dec. 6, 2014, from http://www.business insider.com/google-value-stock-competition-2011-3. Paterson, C., and D. Domingo, eds. 2008. Making Online News: The Ethnography of New Media Production. New York: Peter Lang. Perren, A. 2010. “Business as Unusual: Conglomerate-Sized Challenges for Film and Television in the Digital Arena.” Journal of Popular Film and Television 38(2): 72–78. ———. 2013. “Rethinking Distribution for the Future of Media Industry Studies.” Cinema Journal 52(3): 165–171. Peterson, K. 2006. Sept. 27. “Citizen Journalism: News Brought to You by the Average Joe.” Seattle Times. Retrieved from LexisNexis Academic Database. Pew Project for Excellence in Journalism. 2010. The State of the News Media, 2010. Survey. Retrieved Dec. 6, 2014, from http://stateofthemedia.org/2010. ———. 2011. The State of the News Media, 2011. Survey. Retrieved Dec. 6, 2014, from http://stateofthemedia.org/2011. ———. 2013. The State of the News Media, 2013. Survey. Retrieved Dec. 6, 2014, from http://stateofthemedia.org/2013. Pexton, P. B. 2011. Mar. 25. Behind the Post’s Redesigned Web Site. Washington

BIBLIOGRAPHY 296

Post. Retrieved Dec. 6, 2014, from http://www.washingtonpost.com/opinions /behind-the-posts-redesigned-web-site/2011/03/25/AFC3GXYB_story.html. Pfaffenberger, B. 1992. “Technological Dramas.” Science, Technology, and Human Values 17(3): 282–312. Pickering, A. 1995. The Mangle of Practice: Time, Agency, and Science. Chicago: University of Chicago Press. Pinch, T. J. 2010a. “The Invisible Technologies of Goffman’s Sociology: From the Merry-Go-Round to the Internet.” Technology and Culture 51(2): 409–424. ———. 2010b. “On Making Infrastructure Visible: Putting the Non-Humans to Rights. Cambridge Journal of Economics 34(1): 77–89. Pinch, T. J., and W. E. Bijker. 1984. “The Social Construction of Facts and Arte­ facts; or, How the Sociology of Science and the Sociology of Technology Might Benefit Each Other.” Social Studies of Science 14 (3): 399–441. Plesner, U. 2009. “An Actor-Network Perspective on Changing Work Practices: Communication Technologies as Actants in Newswork.” Journalism 10(5): 604–626. Plesser, A. Executive Producer. 2008a. July 15. “Here’s a Scoop: MSNBC’s ‘Gossip Girl’ Courtney Hazlett Has Customized Micro-Site.” Video interview. Beet.TV. Retrieved Dec. 6, 2014, from http://www.beet.tv/2008/07/ heres-a-scoop-m.html. ———. 2008b. June 2. “MSNBC.com Launches Keith Olbermann Mini-Site.” Video interview. Beet.TV. Retrieved Dec. 6, 2014, from http://www.beet.tv/ 2008/06/msnbc-launches.html. ———. 2009a. Oct. 23. “Beet.TV Exclusive: MSNBC.com Readies New Video Player for Facebook.” Video interview. Beet.TV. Retrieved Dec. 6, 2014, from http://www.beet.tv/2009/10/msnbccom-readies-launch-of-sizable-embed dable-player-for-blogs-facebookalso-inplayer-contextual-vide.html. ———. 2009b. Oct. 30. “Mobile Video Is Getting Higher Ad Rates Than Online.” Video interview. Beet.TV. Retrieved Dec. 6, 2014, from http://www .beet.tv/2009/10/mobile-video-is-getting-higher-ad-rates-than-online -.html. ———. 2009c. Aug. 24. “MSNBC.com Chief Says Video Transcripts Make Editing Scalable with ‘Minimal Human Intervention.’” Video interview. Beet.TV. Retrieved Dec. 6, 2014, from http://www.beet.tv/2009/08/msnbccom-chief -says-video-transcripts-makes-editing-scalable.html. ———. 2009d. Aug. 12. “MSNBC.com to Monetize Its Videos throughout the Web. Video interview. Beet.TV. Retrieved Dec. 6, 2014, from http://www .beet.tv/2009/08/the-huffington-post-is-driving-revenue-to-msnbccom -.html. ———. 2009e. Nov. 5. “MSNBC.com’s Evolving Embeddable Video Player.” Video interview. Beet.TV. Retrieved Dec. 6, 2014, from http://www.beet

BIBLIOGRAPHY 297

.tv/2009/11/msnbccoms-embeddable-video-player-charlie-tillinghast -on-next-steps.html. ———. 2010. June 23. “MSNBC.com Readies First iPad App: Maybe for Rachel Maddow.” Video interview. Beet.TV. Retrieved Dec. 6, 2014, from http:// www.beet.tv/2010/06/msnbccom-readies-ipad-app-for-single-program -maybe-rachel-maddow.html. ———. 2011a. Apr. 8. “ABCNews.com Gets Seventy Percent of Video Views via Links and Search, Says Digital Chief.” Video interview. Beet.TV. Retrieved Dec. 6, 2014, from http://www.beet.tv/2011/04/abcnewscom-gets-seventy -percent-of-video-views-via-links-and-search-says-digital-chief-paul-slavin .html. ———. 2011b. May 8. “Exclusive: Rachel Maddow Leads Video Consumption among NBC News Network Shows.” Video interview. Beet.TV. Retrieved Dec. 6, 2014, from http://www.beet.tv/2011/05/rachel-maddow-tops-online -engagement-among-nbcmsnbccom-shows-.html. ———. 2011c. Jan. 6. “MSNBC.com Finding New Audiences for Video News via Mobile. Video interview. Beet.TV. Retrieved Dec. 6, 2014, from http://www .beet.tv/2011/01/msnbc-finding-new-audiences-for-news-with-mobile -devices-.html. ———. 2011d. Apr. 7. “MSNBC.com Finds Growing Traffic via Its Shared Video Player.” Video interview. Beet.TV. Retrieved Dec. 6, 2014, from http://www .beet.tv/2011/04/msnbcembed.html. Postigo, H. 2012. The Digital Rights Movement: The Role of Technology in Subverting Digital Copyright. Cambridge, MA: MIT Press. Powell, W. W. 2001. “The Capitalist Firm in the Twenty-First Century: Emerging Patterns in Western Enterprise.” Pp. 33–68 in P. DiMaggio, ed., The TwentyFirst Century Firm. Princeton, NJ: Princeton University Press. Pressler, J. 2008. Nov. 2. “The Dr. Maddow Show.” New York Magazine. Retrieved Dec. 6, 2014, from http://nymag.com/news/media/51822. Prior, M. 2007. Post-Broadcast Democracy: How Media Choice Increases Inequality in Political Involvement and Polarizes Elections. Cambridge, Eng.: Cambridge University Press. Purcell, K., L. Rainie, A. Mitchell, T. Rosenstiel, and K. Olmstead. 2010. Under­standing the Participatory News Consumer: How Internet and Cell Phone Users Have Turned News into a Social Experience. Survey. Pew Research Center. Retrieved Dec. 6, 2014, from http://www.pewinternet.org/2010/03/01/ understanding-the-participatory-news-consumer. Putnam, R. D. 2000. Bowling Alone: The Collapse and Revival of American Community. New York: Simon and Schuster. Rao, L. 2011. Feb. 4. “Tremor Media Acquires Mobile Video Ad Platform Transpera.” TechCrunch. Retrieved Dec. 6, 2014, from http://techcrunch

BIBLIOGRAPHY 298

.com/2011/02/04/exclusive-tremor-media-acquires-mobile-video-ad -platform-transpera. Raymond, E. S. 2000. “The Cathedral and the Bazaar.” Catb.Org. Retrieved Dec. 6, 2014, from http://catb.org/~esr/writings/cathedral-bazaar/cathedral -bazaar. Reagle, J. M. 2010. Good Faith Collaboration: The Culture of Wikipedia. Cambridge, MA: MIT Press. Riley, P., and N. Usher. 2010. Nov. 14. “Bridging Journalism’s Past with Its Future: Changes and Continuity.” Paper presented at the National Communication Association Annual Meeting in San Francisco. Roizen, B. F. (host). 2008. June 22. “Vator Box on Hollywood Startups, Mobile Ad Networks.” Video. Vator News. Retrieved Dec. 6, 2014, from http://vator .tv/news/2008-06-22-vator-box-on-hollywood-startups-mobile-ad-net works. Rosen, C. 2004. “The Age of Egocasting.” New Atlantis 7 (Fall/Winter): 51–72. Rosen, J. 2006. June 27. “The People Formerly Known as the Audience.” PressThink. Retrieved Dec. 6, 2014, from http://archive.pressthink. org/2006/06/27/ppl_frmr.html. ———. 2011. Feb. 26. “A Common Mistake in Seeking Out the Next Business Model for News Is to Begin by Positing that Such a Model Exists within the Known Universe.” Twitter post. Twitter. Retrieved Dec. 6, 2014, from https:// twitter.com/jayrosen_nyu/status/41556214747762688. Ryan, C. 1991. Prime Time Activism: Media Strategies for Grassroots Organizing. Boston: South End Press. Ryfe, D. M. 2012. Can Journalism Survive? An Inside Look at American Newsrooms. Cambridge, UK: Polity. Sandoval, G. 2011. May 30. “U.S. Calls Blitz on Alleged Pirated-Sports Sites.” CNET News. Retrieved Dec. 6, 2014, from http://news.cnet.com/830131001_3-20030429-261.html. Schechner, S. 2011. Jan. 22. “MSNBC, Olbermann Call It Quits, Cancel Show.” Wall Street Journal. Retrieved Dec. 6, 2014, from http://online.wsj.com/arti cle/SB10001424052748704754304576096772622398558.html. Scheufele, D. 2000. “Agenda-Setting, Priming, and Framing Revisited: Another Look at Cognitive Effects of Political Communication.” Mass Communication and Society 3 (2, 3): 297–316. Schlesinger, P. 1978. Putting “Reality” Together: BBC News. London: Constable. Schmitz Weiss, A., and D. Domingo. 2010. “Innovation Processes in Online Newsrooms as Actor-Networks and Communities of Practice.” New Media and Society 12(7): 1156–1171. Schudson, M. 1998. The Good Citizen: A History of American Civic Life. New York: Martin Kessler Books.

BIBLIOGRAPHY 299

———. 2000. “Good Citizens and Bad History: Today’s Political Ideas in Historical Perspective.” Communication Review 4(1): 1–20. ———. 2011. The Sociology of News. 2d ed. New York: W. W. Norton. Seelye, K. Q. 2007. May 8. “Muzzling All to Hush a Few.” New York Times. Retrieved Dec. 6, 2014, from http://thecaucus.blogs.nytimes.com/2007/05/08/ muzzling-all-to-hush-a-few. Segan, S. 2011. May 3. “RIM, Microsoft Partner for Bing on Blackberry.” PC Magazine. Retrieved from LexisNexis Academic Database. Shirky, C. 2008. Here Comes Everybody: The Power of Organizing without Organizations. New York: Penguin. Silverstone, R. 1994. Television and Everyday Life. New York: Routledge. Singer, J. B. 2004. “More Than Ink-Stained Wretches: The Resocialization of Print Journalists in Converged Newsrooms.” Journalism and Mass Communication Quarterly 81(4): 838–856. ———. 2010. “Journalism in the Network.” Pp. 277–286 in S. Allan, ed., The Routledge Companion to News and Journalism. New York: Routledge. Slattery, S. 2009. “‘Edit This Page’: The Socio-Technical Infrastructure of a Wikipedia Article.” Pp. 289–295 in Proceedings of the 27th Annual International Conference on Design of Communication, New York. “Social TV Leaderboard.” 2011. June 28. Lost Remote. Retrieved in Dec. 6, 2014, from https://web.archive.org/web/20110628225543/http://www.lostremote .com/top-tv-social-media. Sorgatz, R. 2007. Oct. 7. “Why We Bought Newsvine.” Fimoculous. Retrieved Dec. 6, 2014, from http://www.fimoculous.com/archive/post-3267.cfm. Squires, C. 2005. Dec. 10. “Networks Reinvent TV News for a Digital Age. St. Petersburg Times, 1A. Retrieved from LexisNexis Academic Database. Star, S. L., and G. C. Bowker. 2002. “How to Infrastructure.” Pp. 151–162 in L. A. Lievrouw and S. Livingstone, eds., Handbook of New Media: Social Shaping and Consequences of ICTs. Thousand Oaks, CA: Sage Publications. Star, S. L., and J. Griesemer. 1989. “Institutional Ecology, ‘Translations,’ and Boundary Objects: Amateurs and Professionals in Berkeley’s Museum of Vertebrate Zoology, 1907–39.” Social Studies of Science 19(3): 387–420. Stark, D. 1999. “Heterarchy: Distributing Authority and Organizing Diversity.” Pp. 153–179 in J. H. Clippinger, ed., The Biology of Business: Decoding the Natural Laws of Enterprise. San Francisco: Jossey-Bass. ———. 2009. The Sense of Dissonance: Accounts of Worth in Economic Life. Princeton, NJ: Princeton University Press. Starosielski, N. 2012. “‘Warning: Do Not Dig’: Negotiating the Visibility of Critical Infrastructures.” Journal of Visual Culture 11(1): 38–57. Steinberg, J. 2005. Aug. 25. “An Anchor by Evening, a Blogger Anytime.” New York Times. Retrieved from LexisNexis Academic Database.

BIBLIOGRAPHY 300

Stelter, B. 2007. Oct. 8. “MSNBC to Acquire a Chattier News Site.” New York Times, C5. Retrieved from LexisNexis Academic Database. ———. 2010a. Feb. 4. “CBS News Lays Off Dozens in New Round of Staff Cuts.” New York Times. Retrieved Dec. 6, 2014, from http://www.nytimes.com/ 2010/02/04/business/media/04cbs.html. ———. 2010b. Oct. 6. “Memo Hints at Name Change for MSNBC.com.” New York Times. Retrieved Dec. 6, 2014, from http://mediadecoder.blogs.ny times.com/2010/10/06/memo-hints-at-name-change-for-msnbc-com. Stoll, I. 2011. May 1. “Jeff Zucker on Immelt, Newspapers.” FutureOfCapitalism .com. Retrieved Dec. 6, 2014, from http://www.futureofcapitalism.com/ 2011/05/jeff-zucker-on-immelt-newspapers. Streeter, T. 1996. Selling the Air: A Critique of the Policy of Commercial Broadcasting in the United States. Chicago: University of Chicago Press. Strover, S., and W. Moner. 2013. “The Contours of On-Demand Viewing.” Pp. 234–253 in J. Holt and K. Sanson, eds., Connected Viewing: Selling, Streaming, and Sharing Media in the Digital Age. New York: Routledge. Suchman, L. 2000. “Located Accountabilities in Technology Production.” Paper presented at the Sawyer Seminar on Heterarchies at the Santa Fe Institute, Santa Fe, NM. Sunstein, C. R. 2007. Republic.com 2.0. Princeton, NJ: Princeton University Press. Tang, C. 2005. Dec. 6. “On Newsvine, Inc.” Calvintang.com. Retrieved Dec. 6, 2014, from http://www.calvintang.com/blog/2005/12/on-newsvine-inc. ———. 2006a. Jan. 5. “Newsvine Private Beta.” Calvintang.com. Retrieved Dec. 6, 2014, from http://www.calvintang.com/blog/2006/01/newsvine-private-beta. ———. 2006b. Mar. 1. “Newsvine Public Launch.” Calvintang.com. Retrieved Dec. 6, 2014, from http://www.calvintang.com/blog/2006/03/newsvine -public-launch. ———. 2007a. Oct. 7. “Open Letter to Newsvine Community.” Calvintang.com. Retrieved Dec. 6, 2014, from http://www.calvintang.com/blog/2007/10/ open-letter-to-newsvine-community. ———. 2007b. Oct. 30. “Viki Gonia on MSNBC—Oct. 30th Democratic Debate.” Calvin Tang’s Newsvine Column. Retrieved Dec. 6, 2014, from http:// tang.newsvine.com/_news/2007/10/30/1061099-viki-gonia-on-msnbc-oct -30th-democratic-debate. ———. 2010. Mar. 30. “Newsvine Next Level.” Calvin Tang’s Newsvine Column. Retrieved Dec. 6, 2014, from http://tang.newsvine.com/_news/2010/03/30/ 4088039-newsvine-next-level. Taylor, C. 1996. May 4. “This Just In: Microsoft Hopes to Shape News, Info on Internet.” Dallas Morning News, 1F. Retrieved from LexisNexis Academic Database. Telligent. 2007. “Case Studies: MSNBC Blogs.” Marketing material. Telligent.com. Retrieved Dec. 6, 2014, from http://web.archive.org/web/20080501161241/ telligent.com/casestudies/archive/msnbc.aspx.

BIBLIOGRAPHY 301

Terranova, T. 2000. “Free Labor: Producing Culture for the Digital Economy.” Social Text 18(2): 33–58. Thelwell, E. 2007. Oct. 9. “Microsoft and NBC Buy Social Media Site.” Daily Telegraph, 3. Retrieved Dec. 7, 2014, from LexisNexis Academic Database. Tofel, K. C. 2007. Jan. 19. “Zune Podcast Support Squirting Slowly on the ’Net.” Mobilize. Retrieved Dec. 7, 2014, from http://gigaom.com/mobile/zune_ podcast_su. Transpera. 2009. Mar. 11. “MSNBC.com Mobile Now Showing Video.” Press release. Retrieved from LexisNexis Academic Database. Trout, C. 2011. Mar. 22. “ComScore: Microsoft Becomes Second for Online Video in One Month.” Updated. Engadget. Retrieved Dec. 7, 2014, from http://www.engadget.com/2011/03/22/comscore-microsoft-becomes -second-for-online-video-in-one-month. ———. 2013. On-Demand Culture: Digital Delivery and the Future of Movies. New Brunswick, NJ: Rutgers University Press. Tryon, C., and M. Dawson. 2013. “Streaming U: College Students and Connected Viewing.” Pp. 217–233 in J. Holt and K. Sanson, eds., Connected Viewing: Selling, Streaming, and Sharing Media in the Digital Age. New York: Routledge. Tuchman, G. 1978. Making News: A Study in the Construction of Reality. New York: Free Press. Tumber, H. 1999. News: A Reader. New York: Oxford University Press. Tunstall, J. 1971. Journalists at Work: Specialist Correspondents; Their News Organizations, News Sources, and Competitor-Colleagues. London: Constable. Turner, F. 2005. “Actor-Networking the News.” Social Epistemology 19(4): 321–324. ———. 2006. From Counterculture to Cyberculture: Stewart Brand, the Whole Earth Network, and the Rise of Digital Utopianism. Chicago: University of Chicago Press. Turow, J. 1992. Media Systems in Society: Understanding Industries, Strategies, and Power. New York: Longman. ———. 2004. “Audience Construction and Culture Production: Marketing Surveillance in the Digital Age. Annals of the American Academy of Political and Social Science 597(1): 103–121. ———. 2011. The Daily You. New Haven: Yale University Press. Usher, N. 2010. Nov. 23. “Why Spreadable Doesn’t Equal Viral: A Conversation with Henry Jenkins.” Interview. Nieman Journalism Lab. Retrieved Dec. 7, 2014, from http://www.niemanlab.org/2010/11/why-spreadable-doesnt -equal-viral-a-conversation-with-henry-jenkins. Usher, N., and P. Riley. 2010. Apr. 24. “Reshaping the Public Radio Newsroom for the Digital Future.” Paper presented at the International Symposium on Online Journalism, Austin, TX. Verton, D. 2002. Sept. 9. “Technology Aids Hunt for Terrorists.” ComputerWorld.

BIBLIOGRAPHY 302

Retrieved Dec. 7, 2014, from http://www.computerworld.com/s/article/ 74052/Technology_aids_hunt_for_terrorists. Wallenstein, A. 2011. Feb. 8. “MSN’s Scott Moore Questions ‘Risky Bet’ AOL Made on HuffPo.” PaidContent.org. Retrieved Dec. 7, 2014, from http:// paidcontent.org/article/419-msns-scott-moore-questions-risky-bet-aol -made-on-huffpo. Warner, M. 1970. “Decision-Making in Network Television News.” Pp. 158–167 in Media Sociology. Urbana: University of Illinois Press. ———. 2002. Publics and Counterpublics. New York: Zone Books. Weiner, R., and K. Stalter. 1995. Aug. 14. “TBS, Microsoft Talk.” Daily Variety, 1. Retrieved from LexisNexis Academic Database. Wellman, B., and M. Gulia. 1999. “Net-Surfers Don’t Ride Alone: Virtual Communities as Communities.” Pp. 331–366 in B. Wellman, ed., Networks in the Global Village: Life in Contemporary Communities. Boulder, CO: Westview. White, D. M. 1950. “The ‘Gate Keeper’: A Case Study in the Selection of News.” Journalism Quarterly 27:383–390. Williams, R. 1975. Television: Technology and Cultural Form. New York: Schocken Books. Winer, D. 2011, Apr. 4. “Mysterious Tools behind Engadget Move?” Scripting News. Retrieved Dec. 7, 2014, from http://scripting.com/stories/2011/04/04/ mysteriousToolsBehindEngad.html. “Wireless Content Deals Keep Coming.” 2000. June 13. Electronic Advertising and Marketplace Report 14(12). Retrieved from LexisNexis Academic Database. Wong, E. 2010. Apr. 27. “MSNBC.com Friends Smartphones.” AdWeek. Retrieved Dec. 7, 2014, from http://www.adweek.com/news/advertising-branding/ msnbccom-friends-smartphones-107304. Woolgar, S. 1991. “Configuring the User.” Pp. 57–99 in J. Law, ed., A Sociology of Monsters. New York: Routledge. Young, S. 2008, Jan. 7. “Embed Our Video on Your Site.” Alpha Channel. Retrieved Dec. 7, 2014, from https://web.archive.org/web/20120520062327/ http://alphachannel.msnbc.msn.com/_news/2008/01/07/4466076-embed -our-video-on-your-site?. YuMe Corporation. n.d. Safeguards for Your Brand. Retrieved May 21, 2011, from http://www.yume.com/content/brand-security. Zichermann, G. 2011. Apr. 26. “The Purpose of Gamification.” O’Reilly Radar. Retrieved Dec. 7, 2014, from http://radar.oreilly.com/2011/04/gamification -purpose-marketing.html. Zittrain, J. L. 2008. The Future of the Internet and How to Stop It. New Haven: Yale University Press.

INDEX

ABCNews.com, 4, 60, 61, 123, 158, 232 Action Engine, 95, 96, 98 actor-network theory: and boundaries of systems, 119, 270n9; general symmetry in, 113, 116, 269n2; and heterogeneous engineering, 113, 116, 241, 256–57; and hierarchy, 241; and isometric pressure, 277n2; and media research, 112–15, 270n3; and the social construction of technology, 269n2; and YuMe, 87 actors: and distribution routes, 9, 14, 15, 39; networks of, 42; nonhuman, 113; provincial objectives of, 110–11; recalcitrance of, 40; in systems, 120, 154, 182, 203–8, 252–53, 256. See also actor-network theory Adams, Tyler, 271n9; on the comment-collapsing algorithm, 139; on the Greenhouse, 136; on moderation presence, 144; on MSNBC.com integration, 162; on the Newsvine Code of Honor, 141–42; on Newsvine gangs, 141; on Newsvine Guides, 143; Newsvine 3.0, 241 ad networks, 24, 86, 88–89, 96, 268n10. See also Transpera; YuMe Adobe Flash, 97, 268n9, 268n11 advertisers. See advertising advertising: business model, 251; and the embeddable player, 70, 83, 86–88, 250; and engagement, 214; and mobile distribution, 95, 96, 98; on Newsvine.com, 130, 144, 145, 150, 162; online spending on, 52; target audiences for, 24; and vantage points, 119 affordances: for distribution, 5, 20, 250; of MSNBC.com’s publishing platforms, 188,

231; for public participation, 29–30, 176; and translation, 114; of video players, 57. See also constraints agency: actor-network theory and, 113–14; and automation, 43–44; material, 9, 40, 43–44, 113–14, 264n3; and recalcitrance, 40; and structure, 11, 19, 77, 108, 114, 249, 250, 252 Air America, 197 AJAX, 128 algorithms: and delegation, 43–44; and featured comments, 140; as heterogeneous products, 139–40; and knowledge practices, 43; for moderation, 131, 135, 139–40, 204–205; and page layouts, 135, 145, 267n7; and public relevance, 28–29; for story recommendations, 148; surveillance by, 29; and transparent intermediaries, 76, 87, 250 Amazon, 23, 34, 127, 271n3 American Express ‘Your Business’ video site, 225 America’s Talking, 50, 53, 265n10 analytics. See audience measurement Ananny, Mike, 30, 152 Anderson, Chris, 165; The Long Tail, 165 Anderson, C. W., 24, 26, 70, 114–15, 215, 251 Anderson, Lance, 124, 125, 130, 147–48, 161 Andrea Mitchell Reports, 67, 180 Android, 96, 97, 269n13 And That’s the Way It Will Be (Harper), 50 AOL (America Online): commenting functionality, 152; content management system, 273n3 (chap. 9); dial-up network, 49; editorial strategy, 256; and Huffington Post, 91, 273n3 (chap. 8); video, 58

303

INDEX 304

Apache, 127, 155 APIs (application programming interfaces), 30–31; SkyPad as, 228 Apollo 13, 39–40, 117 Apple, 23, 94, 127, 268n9. See also iPad; iPhone Apple Safari, 268n11 application programming interfaces. See APIs apps. See mobile apps Ardome, 53 artifacts, 14; and distribution routes, 39; heterogeneity of, 43–45; news as, 2; politics of, 40, 55–56; social construction of, 112, 116, 204, 206 Ashley Blogs, 225 Associated Press, 78, 126, 149, 150 audience effects, 6, 7, 19 audience engagement. See engagement audience measurement, 24–26, 85–86, 214–15, 250–52, 265n5. See also audiences; institutionalized audiences audiences: and brands, 166, 207, 217, 231, 245; constructed quality of, 24–26, 251–52; fragmentation of, 213–15, 217, 221, 222; mass vs. niche, 25–26, 217, 222, 229, 254, 275n4; for mobile content, 93; older, 276n6; for the Rachel Maddow Show, 170, 172, 182, 198; on smartphones, 96–97; and social distribution, 245; surveillance of, 24–26, 250–52, 263n1; targeting, 20, 166, 220–21; for Today.com, 217, 220–21; for Today Show, 97–98, 220–21; and user communities, 236; visibility of, 24–26. See also audience effects; audience measurement; engagement; institutionalized audiences audio clips, 50 automation: and agency, 43–44, 113; of AP postings to Newsvine, 126; and editorial control, 73–74, 267n7; and knowledge practices, 43; and moderation of users, 131, 136–38, 139–40; of spam (see spambots); of video transcoding, 82–83. See also algorithms Avid, 52–53

Ballmer, Steve, 51, 52 Battelle, John, 70; The Search, 3–4 BBC, 22, 50 Benen, Steve, 259–60 Benkler, Yochai, 85 Bing, 56–57, 57, 58–60, 82, 198 Bites, 217 Bivens, Rena, 23, 107 black-boxing, 3, 15, 101, 108, 114 bloggers, 79, 82, 85–86, 197, 222–23, 244 blogging software. See content management systems blogs: and branding, 217–18, 222–23, 231–32; and broadcast networks, 268n7; and conversation online, 166; format of, 187, 259; MSNBC.com adoption of, 91, 167, 222–23; and MSNBC.com redesign, 165–69, 193–94; in news routines, 173, 191–92; Newsvine-powered, 122, 188, 193–96; and online communities, 222–23; and peer production, 212; for shows, 67, 74, 122, 173–77, 178, 185–87 blogware. See content management systems Blue Site. See MSNBC.com (website) Boczkowski, Pablo: Digitizing the News, 12; imitation in news work, 23, 107; niche editorial products, 275n4; provisional settlements, 118, 277n2; use cases, 70 bootlegging: and free speech, 27; in underserved markets, 27; of televised sports, 44; of MSNBC clips, 45, 78, 81, 84–86, 119, 248, 270n7; and metering, 270n8; and value, 84–85, 249 Bosk, Charles, 122–23 boundary objects, 215, 253, 274n5 Bourdieu, Pierre, 114 boyd, danah, 128, 273n7 Boyle, Alan, 222, 235 Boyle, James, 27 Boxee, 44, 249 brand-safety, 10, 86–88, 250 brands: blogs as, 222–23; ingredient, 239; language of, 217–18; management of, 218–21, 243–44; master and sub, 217–21, 244; MSNBC and MSNBC.com, 65–67,

INDEX 305

72, 162, 176, 197–99, 245; of MSNBC TV programs, 171, 217–21; Newsvine, 149–51, 159, 161; niche, 218, 231–32, 243–44, 255; and organizations, 207–208, 256; personal, 218, 231; proliferation of, 17, 178, 202, 216–31, 241, 243, 245, 267n6; Rachel Maddow Show, 171, 182; recalcitrance of, 207, 218–20; and software architecture, 231, 238; targeting audiences through, 166, 207, 216–23, 245; on television, 245; television versus online, 217–21; user groups as, 238–39; white-label, 10, 16, 57, 77, 239 breaking news: and blog format, 186–87; and evening news, 33, 68–69, 219–20, 267n6 (chap. 4); and Newsvine, 124, 138, 143; and story pages, 229 Breaking News (startup), 65, 202, 217, 225, 242 Brightcove, 57, 60, 92, 250 broadband penetration, 9, 13, 51, 52 broadcast networks: affiliates, 6–7, 265n4; and blogging, 268n7; and the civil rights movement, 21–22; consumers bypassing, 13; and embeddable video, 85; evening news as flagship, 219; live streaming and, 9–10, 265n5; and user participation, 157–59, 163, 192. See also cable channels; cable providers Brown, Jennifer, 213, 221, 242, 269n15 Bruns, Axel, 212 Bryant Park Project, 192 Budos, Mark, 125, 144, 156, 163, 165, 225 Burke, Kenneth, 39–40, 42, 113 Bush administration, 66 business-to-business distribution firms, 10, 15, 76–77, 250. See also transparent intermediaries C#, 155, 156 cable channels: disintermediation of cable providers, 99; and embeddable video, 85; and media choice, 214; online availability, 9; and public participation, 30, 158–59, 163; revenues, 12. See also MSNBC TV

cable news, 12, 65–66, 107, 170–71, 273n1 cable providers: consumers bypassing, 13; live streaming and time delay, 45, 99–100, 118–19, 248, 259, 265n5; and on-demand content online, 99–100. See also broadcast networks calculated publics, 29 Callon, Michel, 42, 113, 119 Captain, Catherine, 95 captions, 47–48, 80 Carey, James, 3 Carlson, Tucker, 267n5 (chap. 4) cascading style sheets. See CSS categories, 48, 56, 62, 63, 64, 70–71 CBS, 50 CBS Interactive, 232, 267n7 CBS News, 10, 96, 106, 158, 220 CBSNews.com, 157, 158, 232, 267n7, 271n6 cellular carriers, 94, 98 censorship, 26, 27. See also free speech chains of custody, 15, 80–81 Cheezburger, Inc., 256 Chia, Robert, 117, 120, 169, 206 Chrome (web browser), 268n11 chromes (code), 223–25, 230 chyron graphics, 47–48, 245 citizen journalism: and mainstream media, 152, 239; and Newsvine, 124, 125, 138, 154, 205, 239 classes, 223–25 class extensions, 223–25 Clicker, The, 217 Clinton impeachment, 65 clip & share interface, 77, 78. See also embeddable player closed captioning, 89–90, 267n3. See also Nexidia cloud services, 9, 271n3 CNN, 48–49, 53, 66, 118, 152 CNN.com, 66, 118, 152 Code of Honor. See Newsvine Code of Honor Coleman, Gabriella, 34, 111–12 columnists, 217–18, 222–23, 231–32

INDEX 306

Comcast: MSNBC acquisition, 13, 18, 118, 203, 258–61, 263n6, 274n6; and the MSNBC joint venture, 52, 118, 203, 258–61; NBC Universal acquisition, 52, 258–61; comment-collapsing algorithm, 139–40 comment moderation, 132–36, 139–40, 157–59, 163, 184, 207, 232–41, 276n1 (chap. 13) commercials. See advertising commons-based peer production, 115, 212–13, 254–55 communication infrastructure theory, 20 communities, 20–22, 34, 183–84, 212–13, 223. See also Newsvine user base; publics communities of the affected, 31. See also publics Community Server, 91, 167, 174–75, 183, 188, 194 Conaway, Laura, 190–94, 196–99, 202–203, 244–45, 259–60 conditions of congregation, 20–22, 72, 166, 257, 268n12 constraints, 5, 20. See also affordances consumer electronics industry, 23, 249 content: access to, 5–6, 10, 34, 44–45, 248; diversity of, 106; path of (see voltas); quality of, 130, 143; ratings, influence of, 24–26; repurposing of, 231–32; as scholarship focus, 1–2, 19. See also content management systems content management systems: at AOL, 273n3 (chap. 9); faking, 185–87, 189, 195, 204; flexibility of, 223–32, 241, 243–45, 275n3 (chap. 12); heterogeneous engineering of, 45; MSNBC.com blogging, 91–92, 167, 188–89, 193–96, 222–25; MSNBC.com general, 14, 16, 17, 53, 54, 204, 223–30; syndication through, 56, 198; for television, 52–53; at the Washington Post, 274n3 (chap. 9). See also Community Server; M3 architecture; Newsvine (software platform); SkyPad; TypePad convergence, 50 conversation economy, 3–6, 17, 100, 200; and embeddable player, 79, 92, 216; and

epistemic encounters, 213; and MSNBC brands, 216–21; and Newsvine, 166–67; and the Rachel Maddow Show, 183. See also epistemic encounters; social distribution copyright. See intellectual property correspondents, 107, 171, 275n3 (chap. 11) Cosmic Log, 222, 235 Couldry, Nick, 114, 207, 241, 244 Countdown with Keith Olbermann: brand, 171; metadata for video clips, 265; microsite template, 175–76, 190; and the mobile website, 95; and MSNBC TV’s editorial strategy, 66; ratings, 66, 200; web producer for, 180, 190; and Windows Mobile, 94; YouTube bootlegging of, 78 Creative Development. See under MSNBC. com (company) cross-platform video distribution, 93–100 cross-promotion, 58–60, 59, 66–67, 266n1 crowdsourcing, 134, 151 CSS (cascading style sheets), 98, 123, 225, 269n14 culture: access to, 7, 10, 18; exporting of, 22; media, 32–34; and network forums, 29–30; visibility of, 21, 22 Dani, Sarika, 71, 269n15 databases, 126–28, 155, 228, 272n1 Dateline, 95, 97, 180 DavidGregoryTV.com, 275n5 Davidson, Mike: on the arms-length community concept, 159; on blogs, Newsvine-powered, 174, 222, 233; on comment moderation, 133–34, 136, 232; on development delays, 174, 240; on the “Discuss on Newsvine” feature, 163–64; on general-interest social networks, 166; on leaving Disney, 123–24; and the Mike blog template, 194; on news explorers, 153; on Newsvine 3.0, 235–40; on the Newsvine acquisition, 149–51, 156–57, 236; on the Newsvine community, 129, 235–40; and Newsvine concept, 124; on the New York Times-Newsvine partner-

INDEX 307

ship, 144; on online communities for legacy media, 236, 276n1 (chap. 13); on posting guidelines, 132; on Random Acts of Vineness, 136–37; and spam, 144 Deciding What’s News (Gans), 1, 106, 170 deflation, 139 delegation, 43, 113, 265n3 Democratic National Convention, 151 design: graphic, 167, 174 (see also chromes; CSS); and heterogeneity, 41, 43, 139; and participatory affordances, 30; and playlists, 71; of public-relevance algorithms, 28–29; and technological dramas, 44–45; and technology studies, 55–56; users, as inscribed in, 62, 70; values in, 35, 40, 140 “Determination of News Photographs, The” (Hall), 2 Deuze, Mark: Media Work, 23; on media work, 22–23; on the news cycle, 107; on organizational subcultures, 109, 169; on permanent revolution, 115; on professional ideology, 269n1; on project teams, 243 Digg, 124, 144 Digital Life, 217 digital rights management (DRM), 5, 23, 34, 249 digital word of mouth (DWOM), 91 Digitizing the News (Boczkowski), 12 Dipity, 183 Disney, 10, 96, 123–24 dissociation, 42 distribution: conceptual tools for studying, 248–57; as control over information and culture, 7, 18, 78, 248; and conversation economy, 3–6, 100, 200, 245; cross-­ platform, 93–100; defined, 46; invisibility of, 3; network perspective on, 109; practices, 6; research on, lack of, 6–7, 19, 248; significance of, 19–35; windows, 100. See also distribution systems; social distribution; voltas distribution studies, 19–35, 248–57 distribution systems: conceptual tools for studying, 248–57; creation of, 17; and heter-

archy, 252–53; and isometric pressure, 256; monetization of, pressure surrounding, 12; MSNBC cable site and, 203; norms invested in, 34–35; provincialism in, 204–205, 245; research on, lack of, 6–7, 8; and systems of knowledge, 254–56; variation in, 13. See also system building; systems; voltas domain-blocking, 5, 267n4 Donahue, Phil, 66 Doonesbury, 59 dot-com crash, 151 DRM. See digital rights management Drupal, 258 Dunbar’s number, 276n2 Duneier, Mitchell, 21, 26–27 DWOM. See digital word of mouth Dwyer, Tim, 5, 12 Dylan Ratigan Show, 67, 180 Economist, 4, 220 editorial control, 51, 73–74, 85–86, 201–203, 267n7 editorial products, 218, 223, 229, 231, 244–45, 255. See also brands Ed Show, 180, 202, 226, 266n4 Education Nation, 48, 258 elections, 65, 157, 162, 232 Elkhart Project, 193–94, 225 Ellison, Nicole, 128, 273n7 embed codes, 77, 78, 91, 92 embeddable clips. See embeddable player embeddable player: and advertising, 70, 83, 86–88, 250; and conversation economy, 216; development of, 76, 77–92; dimensions, 78, 79, 267n2; monetization challenges, 83, 250; and MSNBC.com publishing platforms, 91–92; social media integration, 92; sub-clip functionality, 77–78, 79, 82, 83; as traffic driver, 45–46, 77; and traffic-sharing agreements, 90–91; user interface, 77–78, 78, 79; and YouTube, 45 embeddable slideshows, 183, 185 embeddable video player. See embeddable player

INDEX 308

embedded advertising, 86–88 engagement, 17, 80, 152, 184, 192, 205, 214–15 enrollment: in MSNBC.com’s editorial strategy, 231, 244; in MSNBC’s system for flipping video, 46, 61, 62; in Newsvine’s system, 155, 157, 254; in the Rachel Maddow Show’s system, 177; and system building, 42, 206–207 epistemic concerns, 231. See also epistemic cultures; epistemic encounters epistemic cultures, 211–12, 215, 223, 253, 254–56, 269–70n2 (chap. 6) epistemic encounters, 213, 215, 218, 254–56. See also epistemic cultures ESPN, 123, 124, 128, 163 evening newscasts, 33, 68–69, 170–71, 219–220, 267n6 (chap. 4) EveryBlock, 65, 202, 217, 242 Facebook: and Bing search, 58; and the conversation economy, 4, 5; and the embeddable player, 77, 92; and MSNBC, 184; and NBC News, 184; News Feed, 28–29, 144; New York Times partnership, 144; and niche social networks, 166–67, 273n7; subscribers, 271–72n10; and twoscreen experiences, 165. See also social distribution; social media FCC (Federal Communications Commission), 48 Femia, Will: on brands, 158, 207, 218; and Countdown with Keith Olbermann, 190; on development delays, 174–75, 185–86; and faking blogware, 185–87, 189, 195; Rachel Maddow Show, autonomy of, 188–89; Rachel Maddow Show, work with, 179–87, 190–91, 194–95, 197, 205–206, 260, 274n4 (chap. 9), 274n1 (chap. 10); on television, 1, 3; on user participation, 158, 234; and video clips, 45, 71, 86, 198, 218 field theory, 114, 277n2 file formats, 47, 96 file transfers, 198

Finkler, Cathy, 71, 175 Firefox, 268n11 First Read, 226, 235 flames, 163–64 Flamoe, Eric, 155–57 Flash, 97, 268n9, 268n11 flexible labor, 213 flexible leisure, 213, 220 Flickr, 98, 183–85, 197, 212 FlipCams, 199, 200, 274n3 (chap. 10) FlipFactory, 47 flipping video, 46–48, 197–98 Fox Business News, 78 Fox News Channel, 65, 266n5 (chap. 4) frameworks, 16, 271n4. See also software architectures free speech, 27, 249, 257. See also censorship Frost, Sandy, 138 Gadgetbox, 217 Galison, Peter, 215 Galtung, Johan, 2; “The Structure of Foreign News,” 2 Gans, Herbert, 2, 3, 106; Deciding What’s News, 1, 106, 170 “Gate-Keeper, The” (White), 1 gatekeeping, 1–3, 159, 233, 255 Gates, Bill, 49, 50 Gawker, 238, 256, 276n4 General Electric, 53, 206 general symmetry, 113, 116, 269n2 geofencing, 5, 84, 267n4 Giant List of Scum, 142 Gillespie, Tarleton, 27, 28–29, 34, 263n1 (chap. 2) Girard, Monique, 116, 118, 177, 243, 277n2 Gitlin, Todd, 2 GLoS (Giant List of Scum), 142 Go, Jesamyn “Sam”: on automation, 73; on MSNBC TV staff structure, 203; on Rachel Maddow, 172; resources, use of, 181–82, 190, 200; on use cases, 70–77; video clips, categorizing, 70–71; and web producers’ work routines, 179, 180 Gonia, Viki, 151

INDEX 309

Google, 127; App Engine, 271n3; Buzz, 273n3 (chap. 9); Chrome, 268n11; Gmail, 273n3 (chap. 9); News, 4, 124; Plus (+), 260; TV, 249; Wave, 273n3 (chap. 9) Greenhouse, 135–36, 142, 145, 161–62 Green Site. See Newsvine.com (website) Gregory, David, 275n5 Griesemer, James, 119, 274n5 Griffin, Phil, 172 Grio, the, 48, 258 Guardian (London), 30 Hall, Stuart, 1–2, 106; “The Determination of News Photographs,” 2 Hardball with Chris Matthews: blog, 173; brand, 171; show front, 63, 74; subdomain for, 67; web producer for, 180, 182; and Windows Mobile, 94 Harper, Christopher, 50; And That’s the Way It Will Be, 50 headlines, 47–48, 62, 64, 80, 218 Hemmingway, Emma, 114, 270n5 Hermida, Alfred, 158–59, 233 heterarchy: and epistemic cultures, 254; features of, 109–11, 247, 252–53; in media firms, 11, 115–16, 177–78, 252–53; at MSNBC, 17, 169, 177–78, 199, 205–206, 241–44, 260–61 heterogeneities. See heterogeneous engineering heterogeneous engineering: and actor-network theory, 113, 116, 241, 256–57; and audience measurement, 214; and the cable site (standalone), 201–204, 244–45; under Comcast, 261; and the “Discuss on Newsvine” feature, 160–64; and distribution research, 12, 45–46, 244–45, 256–57; and entropy, 41, 42; features of, 8–10, 40–43; and flexible architectures, 231–32, 241, 244; and flipping video, 48, 83; and heterarchy, 111, 116–17, 241–44; and hierarchy, 241–44; and local teleology, 168–69, 206; in media organizations, 106; and media texts, 215–16; and Meet the Press, 275n5; and mobile distribution, 94; by MSNBC.com, 151–54, 176, 231,

241–44; and recalcitrance, 40, 42, 206; of Newsvine’s solution stack, 126–28, 154–57; of Newsvine’s user base, 129, 134, 137–38, 140, 143, 151, 154, 238–39; and on- and off-site distribution, 100; and promotion on-air, 244–45; and the Rachel Maddow Show’s web presence, 170–78; and social media integration, 92; time delay as example of, 45–46; and Today. com, 242; and transparent intermediaries, 88–89, 252; by users and audiences, 231, 245, 274n1 (chap. 11); of the video player, 61–62, 64–65, 72–73, 216, 261; volta as example of, 8–10, 39 hierarchy: failure of, 270n6; flattening of, 110–11, 206, 252; and heterarchy, 11, 115–16, 199, 206, 241–44, 247, 252–53, 261; in systems, 119, 168. See also heterarchy Holt, Jennifer, 23, 263n2 HTML, 77, 97, 99, 126 Hubris: Selling the Iraq War, 270n7 Huffington Post, 91, 124, 201–202, 218, 259, 273n3 (chap. 8) Hughes, Thomas: heterogeneity, 40–41, 75; human-built world, 44; Rescuing Prometheus, 247; reverse salients, 41, 168, 177–78; systems, boundaries of, 119, 270n9; systems, definitions of, 264n1 (chap. 3); systems, modern, 119, 247, 270n6; systems, postmodern, 11, 101, 115–16, 169, 247, 264n2 (chap. 3), 270n6; systems, shift in view of, 101, 264n2 (chap. 3) Hulu: and media culture, 34; and regional media, 22; revenues, 99; syndication of NBC News video to, 56–57; viewing numbers, 58; viewing restrictions, 44, 84, 85, 267n4 Hutchins, Edwin, 43–44, 113, 264n3 identity politics, 21, 25, 257 imagined communities, 21, 257 immediacy, 3, 23 inauguration speeches, 81

INDEX 310

information, 2, 3, 9, 13, 22–23 infrastructures: and audience measurement, 24–26; and behavior, 8, 40; content management systems as, 17, 236; and distribution routes, 3, 9, 100; invisibility of, 8, 15, 249; obduracy of, 40–41; and publics, 20–22, 31, 152; as scholarship focus, 8, 19, 248, 277n1 (chap. 14); software as, 76–77; stakeholders in, 35; and television news, online distribution of, 1, 14, 16, 261; and transparent intermediaries, 76–77, 87, 90, 250; for web applications, 126–28, 149 In-Game, 217 institutionalized audiences, 24, 30, 214–15, 250–52, 255 intellectual property, 22–23, 27, 211, 249, 267n5 (chap. 5) interactive television, 50, 51, 53, 265n9 Internet: access, 9, 10, 13, 51–52; censorship, 27; and dial-up networks, 49; framings of, 30; and news media, 7, 12, 75; and organizational forms, 212, 248; provincial uses of, 112 Internet Explorer, 58, 266n2, 268n11 interpellation, 21, 26, 30, 31 interpretive flexibility, 112 iOS, 268n9. See also iPad, iPhone iPad, 97–98, 99–100, 181, 268n9, 269n13 IP-blocking, 5, 135 iPhone, 95, 96–98, 268n9 Iran, 27, 84, 267n5 (chap. 5) Iraq War, 66 iReport, 152 isometric pressure: and actor-network theory, 277n2; features of, 117–20, 120, 253–54; and MSNBC.com, 154; and Newsvine, 154, 163, 168, 205, 254; of web production for MSNBC TV, 181, 199–201, 205, 254, 260 isometric stability. See isometric pressure iTunes, 94 Jansing & Co., 180 JavaScript, 128, 240 Jenkins, Henry, 33, 213

John, Richard, 3, 20, 22, 247, 263n1 (chap. 2) journalism studies: and actor-network theory, 112–15; and gender, 275n7; and material agency, 114; news work, 106–108; research foci, 6, 8, 75; and sociology of knowledge, 211 journalists: and comments, 158–59, 232, 234; and the community value of news, 213, 215; and distribution, 7, 22–23, 74, 106; and embeddable content, 85; news judgment, 5, 106–108; reading in, 173; television, 173, 275. See also media work Joyce, Cynthia, 69, 275n6 key words, 80–81 Killfile (Chris Thomas), 138, 150, 151 King, Martin Luther, Jr., 21–22 kludges, 14; and Newsvine, 169; and the Rachel Maddow Show’s web presence, 182; recalcitrance and, 39–41; transparent intermediaries as, 88; and video files, 198; and the video player, 62, 261 Knorr Cetina, Karin, 211–12, 254–55, 269–70n2 knowledge: and automation, 43; social construction of, 113; sociology of, 211–12, 253, 254–56; systems of, 211–12, 216, 254–56. See also epistemic cultures; epistemic encounters Kokas, Aynne, 33 LAMP stack, 127, 206 Larkin, Brian, 7, 22, 27 Last Word: brand, 171; web presence, 63, 74, 202, 226, 266n4; web producers for, 180, 202 Latour, Bruno: chreod, 249; deflation, 139; delegation, 43, 264n3; general symmetry, 113; Janus face, 182; material agency, 264n3; socio-technical hybrids, 115; and technology studies, 55–56 launcher links, 64, 69, 74, 195 Law, John: and actor-network theory, 113, 256–57; aircraft design equations, 139;

INDEX 311

entropy in systems, 41, 42, 117; heterogeneous engineering, 8–9, 40–43, 116–17, 138, 176, 256, 274–75n1 (chap. 11); on Hughes, 41, 116–17, 264n2 (chap. 3); on organization, 241; and organizational studies, 117, 241; and the social construction of technology, 116; and technology studies, 55–56; the volta, 8–9, 39, 100 layout engines, 96, 268n11 Lean Forward campaign, 66–67, 205 legacy media organizations: attitudes toward change, 11–12, 14; financial difficulties, 12–13; organization of, 108–109; provincialism within, 108–109, 273n3 (chap. 9); and user participation, 157–59, 163, 205, 232–36, 239; Levi, Michelle, 158, 160 Lewin, Kurt, 2 Life, Inc., 217 links: and conversation economy, 4, 212–13; on microsites, 175, 185; Newsvine’s moderation of, 131, 135; Rachel Maddow Show’s use of, 185–87, 195; and video player, 64, 69, 74, 195. See also Newsvine. com: seeds Linux, 127, 155–57, 174, 273n5, 249, 255 liveness, 3, 23 live streams, 45, 99–100, 118–19, 269n16 Lobato, Ramon, 27 local media, 12, 20, 22, 48 local news. See local media local teleologies, 168–69, 252–53; and the Maddow Blog, 177–78, 192; and Newsvine, 168–69, 204–205, 239; and the Rachel Maddow Show, 173, 177–78, 196, 205–206; and show fronts, 176–78; and social distribution, 215–16, 245, 274n1 (chap. 11); and systems, 168–69, 204–206; and Today.com, 242; and web producers, 186 long tail, 82, 165, 273n7 Long Tail, The, 165 Look, The, 217 Lost in the Vines, 150

Lotz, Amanda, 7, 25–26, 31, 268n12 Lullabot, 258 M2 architecture, 146–48, 194, 225, 272n1 M3 architecture, 17, 194, 223–27, 272n1 Maddow Blog, 196; and blogs for other shows, 175, 204; and the cable site (standalone), 259–60; comment moderation on, 235; creation of, 188–208; development delays, 173–77, 185–87, 194–95, 227; and editorial process for the show, 191–92, 199, 259–60; faking blogware, 185–87, 189, 195, 204; and the Newsvine platform, 194–96, 207–208; on-air exposure of, 244; social distribution of, 197–98, 244, 260; and social media integration, 196–99, 204; and third-party tools, 183, 188–89, 196–99; and user participation, 191, 198–99 Maddow, Rachel: editorial process, 171, 191, 193, 274n3 (chap. 10); fan sites, 183, 194; hire of, 66, 172, 181; influence at MSNBC, 172, 188–89, 274n1 (chap. 10); media and guest appearances, 183–84, 188; MSNBC .com show front, 176–77, 205; prioritization of the web, 16–17, 172–73, 176–77, 181, 188–93; Redmond visit, 180; “Talk Me Down” Q&As, 182; Twitter following, 170, 200; vantage point, 118, 176–78; and Vimeo, 197 Making News (Tuchman), 1, 211 market segmentation, 97, 165–66, 168, 202–203, 217, 246 Marres, Noortje, 31 Martin Blogs, 194, 225, 226 Marvel, Mark, 99 mass mediation, 231, 244–45, 255, 256 mass reach, 231, 244–45, 256 Maurone, Jeff, 97 McClellan, Martin, 194 McKinney, Tricia, 192 McMillian, John, 20–21, 29 media activism, 2 media choice, 31–32, 213–15, 217, 222

INDEX 312

media concentration, 11, 23, 107, 177 media conglomerates, 13, 106, 109, 169, 177–78, 275n4 media consumption: and media choice, 31–32, 213–15; norms surrounding, 32; online shift, 12; and platforms, 20; and polarization, 32, 238; and produsage, 216; representations of, 24–25; as scholarship focus, 6, 7, 19; and social distribution, 245. See also audience measurement media content. See content media culture, 32–34, 41, 72, 251, 257, 268n12 media distribution systems. See distribution systems media economy, 5, 11, 115 media industries: and distribution, 22–23, 257; and institutionalized audiences, 214–15, 250–52; in power-role theory, 41; structural characteristics of, 108–109; and systems of knowledge, 211, 253; and 2008 financial crisis, 156–57; volatility of, 13, 156–57, 177 media organizations. See organizations media production: and digital technologies, 48; and distribution, 11–12, 33, 246, 255–56; and organizations, 106–9; as scholarship focus, 1–2, 6–7, 19, 108, 114; as system building, 41 media sociology: and actor-network theory, 114–15; contributions of, 1–2; and distribution, 6–7, 19–20, 75, 108, 114, 246, 257; and organizational studies, 115; problems considered by, 5, 255; professional socialization focus, 11, 106–108; and sociology of knowledge, 211, 255; structural focus, 11, 106–108 media work: actor-network theory and, 114–15, 270n3; as bellwether for other industries, 22–23; and distribution, 7, 22–23, 74, 105, 257; and epistemic encounters, 215; imitation and monitoring in, 107; monism of, 107; professional socialization in, 106–109, 269n1 Meet the Press, 95, 180, 219, 275n5

MegaVideo, 270n8 metadata: and playlists, 64; and syndication, 56; for video clips, 47–48, 53–54, 265n6, 56, 70–71, 80 metering, 4, 119, 270n8. See also paywalls micro-sites, 67, 72, 73, 100; Rachel Maddow Show’s appropriation of, 182–83, 185, 195; as show templates (see show fronts) Microsoft: branding, 158, 188, 207; C#, 155, 156; campus (see Redmond, WA); and Comcast, 203, 258, 260; data centers, 149, 155, 157; and interactive television, 50, 51, 53, 265n9; Internet Explorer, 58, 266n2, 268n11; and MSNBC joint venture, 15, 48–53, 101, 200, 248, 263n6, 265n10; MSNBC TV withdrawal, 51–52, 265n11; and MSN dial-up service, 49; and the MSN video player, 57; .NET, 156; Operations Support, 155–157; server operating system and software, 127, 155, 174, 270n2 (chap. 7); syndication arrangements, 58, 198, 266n1; Windows Mobile, 94, 95; and Windows 95, 49. See also MSN Microsoft Network. See MSN Mike Blogs, 194–96, 202, 225, 226 Minneapolis bridge collapse coverage, 228 mobile apps, 95, 97–98, 178–79, 198, 269n13 mobile devices: audiences on, 93, 97, 259, 268n12; display dimensions, 96; distribution to, 3, 15, 76, 93–100; formatting media for, 10, 95–96, 259; phones, 3, 15, 93–100; tablets, 3, 15, 97 mobile web browsers, 95–96, 268n11 mobile website, 94–97, 259, 268n11, 269n14 MobiTV, 94, 95, 98, 268n8 modules, 146–48, 147, 149, 160, 272n1 monitorial citizenship, 32, 267n1 Morning Joe, 180 Mozilla Firefox, 268n11 MSN (Microsoft Network): branding of, 188; and cross-promotion, 203, 266n1; dial-up service, 49; other Microsoft products and, 58–60; and the MSNBC Digital Network, 65; MSN News, 49–50, 53; transition to standalone website, 265n8; transition to

INDEX 313

web, 49–50; video player, 57, 60–61; video syndication to, 56–60, 198, 248 MSNBC (cable channel). See MSNBC TV MSNBC (co-branded organizations): branding tensions surrounding, 65–67, 72, 176, 200–204, 207, 220, 248–49; cross-promotional activities, 58–60, 59; provincialism within, 105, 109, 119, 168–69, 199; symbolism of brand, 158, 207 MSNBC cable site: under Comcast, 195, 244–45, 258–61, 274n6; MSNBC Digital Network plans for, 169, 199–204, 219; Newsvine’s role in, 169, 204, 239, 258, 260, 274n6; social media team, 259, 261 MSNBC.com (company): and Bing, 58–60; branding decisions at, 218–21, 243–44; business strategy, 165–66, 168, 200, 202–204, 216–32; and cable providers, 99–100; Creative Development team, 164, 193–94; data centers, 149, 155, 178; development cycles, 174–75, 185–86, 227–28, 275n1 (chap. 12); development delays at, 174–75, 185–86; dissolution, 18, 258–61; heterarchy within, 241–44; heterogeneous engineering by, 151–54, 176, 231, 241–44; and Maddow, 172, 180–82, 188, 205; managerial structure, 241–44; and MSN, 58–60; NBC Universal, acquisition by, 276n5; Newsvine acquisition, 16, 122, 145, 147–54, 168–69, 234, 236, 242; and television, 178, 185–86, 188–208, 219–21; traffic-sharing agreements, 90–91; vantage point, 151–54, 166–68, 174–78, 181, 199–204, 240, 253. See also MSNBC Digital Network; MSNBC joint venture MSNBC.com (website): blogging, early efforts at, 91, 167; blog redesign, 165–69, 173–75, 188, 193–96, 232–33; bootlegging, responses to, 45, 78, 248; branding tensions with MSNBC TV, 65–67, 72, 176, 200–204, 207, 220, 248–49; and citizen journalism, 152, 154, 205; comment moderation on, 232–35, 276n1 (chap. 13); cover (homepage), 65, 68, 176, 200, 249; editorial strategy, 66–67, 201; editors

of, and blogs, 222–23, 231, 233–34; editors of, and television content, 66–67, 68–69, 73, 105, 201, 220; Gadgetbox, 217; In-Game, 217; investments in, 265n10; and live streaming, 45–46, 118–19, 265n5; Minneapolis bridge collapse coverage, 228; mobile apps, 97–100, 198; mobile website, 94–97; MSN, cross-promotion with, 266n1; MSNBC TV, editorial independence from, 50–51, 52, 65–66, 201, 207; MSN News, origins as, 49–50, 53; and NBCNews.com, 67, 244, 258–61; news cycle of, 67–69; Newsvine, relationship to, 122, 164–69, 193–94, 205, 225–27, 238–41, 273n6; Newsvine blogs on, 167–68, 193–96, 202, 225–27, 226, 231–36, 240, 244; Newsvine platform integration, blogs and branded sites, 167–68, 193–96, 202, 208, 225–27, 226, 231–41, 276n4; Newsvine platform integration, initial, 152, 159; question boards, 165, 239–40; redesign, 174–75, 193–94, 222, 227–29; SkyPad, 14, 17, 227–30; sub brands, 217, 219–20, 244, 275n6; Technolog, 217; television-­related content on, 67–69, 72–73, 176, 200–201, 205, 219–21, 248; traffic-sharing agreements, 90–91; user base, 149, 160; and user participation, 152, 159–68, 176, 183, 232–41, 276n1 (chap. 13); video traffic, 4, 77. See also micro-sites; show fronts; story pages MSNBC.com East Coast offices, 14; and branding, 217–18; and MSNBC TV web staff, 189–91; vantage point, 180–81; web producers assigned to television, 178–82, 185–86, 190–91, 200, 274n4 (chap. 9), 274n1 (chap. 10); and West Coast offices, 181–82 MSNBC Digital Network (web properties owned by MSNBC.com): brands, proliferation of, 178, 202, 216–31, 241, 243, 245, 267n6; cable site (standalone), 202–204; and development tasks, 164; flexible software architectures, 222–32, 241, 243–45; NBC News Digital takeover

INDEX 314

MSNBC Digital Network (continued) of, 14, 118, 276n5, 258–61; NBC Universal, acquisition by, 258–61, 263n5, 276n5; Newsvine acquisition, 16, 122, 145, 147–54, 168–69, 217, 234, 236, 242; Newsvine technologies and, 223–27, 229–30, 239–41, 276n4; origins of, 65, 151, 217; SkyPad, 14, 17, 227–30; and television, 178, 185–86, 188–208, 219–21, 248–49. See also Breaking News; EveryBlock; MSNBC.com (company); MSNBC .com (website); Newsvine (company); Newsvine (software platform); Newsvine. com (website); Today.com MSNBC joint venture, 263n6; dissolution of, 52, 118, 203, 258–61; dot-com crash, 151; financial terms of, 51, 265n10; history of, 48–53, 200, 261; MSNBC TV, departure of, 51–52, 151, 265n11. See also MSNBC.com (company); MSNBC Digital Network MSNBC TV (cable channel), 14, 18; America’s Talking, origins as, 50, 53; and Avid software, 52–53; blogs, 167, 173–78; branding tensions with MSNBC. com, 65–67, 72, 176, 200–204, 207, 220, 248–49; editorial strategy, 65–66, 267n5 (chap. 4); and embeddable player, 77; flipping televised video to the web, 45–54, 178, 190, 199–205, 248–49; heterogeneous engineering by, 201–202; and Huffington Post, 201; Lean Forward campaign, 66, 205; live streaming and, 45–46, 118–19, 265n5; Maddow, influence on, 172, 188–89; micro-site on MSNBC.com, 67, 182–83, 185, 195; Microsoft, divestment of, 51–52, 265n11; MSNBC.com website, editorial independence from, 50–51, 52, 65–66, 207; Newsvine, personalities on, 161, 182; Newsvine, relationship to, 16, 122, 165, 169, 235, 238; and Newsvine 3.0, 235, 258, 259; Newsvine-­powered blogs, 165, 174–75, 193–96, 202, 204, 226; offices of, 171, 190, 265n12; relocation of, 265n12; and TV Everywhere, 118–19, 259;

vantage point, 203; web content from, 175, 202–203; web staff, internal, 189–93, 202–203, 219; YouTube bootlegs of, 45, 78, 119, 248, 270n7. See also MSNBC cable site MSOs (multiple-system operators). See cable providers Muller, Judy, 273n2 multiple-system operators. See cable providers MySpace, 128, 146, 273n7 MySQL, 127–28 Napoli, Philip, 24, 26, 30, 214–15 NBC. See NBC Universal NBC.com, 48 NBC News, 14; and Avid software, 52–53; blogs, 167, 173, 226, 275n5; branding, 158, 207, 219–21; and cross-promotion, 58–60, 59; Elkhart Project, 193–94; Gans on, 106; internal web staff, 274n4 (chap. 10); and live streaming, 45–46, 265n5; on MSNBC.com, 68–69, 219–21 (see also Today.com); and MSNBC joint venture, 48, 51–53, 58, 65–66, 101, 199, 219–21; and NBCNews.com, 67, 244, 258–61; Newsvine, personalities on, 160, 161; Newsvine, relationship to, 16, 122, 165, 169, 235, 238; and Newsvine 3.0, 235; and TV Everywhere, 259 NBCNews.com, 67, 244, 258–61 NBC News Digital, 14, 18, 118, 238, 240, 258–61, 265n5 NBC Nightly News: blog, 173; branding issues surrounding, 219–20; and the mobile website, 95, 97; Newsvine Discussion Club, 165; placement online, 68–69, 219–20, 275n6; production of, 171; time zones, 265n7; web producer for, 180 NBCUniversal. See NBC Universal NBC Universal: and Comcast, 52, 100, 118–19, 258; and General Electric, 206; MSNBC.com, acquisition of, 258–61, 263n5, 276n5; MSNBC.com, staff relationships with, 188–93, 202–203; and MSNBC joint venture, 15, 48–53, 60, 101,

INDEX 315

248, 263n6, 265n10; MSNBC TV, buyout of, 51–52, 265n11; MSNBC TV, relocation of, 266n12; and TV Everywhere, 118–19, 259; video-intensive websites, 48 Neff, Gina, 115, 116; Venture Labor, 23 .NET, 156 Netflix, 22, 34, 249 network forums, 29–30, 215–16, 253 network news, 12, 45, 170–71, 265n4. See also evening newscasts; NBC News News: A Reader (Tumber), 6 news explorers, 153, 154, 217 news magazines, 12, 220 newsmagazine shows, 171, 220 NewsNation with Tamron Hall, 67, 180 newspapers, 12, 21–22, 85, 173, 185, 213 news values, 2, 106 Newsvine (company): branding, 149–51, 159, 161; business model, 143–45, 239; cable site (standalone), 204, 258, 260, 274n6; comment moderation stance, 132–34, 232–33; finances, 125, 148–49, 151, 154, 205; investors in, 124, 149; and local teleology, 168–69, 239–41; MSNBC. com, acquisition by, 16, 122, 145, 147–54, 168–69, 217, 234, 236, 242; MSNBC.com, blogs for, 165–69, 174–75, 193–94, 202, 204, 208, 231–36, 240, 275n5; MSNBC. com, role within, 14, 17, 164–69, 193–94, 204, 208, 238–41; NBCUniversal, acquisition by, 263n5, 276n5; New York Times partnership, 144–45, 160; origins of, 122–45; speed to market, 156, 194; staff size, 133–34, 156–57, 239–40; and vantage point, 122, 124, 153–54, 174, 233, 253 Newsvine (software platform): and cable site (standalone), 204; and chromes, 223–25, 230; commenting system, 233–41, 258; and cross-posting, 231–32; and the Elkhart Project, 193; flexibility of, 194, 223–27, 229–32; groups on, 235–41; and the Maddow Blog, 193–196; and MSNBC. com blogs, 92, 174–75, 188, 193–96, 202, 222–23, 231–36, 240, 275n5; and MSNBC TV programs, 193–96, 202, 204; M2

architecture, 146–48, 194, 225, 272n1; M3 architecture, 17, 194, 200, 223–27, 272n1; and Newsvine.com, 167, 223, 235–36, 239–41; and Newsvine 3.0, 235–41, 258, 259, 276nn3–5 (chap. 13); products built with, 225–27; Vimeo support, 200. See also Newsvine.com (website) Newsvine Anti-Spam user group, 142 Newsvine blog, 129, 131–32, 136, 149–51, 161, 164 Newsvine.com (company). See Newsvine Newsvine.com (website): abuse, policing of, 130–31, 134–36, 138–42; beta features, 140, 147, 238; beta testing of, 125, 128–30, 131–32, 238, 270n1; codebase, 127–28, 155–57, 239–41; columns, 125, 135, 150, 162, 167, 225, 235–36; comments, and citizen journalism, 124, 239; comments, moderation of, 132–36, 139–40, 162–63, 235; comments, quality of, 125, 237–38; comments, visibility of, 130, 140; concept for, 123–26; cover (homepage), 125–26, 130, 135, 145, 146, 147, 161; development delays, 162–63, 240; “Discuss on Newsvine” button, 159–64, 273n6; events, 136; Greenhouse, 135–36, 142, 145, 161–62; groups, 235–41, 276nn2–3 (chap. 13); Guides, 143; heterogeneous engineering of, 126–28, 129, 134, 137–38, 140, 143; language restrictions, 138; layouts, 146; Leaderboard, 161; M2 architecture, 146–48, 147, 148, 272n1; moderation policies (see also Newsvine Code of Honor), 132–34, 138, 141–42, 162; moderation technologies, 130–31, 134–36, 138–41, 235, 272n11; modules, 146–48, 147, 149, 160, 194, 272n1; and MSNBC.com comments, 159–64; and the MSNBC Digital Network, 16, 65, 122, 164–68, 202; and Newsvine 2.0 (see M2 architecture); and Newsvine 3.0, 235–40, 258, 259, 276n3 (chap. 13); origins of, 122–45; profanity filter, 130–31, 204–205, 240, 271n6; profile pages, 146–47, 148; public launch, 132, 150; and the Rachel

INDEX 316

Newsvine.com (website) (continued) Maddow Show iPhone app, 98; Random Acts of Vineness, 137–38; repurposing of software, 194; revenue-sharing program, 129, 130, 150–51, 162; section fronts, 125, 126, 130, 135, 145; seeds, 125, 135, 141, 145, 160, 162–63, 273n6; servers, 149, 155, 223, 270n2 (see also solution stacks); share button, 144–45, 160–64; solution stack, 126–28, 154–57; Tax Man Project (see M2 architecture); tokens, 131, 140, 161; trusted user system, 131, 135–36, 140, 145; user accounts, 132, 134, 135, 140–41, 160; user agreement, 138; user invites, 129; Vineacity, 136–38, 137; voting, and content moderation, 138, 139–40; voting, and content visibility, 125–26, 130, 132, 140, 271n5; voting, and the Greenhouse, 135 Newsvine Code of Honor, 133; development of, 132; gaming of, 141–42; and the Greenhouse, 135; language restrictions in, 138; and MSNBC.com users, 161–62, 164; purposes of, 132–34 Newsvine developers: and cable site (stand­ alone), 204; comment system design, 233; and Disney, 123, 163; and MSNBC .com, 149, 153, 162–63, 239; and the M2 architecture, 146–48; and the M3 architecture, 223–27; as a small team, 156–57, 239–40; and sports, 125; and user base, 128–30 Newsvine Guides, 143 Newsvine moderators: and the Greenhouse, 135, 142; and language restrictions, 138; and Newsvine gangs, 141; paucity of, 134, 143–44, 271n9; tools used by, 140–41, 204; users, work with, 138, 142, 234 Newsvine Next Level, 136 Newsvine user base: and beta testing, 128–29; culture of, 137–38, 163; heterogeneous engineering of, and Newsvine acquisition, 151, 154, 161–64, 234, 273n3 (chap. 8); heterogeneous engineering of, other, 129, 134, 137–38, 143, 145, 162; and language

restrictions, 138; and MSNBC.com users, 160–64, 208, 235; and news explorers demographic, 153, 154; and Newsvine 3.0, 235–41, 276nn2–3 (chap. 13); norms of, 136, 138, 142, 143, 164, 205, 234, 241; as online community, 129, 132–33, 136–38, 140, 160, 164, 234, 236; public, 130–43; self-policing by, 129, 132–33, 139–40, 143, 161, 205, 234; size of, 134, 144, 148, 153 Newsvine users: abuse by, 131, 140; discussions among, 134; and the Greenhouse, 135, 142, 145, 161–62; invites for, 129; moderation by, 134, 139, 162; moderation of, 140–41; MSNBC, exposure on, 150–51; MSNBC, reactions to acquisition, 150, 272–73n2; new, 136, 142–43, 145, 161–64, 237; and the Newsvine Code of Honor, 132–33; pro-­social behavior by, 142–43, 164, 235; recruitment of and by, 130; status of, 126; trusted and untrusted, 131, 135–36, 140, 145 Newsweek, 12, 90, 106, 220 New York City: book vendors in, 21, 26–27; MSNBC.com offices in, 45, 46–48, 54, 70, 178–79, 186, 217, 229; MSNBC TV offices in, 53, 122, 198, 265n12; NPR studios in, 192; Redmond, WA, coordination with, 46–48, 53–54 New York Times: API, 30; bulletin boards, 272nn13–14; community features, outsourcing of, 144, 272n13; and digital word of mouth, 91; Maddow profile, 188; Newsvine partnership with, 144–45, 160; paywall, 44; and Schiller, Vivian, 274n4 (chap. 10); social media partnerships, 144; traffic-sharing agreements with, 90 Nexidia, 10, 81–83, 88–90, 253 Nielsen ratings, 24–25. See also audience measurement Nightly News Discussion Club, 165, 225 Nizan, Amit, 216, 218 norms: and conditions of congregation, 22; and distribution systems, 15, 34–35, 204; of journalism, 11, 23; of media consumption, 32; of Newsvine, 136, 138, 142, 143,

INDEX 317

164, 205, 234, 241; of online spaces, 128. See also socialization NPR (National Public Radio), 30, 192, 274n4 (chap. 10) NYTimes.com. See New York Times NYTimes Forums Refugees user group, 272nn13–14 Obama, Barack, 81, 157 obduracy, 40–41 object-centered sociality, 212 object-oriented programming, 62, 223–27, 266n3 obligatory points of passage, 90 O’Donnell, Lawrence, 66, 74, 202 off-site distribution, 15, 76, 100. See also cross-platform video distribution; embeddable player; mobile devices Olbermann, Keith, 66, 175–76, 181, 200–201 on-demand viewing, 24, 33, 250–52 online communities. See communities Ooyala, 57, 250–51 open source, 127, 134, 154–57, 212, 258 operating systems, 126–27, 250, 270n2 (chap. 7) Opera web browser, 268n11 organizational hierarchy. See hierarchy organizational studies: and heterarchy, 11, 109–11, 119, 177, 204, 206, 243, 244; and science and technology studies, 11, 111, 112, 115–16, 215, 247, 252–53; and vantage point, 120 organizations: boundaries of, 109–11, 119–20, 120, 205–208, 245, 252; forms, 109–11, 115–17, 241–44, 247, 252–53; logics within, 110–11, 119–20, 177–78; and information technologies, 274n5; provincialism within, 106, 109–21, 168–69, 177, 205–206, 241–44, 261. See also heterarchy Parks, Lisa, 7, 35, 249–50, 263n1 (chap. 2) path of content. See voltas (paths of media content) paywalls, 4, 9, 44, 251. See also metering

peer production, 115, 212–13, 254–55 permanently beta organizations. See heterarchy Perren, Alisa, 7, 19, 248 personalization, 24, 28–29, 146 Pfaffenberger, Bryan, 44–45 phone browsers. See mobile web browsers PHP, 127, 155–56, 271n4 Pinch, Trevor, 55–56, 112, 269n2 piracy. See bootlegging Planet Money, 192 platforms: algorithms, curation by, 28–29; and audience segmentation, 20, 255–56; and business models, 12, 255–56; and copyright enforcement, 27; and media choice, 31–32; politics of, 6; and public participation, 29–31, 157–59. See also content management systems; social media; software architectures playlists, 64, 70–72, 73–74, 197, 231 Plesner, Ursula, 112–13, 270n5 podcasts, 48, 94, 183 political activism, 31–32, 72, 79, 257, 267n1. See also social movements postal service, 20, 22, 247 postbureaucratic organizational forms. See heterarchy Postigo, Hector, 34 post-industrial organizational forms. See heterarchy postmodern organizational forms. See heterarchy postmodern systems. See under systems Powerwall, 266n1 primetime programming: and News­vine infrastructure, 16, 122; placement online, 67–69, 72–73, 200–201; web resources assigned to, 179–80, 274n4 (chap. 9) Prior, Markus, 32, 72, 214, 267n1 produsage, 212 profanity filters, 130–31, 204–205, 240, 271n6 professional socialization. See under socialization projects: and organizational forms, 110, 116, 206, 247; teams, 60, 110, 241–44, 252; tradeoffs in managing, 243, 275n1 (chap. 12)

INDEX 318

provincialism: and branding, 218–21; and distribution systems, 14, 15–16, 35, 62, 216, 231, 245, 256; and hierarchy, 241–44; in organizations, 101, 105–106, 109–21, 168–69, 199, 204, 241–44, 252–53; strategic support for, 242–44; and systems of knowledge, 216, 253. See also local teleologies provisional settlements, 118, 277n2 public discourse, 29–31, 44–45, 140, 246, 251 public participation: distribution and, 29–31, 72, 257; news sites and, 152, 157–59, 163–65, 176, 192, 232–36, 239; and the Rachel Maddow Show, 191, 198–99 publics, 20–22, 25–26, 29, 31. See also communities publishing platforms. See content management systems punctualization, 169 Putting Reality Together (Schlesinger), 170 question boards, 165, 239–40 Rachel Maddow Show, 14, 16–17; brand, 171; heterogeneous engineering by, 170–208, 254, 275n6; launch, 172, 181; link blog, 185–87; Maddow Blog, promotion of, 244; mobile apps, 98, 198, 269n13; and the mobile website, 95; production of, 170–73, 190–91; on social media, 98, 182–85, 197–99, 260; sub­domain for, 67; success of, 170, 172, 259–60; “Talk Me Down” segments, 182; and third-party tools, 182–85, 186, 188–89, 190, 197–99, 204; user participation, 182–85; vantage point, 122, 176–80, 193, 194–96, 253; web-original content, 191–92, 197–99, 202, 274n3 (chap. 10); web presence, pre-blog, 170–87, 204; web producers, 178–82, 189–93, 274n1 (chap. 10). See also Maddow Blog; Rachel Maddow Show staff Rachel Maddow Show staff: and branding, 218, 275n6; and cable site (standalone), 259–60; expansion of, 189–93; and the

Newsvine platform, 194–96; and online publishing tools, 185, 188–89, 194–96; and other shows, 203; production work, 171–73, 190–92, 274n3 (chap. 10); reading in, 173; and show fronts, 176–77, 205; vantage point, 122, 176–80, 193, 194–96, 253; web producers, 178–82, 189–93, 196–97. See also Maddow Blog; Rachel Maddow Show radio, 22, 171, 192, 197 RAID (Redundant Array of Independent Disks), 155, 273n4 Random Acts of Vineness, 137–38 ratings: competition for, 48–49, 65–66; and engagement, 214–15; for MSNBC TV, 65–66, 200; for the Rachel Maddow Show, 172, 176, 180–81, 189, 218. See also audience measurement reading in, 173 recalcitrance, 39–42, 117, 204–205, 253; of advertisers, 88; of brands, 207, 232; of broadband market, 51; of the “Discuss on Newsvine” feature, 160–64; of Disney, 124; of embeddable video, 83–88; of flexible architectures, 232–35; and heterogeneous engineering, 40, 42, 206; and local teleology, 168–69, 206, 239; and mobile distribution, 94, 96, 97; of MSNBC.com, 154–56, 205, 233; of Newsvine, 156, 208, 240; and the Newsvine user base, 151, 154, 238; of the profanity filter, 204–205, 240; of the Rachel Maddow Show, 182, 200; and reverse salients, 178; of the story page template, 186; and time delay, 45–46; and traffic-sharing agreements, 91; and translation, 113; and the 2008 financial crisis, 156; of users and audiences, 45–46, 81, 88, 94, 231; video player as response to, 65; and the volta, 100 reddit, 27, 124 Redmond, WA: developers in, 174, 177; Maddow’s visit to, 180; Microsoft and MSNBC.com offices in, 48, 49, 50, 51, 53, 54; web video editors in, 47–48, 64, 265n7 Red Tape Chronicles, 222–23

INDEX 319

Rescuing Prometheus (Hughes), 247 responsive distribution, 28–29, 257 retransmission fees, 45 revenue sharing. See under Newsvine.com (website) reverse salients, 41, 168, 176–78, 253 Rosen, Christine, 33–34, 72 Rosen, Jay, 13, 245 RSS, 74, 94, 127 Ruge, Mari, 2; “The Structure of Foreign News,” 2 Safari, 268n11 Sager, Ian, 184, 234–35 Scarborough, Joe, 267n5 (chap. 4) Schlesinger, Philip, 170; Putting Reality Together, 170 Schudson, Michael, 32, 107, 267n1; The Sociology of News, 6 Schultz, Ed, 66, 202 science and technology studies: and infrastructure, 8; media distribution, framework for examining, 8, 14, 254; and the mundane, 55–56; and postmodern systems, 11; social construction of scientific knowledge, 113; sociology of knowledge, 211–12, 253, 254–56. See also systems scholarship Scoop, 217, 275n2 (chap. 11) Scout Media, 125 scripting languages, 126–27, 243, 271n4 scripts (television), 47–48, 52, 53, 265n6 search: and conversation economy, 4–5, 79, 212–13; engines, 4, 5, 79, 80, 212–13; optimization, 7, 10, 135 Search, The (Battelle), 3–4 Seattle, WA, 16, 124, 125, 138, 149, 178, 222 Second Avenue Partners, 124, 149 SEO (search engine optimization), 7, 10, 135 section fronts, 62, 96, 125, 135, 145 SendThisFile, 198 server software, 126–27, 270n1 share buttons, 5, 92, 144–45, 160–64, 272n12 sharing: and audience measurement, 214–15, 250; capitalizing on, 5, 17, 78–81; and

conversation economy, 4–5, 79, 166–67, 248; and epistemic encounters, 213, 255; of files, 7, 27; by news organizations, 5, 7, 17, 213; and search results, 4; and social media, 4, 79, 166–67, 214–15, 250 show blocks, 47 show fronts, 63, 195; abandonment by some shows, 195–96, 195, 202; maintenance of, 178, 185; marginalization of, 67, 200, 219, 249; on mobile website, 96, 97, 268n11, 269n14; playlists for, 64, 71; recalcitrance of, 205; static nature of, 175–77, 182, 190, 200; and video player, 62, 175–76 Silverstone, Roger, 31, 35 Silverton-Peel, Vanessa, 173, 197 Sims, Mike, 157 Singer, Jane, 232 SixApart, 91, 167 60 Minutes, 220 SkyPad, 14, 17, 227–30, 243, 258 Slade, Mike, 124 Slashdot, 124, 129 Slate.com, 50 smartphones. See mobile devices Smith, Jamil, 197 social construction of technology (SCOT), 112, 116, 269n2 social distribution: and epistemic encounters, 213, 215, 255; and monitorial citizenship, 32, 267n1; and Newsvine, 166–67; for television shows, 179, 245, 260, 274n1 (chap. 11); and time delay, 45–46, 248. See also conversation economy socialization: of MSNBC.com users, 239; of Newsvine users, 138, 140, 145, 162, 164, 239; professional, 11, 106–109, 269n1. See also norms social media: algorithms, curation by, 28–29; and conversation economy, 4, 79, 166–67, 183, 213; and engagement, 214–15; and monitorial citizenship, 32, 267n1; MSNBC TV shows, use by, 74, 98, 182–85, 178–79, 197–99, 260; news content, encountering on, 70; news organizations, use by, 5, 7, 152, 213, 259–61; niche vs.

INDEX 320

social media (continued) general interest, 165–66, 184, 273n7; participatory affordances of, 30; and public discourse, 246; and publishing tools, 12, 157–59; and the Rachel Maddow Show, 176–77, 183–85; social network sites, 4, 152, 165–66; social news sites, 16, 124, 130, 152; as technologies, 152, 164–67, 185. See also Facebook; sharing; Twitter social movements, 2, 107 social news sites, 16, 124, 130, 152 social norms. See norms sociology of knowledge, 211–12, 253, 254–56 Sociology of News, The (Schudson), 6 sociotechnical systems. See systems software architectures: and branding, 223, 225–27, 230–32; developer familiarity with, 156; flexibility of, 69, 70, 223–32, 241, 243–45, 275n2 (chap. 12). See also M2 architecture; M3 architecture; SkyPad software developers. See web developers software frameworks. See frameworks solution stacks, 126–28, 154–57, 271n3 Sorenson, Erik, 66 spam, 130, 131, 135–36, 138, 142, 235 spambots, 131, 135 sports, 44, 125 spreadability, 5, 256 Sprint, 94 SQL, 127–28 SQL Server, 127–28 stakeholders, 3, 31, 35, 89, 204, 206, 214 Stark, David: complex adaptive systems, 42–43, 177; heterarchy, 115, 116, 243; provisional settlements, 118, 277n2 startup companies, 124, 125, 126, 127, 163 state oppression, 26–27 Stearns, Randy: geofencing, 84; MSNBC TV, 201; the Rachel Maddow Show, 181, 186; video player, 60, 61, 78 Stelter, Brian, 91 story pages: and mobile website, 96; Newsvine comments on, 234; the Rachel Maddow Show, appropriation by, 185–87;

redesign of, 69, 174, 194, 227–30; and video player, 62, 71–72 structure: and distribution, 19, 77, 249, 250, 252; sociological focus on, 11, 108, 114 “Structure of Foreign News, The” (Galtung and Ruge), 2 STS. See science and technology studies sub-clips. See video clips subdomains. See URLs Sullivan, Bob, 49, 222 Summer Writing Challenge, 142 Sunstein, Cass, 25, 32, 72, 214 surveillance, 24, 29, 184, 250–52, 257, 263–64n1 (chap. 2) syndication, 56–60, 198 system building: and enrollment, 42, 206–207; and the Greenhouse, 135–36, 145; as heterogeneous engineering, 41–42, 116–17, 206–208, 242; and local teleology, 168–69, 204, 239, 253; by MSNBC.com, 151–54, 241–45; by Newsvine, 122–69, 239–41; and norms, 128; and organizational forms, 115–20, 204–208, 242–44, 252–53; postmodern style of, 11, 101, 115–16, 119, 169, 247, 264n2 (chap. 3); in power-role theory, 41; provincialism and, 15–16, 116–21, 168–69, 204–208, 242–44, 252–53, 256, 261; by the Rachel Maddow Show, 170–208; and reverse salients, 177–78; by transparent intermediaries, 88–90; and vantage points, 116–21, 124, 182, 205–208, 252–53; and video player, 62; and video syndication, 58–59; volta as, 8–9; by YuMe, 87–88. See also heterogeneous engineering systems: complex adaptive nature, 43, 177; conceptualizing, 17, 117–21, 168–69; and distribution routes, 9; entropy in, 40–43; and heterogeneous engineering, 8–10, 40–43, 206–8, 252; interactions between, 42, 204–8, 253–54; isometric pressure in, 117–20, 120, 154, 253–54, 256; of knowledge, 211–12, 216, 254–56; and local teleology, 168–69, 204, 253; modern, 119, 247, 270n6; obduracy of, 40–43;

INDEX 321

postmodern, 11, 101, 115–16, 119, 169, 247, 264n2 (chap. 3); relativistic nature of, 119–20, 120, 169, 199, 203, 252–54, 274n1 (chap. 11); sociotechnical nature, 14, 39–45, 235, 249; vantage points on, 119–20, 120, 168–69, 182, 203, 252–53, 274n1 (chap. 11). See also distribution systems; system building systems of knowledge, 211–12, 216, 254–56 systems scholarship: and actor-­network theory, 113, 119, 270n9; and common system goals, 101; and heterogeneity, 138, 252; and media distribution, 8, 11–12, 14–15; and organizational forms, 115, 119–21, 204, 245, 252–53; and the sociology of knowledge, 254, 256 tablets. See under mobile devices talk radio, 171 Tang, Calvin, 124–25; and moderation, 271n9; and Newsvine acquisition, 150; and Newsvine Code of Honor, 132; and Newsvine concept, 123; on user invites, 129 Tax Man Project. See M2 architecture TBS, 50 teases, 47–48 Technolog, 217 technological dramas, 44–45 technologies: heterogeneous nature, 41, 43–44; and news judgment, 106; regulation through, 44–45; social construction of, 61–62, 112, 116; theories of, 43–45, 55–56, 111–17; television studies, 7, 8, 12 Telligent Systems, 91, 167 thePlatform, 259, 260, 261 Thomas, Chris (pseud. Killfile), 138, 150, 151 thumbnail images, 48, 62, 64, 80 Thurman, Neil, 158–59, 233 Tillinghast, Charles: on blogs, 223; on branding, 67, 217, 219–20, 255; on business strategy, 166, 217; on cable providers, 45; on cable site (standalone), 202–203; on the embeddable player, 78, 79; and Mad-

dow Blog, 274n1 (chap. 10); on metadata, 80–81; on mobile audiences, 93; on Newsvine acquisition, 152–53, 159, 164; on Newsvine, role of, 166–69; on ownership and innovation, 243; on perishability of news content, 84–85; on SkyPad, 227–28; on Today Show demographics, 269n15; on transcript sync, 82–83 Time, 12, 106 time delay, 45–46, 99–100, 118–19, 248 Time Warner, 50 Time Warner Cable, 99–100 Today.com: audience demographics, 217, 220–21; blogs, 217, 226; and branding, 217, 219–21, 223; management strategy for, 242–43; and the MSNBC Digital Network, 65, 202, 219–21; redesign under NBCUniversal, 258–61; and social media, 213; spin-off, 65, 71, 217, 219–21, 242–43 Today Food, 217 Today Moms, 217, 226 Today Show: audience demographics, 97–98, 220–21, 269n15; brand, 219–21; mobile app, 97–98; and the mobile website, 95, 97; playlists for, 64, 71–72; time zones, 265n7; and Windows Mobile, 94. See also Today.com Todd, Chuck, 160 tracing distribution pathways, 4, 9–10, 39, 248–50, 277n1 (chap. 14) trading zones, 215, 253 traffic. See audience measurement traffic-sharing agreements, 90–91 transcoding. See flipping video transcripts, 50, 77, 81–83, 267n3 translation, 87, 113, 114, 202 transparent intermediaries: defined, 15, 76–77, 250; as heterogeneous engineers, 87–90; importance of, 250–52; for mobile distribution, 93, 267n8; and NBC News Digital, 261; Nexidia as, 81; and tracing the path of content, 256 Transpera, 88–90, 96, 98, 119, 250, 268n10 Tremor Video, 10, 58, 88–89, 250, 268n10 trolling, 130, 132, 135–36

INDEX 322

Trudeau, Gary, 59; Doonesbury, 59 Tryon, Chuck: on cable programming, 30; on conditions of congregation, 20; on cultural exports, 22; and distribution, 7; on flexible labor and leisure, 213; on media choice, 20, 31, 213; on media culture, 32–34; on surveillance, 24 Tuchman, Gaye, 2, 211–12, 254, 275n7; Making News, 1, 211 Tumber, Howard, 6 Tumblr, 166, 188, 260 Turner Broadcasting System. See TBS Turner, Fred, 29–30, 111, 114–15, 215–16 Turow, Joseph, 24, 26, 41, 44, 74 TV Everywhere, 118, 259 Twitter: account creation, 244; and Breaking News startup, 225; and the embeddable player, 71, 77, 78, 92; flexibility of, 275–76n3 (chap. 12); hashtags, 213, 245; and niche social networks, 166–67; and the Rachel Maddow Show, 98, 183, 185, 195, 197–98, 200; stakeholders in, 206; and television ratings, 214; and thirdparty developers, 128; as a third-party tool, 245; and two-screen experiences, 165; user base, 128 2008 financial crisis, 156–57, 192, 193, 239 TypePad, 91, 167, 194 underground press, 20–21, 26, 29 Unix, 155, 273n5 URLs: and the Maddow Blog, 177, 196; for story pages, 229; subdomains, 67, 177, 185, 195–96, 249; for videos, 54, 92 user-agent banning, 5, 267n4 user-generated content, 130, 152, 157–59, 183 user interfaces: abuse reporting, 134; comment-collapsing, 139; embeddable player, 77–78, 78, 79; Giant List of Scum, 142; mobile website, 96–97; moderation, 140–41; Newsvine groups, 236–38; post-authoring, 200; and SkyPad, 228–29; video player, 62–64, 197–99, 245 users: of embeddable video, 10, 15, 88; end, 10, 15, 16, 17, 83, 98, 250, 256; Facebook,

4, 28–29; as heterogeneous engineers, 231, 244–45, 274n1 (chap. 11); peer production by, 212–13, 254–55; provincial demands of, 216–18, 231, 245; and the Rachel Maddow Show, 183; restrictions on, 5; savviness of, 201; web, 4, 46, 71, 91. See also audiences; Newsvine users vantage points, 15; features of, 119–20, 120, 178, 252–53; and isometric pressure, 120, 120, 200; of MSNBC.com, 151–54, 155–56, 166–68, 174–78, 199–204, 208, 240, 253; of MSNBC.com East Coast offices, 180; of MSNBC TV, 203; of Newsvine, 122, 124, 153–54, 174, 233, 253; of Nexidia, 253; of the Rachel Maddow Show, 122, 176–80, 194–96, 253; and web presence for shows, 173–80, 199–205, 253 Viacom, 58, 99, 269n16 video clips: behind-the-scenes, 74, 197; creating, 46–48, 53–54, 190, 248, 265nn6–7; in mobile apps, 97–98; mobile devices, formatting for, 96; and show fronts, 175; sub-clips, 77–78, 79, 82, 83; and video player, 56, 63, 71, 231; and Vimeo, 197–99; web-extra, 74, 80; web-original, 74, 80, 197; wire footage, 80 video player (MSNBC.com): development of, 14, 15, 56–75, 174–75, 242–43; flexibility of, 62–64, 73, 76, 230, 231, 245; heterogeneous engineering of, 64–65; interface, 62–64, 63; phasing out of, 259, 261; on show fronts, 175–76, 196; use cases for, 70–71. See also embeddable player video players (general), 5, 14, 15, 24, 45, 57. See also video player (MSNBC.com) video transcoding. See flipping video Vimeo, 76, 197–99, 206, 245, 274n2 Vineacity, 136–38, 137 Virginia Tech shooting, 138, 150 volta (original Portguese trade route), 8–10, 39, 100. See also voltas (paths of media content) voltas (paths of media content): automation and, 74, 82–83; as complex adap-

INDEX 323

tive systems, 42–43; for cross-platform distribution, 98, 100–101; and heterogeneous engineering, 8–10, 39; for MSNBC content, 15, 46, 54, 100–101, 205, 248–49; for online video, 100–101, 205; and provincialism, 109, 205, 208; syndication arrangements as, 58–60; tracing, 4, 9–10, 39, 248–50, 256, 277n1 (chap. 14); and video player, 56 volta do mar. See volta Vox Media, 256 Washington Post, 78, 90, 273–74n3 (chap. 9) Watchlists, 136, 272–73n10 Watts, Steve, 150 Way Too Early, 180 web applications, 126–28, 154–57, 223, 271n3 web browsers: Apple Safari, 268n11; and CSS, 269n14; detection of, 250; Google Chrome, 268n11; layout engines for, 268n11; Microsoft Internet Explorer, 58, 266n2; Mozilla Firefox, 268n11; Opera, 268n11; and streaming video, 61; and web applications, 126, 128. See also mobile web browsers web developers, 13, 126, 128, 243, 268n11, 271n4. See also Newsvine developers WebKit layout engine, 96, 268n11 web producers, 1, 13, 14; and cover editors, 69, 218; digital word of mouth, 91; engagement strategies, 184; and launcher links, 64; and metadata, 70–71; and playlists, 64, 70–71; for the Rachel Maddow Show, 178–82, 189–93, 274n1 (chap. 10); television, team assigned to, 178–82, 185–86, 200, 274n4 (chap. 9); and video clips, 46–48, 54, 83, 178, 190, 265n6; workload of, 73, 179–82; web servers, 126–28, 149, 155, 243, 261, 270n2 web standards, 97, 99, 123, 268n11

web video editors, 47–48, 54, 56, 64, 82–83, 265n7 Wells, Ashley, 193–94 White, David Manning, 2; “The GateKeeper,” 1 white-label brand. See under brands widgets. See modules Wikipedia, 27, 271n8 Wikiquote, 98, 183–84 Williams, Raymond, 6 Windows 95, 49 Windows Mobile, 94, 95 Windows servers, 127, 155, 174, 270n2 (chap. 7) wire footage, 80 Wolff, Bill, 171, 179–80, 181, 189–91, 192, 193 WordPress, 188, 206, 275–76n3 (chap. 12) Yahoo!, 66, 124, 256 Yahoo! News, 66, 124 Yockey, Josh, 125; on comment systems, 233; on dual-use code, 239–40; ESPN experience, 163; on language restrictions, 138; on moderation, 138, 139; on MSNBC.com development cycles, 227; on MSNBC.com users, 164, 235; on M2 framework, 148; on the Newsvine Code of Honor, 132–33, 271n8; on Newsvine community, 132–33, 164, 232, 234–35; on the solution stack, 127–28, 155–56 Young, Stokes, 58–59, 73, 80, 81, 200, 203 YouTube: bootlegging to, 45, 78, 81, 84, 119, 270n7; embedding functionality, 76, 92; flexibility of, 275–76n3 (chap. 12); and MSNBC.com, 45, 78, 81; stakeholders in, 206; and time delay, 45; and TV Everywhere, 119, 270n7; YuMe, 10, 86–90, 119, 250, 251 Zucker, Jeff, 66, 266n5 (chap. 4) Zumobi, 97