Read only: The persistence of lurking in Web 2.0
First Monday

Read only: The persistence of lurking in Web 2.0 by Scott Kushner

Social media is supposed to be all about participation. But most users don’t participate very much. This essay argues that lurking poses a threat to the prevailing logic of corporate social platforms. It explores the leading discourses of participation and lurking in order to theorize how this threat functions, contributes to the political economy of communication in order to account for both user generated content and lurking, and examines strategies that platforms deploy in order to combat lurking and stimulate steady user participation. Finally, it speculates that platforms may be planning for a future where participation figures less centrally, thereby blunting lurking’s threat.


There went everybody
The curve’s angle
The blindspot
Digital political economy
The blindspot of the blindspot
Making lunch
Anti-lurking ordnance
Beyond the curve?



There went everybody

There has been no shortage of noise made about Web 2.0’s most salient surface feature: user-generated content. Much of the attention has come in the form of techno-utopian pronouncements of participatory media’s democratic potentials. Facebook CEO Sheryl Sandberg, in full-throated corporate optimism, proclaimed a “historical shift from the historically powerful to the historically powerless, because everyone has voice” before an audience composed of the historically powerful at the World Economic Forum in Davos (Schmidt, et al., 2015). Sandberg’s rising masses are empowered in much the same way that Clay Shirky imagines “everybody” conspiring to topple institutions, for example in Shirky’s claim that “the mass amateurization of publishing undoes the limitations inherent in having a small number of traditional press outlets” [1].

But consumer habits research reveals that a wide swath of the social media user base lurks: these users read, watch, and listen to content, but they do not contribute any of their own. Lurking stands as a constant threat to the cultural logic of participation, which underwrites social media in both economic and rhetorical terms. As Web 2.0 enters its second decade as an organizing structure for online cultural and commercial experiences, its forms and costumes grow more sophisticated. The wilds of MySpace have yielded to the manicured feeds of Facebook, while the eagerness of sharing has yielded to the sharing economy, positioning social activity as a resource to be tapped. In the context of these shifts, this essay explores the status of lurking. I begin by placing the popular and scholarly discourses of participation in contrast with existing scholarship on lurking. Bringing these two strands of thought into conversation reveals the ways that lurking plays into and defies the political economy of social media. Next, I trace the lineage of the political economy of communication and contribute a new wrinkle by arguing that social media represent a shift in the nature of the so-called “free lunch,” the programming that draws viewers, listeners, and readers to advertising-supported media. Whereas legacy media companies purchased or produced the free lunch, social media platforms rely upon users to make free lunch for one another. It is because lurking constitutes a withholding of this lunch-making function that social platforms quietly work to discourage lurking and stimulate participation. Finally, I examine one exemplary mechanism that a social media platform has deployed in the fight against lurking: the Facebook Like button. The story that this essay tells is not only about the play between participation and lurking, but also about the play between the quantitative demographic and consumer preference data that platform servers crunch and the qualitative cultural data that platform users crave. What emerges is an ambivalent cultural politics that stands in stark contrast to the enthusiastic rhetoric of “historical shifts” and democratization.

Democratic-utopian narratives like Sandberg’s and Shirky’s have echoes in one strand of the academic literature on participatory media. Henry Jenkins sees user-generated content as the heir to the fan culture that he described as a form of “textual poaching,” where fans claimed some measure of control over cultural products (Jenkins, 1992). In his early analysis of social media, users gain the opportunity to claim some measure of control over the narratives that shape their everyday experiences of culture, politics, and sport. Even in that formative moment, Jenkins was quick to point out that “not all participants are created equal,” with corporations and thought leaders more equal than the others [2]. In more recent work, Jenkins has increasingly focused his attention on the recognition “that new grassroots tactics are confronting a range of corporate strategies which seek to contain and commodify the popular desire for participation” [3]. Of course, one might question whether an analysis of social media rooted in the world of devotional fan culture is appropriate to understanding the banal contours of online engagement and disengagement (Couldry, 2011).

Tim O’Reilly, one of Web 2.0’s most visible exponents, sang a slight variation on this theme when he declared in an agenda-setting essay that “network effects from user contributions are the key to market dominance in the Web 2.0 era” (O’Reilly, 2005). Where Sandberg and Shirky spoke of democratic potentials and Jenkins described the cultural power of everyday acts of writing, O’Reilly made no bones about the fact that Web 2.0 represents a business model. O’Reilly later expanded on his vision for Web 2.0, and claimed that its power lay in what he calls collective intelligence (O’Reilly and Battelle, 2009). Collective intelligence describes knowledge that can be mined from content provided by users. Extracting economic value from collective intelligence is the hard work that successful Web 2.0 companies must undertake: “Collective intelligence applications depend on managing, understanding, and responding to massive amounts of user-generated data in real time.” Missing from O’Reilly’s list are “coaxing” and “compelling.”

The notion that media platforms coax and compel their users to do things is the key insight that drives a diverse set of scholars who critique Web 2.0. Surveillance studies theorists reveal the ways that social media platforms track users’ behavior using sophisticated data collection apparatuses. Mark Andrejevic deftly sidesteps the questions of privacy that often characterize popular responses to internet tracking in order to concentrate attention on the power structures that surveillance imposes [4] and focuses on “the fact of private ownership and control of the network” that enables social media platforms to “control personal information” that their platforms collect [5]. As Robert Gehl notes, this control allows platforms to imagine users as elements in a vast affective “superprocessor” (Gehl, 2011). The power that surveillance studies critiques is real, but it rests on an assumption that must be tempered by a banal social fact of digital culture.

What the surveillance studies critique and all of these popular and academic perspectives share is the assumption that users are inclined to engage with Web 2.0 platforms in a manner that matches the rhetoric: they assume that the historically powerless are declaiming, that “everybody” is publishing widely, that fans are writing back to institutional cultural producers, that user contributions are waiting to be transformed into collective intelligence, that users are eagerly processing and producing the cultural data that fills news feeds and mobile screens. In short, the legend of social media bumps up against the cold reality of empirical evidence: most people just don’t participate all that much.



The curve’s angle

“Your only real choice here is in how you shape the inequality curve’s angle,” writes Jakob Nielsen (2006) in one of the seminal texts on participation in Web 2.0 spaces. This inequality curve describes a digital divide, but it does not describe the differential access to networked devices or resources. Rather, Nielsen’s inequality curve describes differentials of participation in networked spaces that are marked as participatory. Importantly, Nielsen is not urging users to participate more. Instead, he is addressing platform owners and UX designers, and he challenges them to think about how they can compel or coax users to participate more. What is curious here is that Nielsen deploys the word “inequality” to describe a dynamic that he sees driven by user choice: “If lurkers want to contribute,” he writes, “they are usually allowed to do so” (Nielsen, 2006). Nielsen’s tacit assumption is that lurking is not a matter of social exclusion, but rather of indifference. He places the burden of shaping the curve or producing user participation in social media squarely on the shoulders of the party for whom that production is most vital: the platform.

Nielsen’s work on participation inequality is one key touchstone in critical assessments of lurking in Web 2.0, the practices of consuming content without generating any. Others have constructed lurkers as problems to be solved (Preece, et al., 2004; Nonnecke and Preece, 2003) or poorly understood participants whose contributions as listeners are essential to the social fabric of the Web (Crawford, 2011, 2009). Meanwhile, scholars investigating the broader practices of non-participation have concentrated on the social and technical complexities of leaving a social media platform (Brubaker, et al., 2016); the philosophical implications of no longer contributing (Kushner, 2011); the ethics and politics of abstaining from social media platforms (Portwood-Stacer, 2013, 2012); the political potentials of non-participation (Casemajor, et al., 2015); and the possibility that the use and non-use of technologies are wrapped up in technical, legal, and social norms (Levy, 2015), organized within complex processes of establishing social status (Thorén and Kitzmann, 2015), or discursively bound up in deep histories of railroad infrastructure and regimes of common carriage (Banks, 2015). While critical scholars have been expansive in their treatments of lurking, Nielsen’s work has remained the most influential among those who develop platforms, due perhaps to Nielsen’s stature within the usability design community. His clear formulation of lurking dynamics provides a ready counterweight to the conventional wisdom about participation. Nielsen is blunt: “Most users don’t participate very much.” The participation curve that Nielsen describes is stark, “a 90-9-1 rule” where 90 percent of users lurk, nine percent “contribute from time to time,” and merely one percent “participate a lot and account for most contributions” (Nielsen, 2006).

This participation pattern has been persistent. Nielsen formulated his 90-9-1 rule in 2006, the early moments of Facebook, the dawn of Twitter, and the time before Instagram. As much as he was responding to the rising popularity of user-generated content, he was also looking back at the long history of participatory online media. In the late 1990s, a group of researchers from AT&T Labs led by Steve Whittaker analyzed six months of Usenet postings — some 2.15 million messages — and found that “conversations in newsgroups are dominated by a minority of highly verbose participants” [6]. In the Whittaker group’s sample, 27 percent of messages represented the sole contribution of a given member over the study period, while less than three percent of the users generated a quarter of the content [7].

More recently, researchers have added nuance to Nielsen’s 90-9-1 rule, especially in the wake of a 2012 BBC study of U.K. Internet users that claimed to undermine the model. The BBC characterized 77 percent of the online population as “active in some way” (Goodier, 2012). Bobbie Johnson throws some water on the BBC’s findings, noting that the participation inequality thesis advanced by Nielsen “was never intended to dictate a single pattern across the entire Web” (Johnson, 2012). Instead, the 90-9-1 rule describes a specific user’s behavior at a specific site. On any given platform, most users lurk and a few contribute some content, while most content is generated by a tiny fraction of the overall user base. But the sports blog lurker may be the active Facebooker. The reason Nielsen’s rule focuses on participation at the level of the platform is because it is at that level that participation matters to the audience he writes for: user interface designers and software engineers. Facebook doesn’t much care how much content users generate off-platform, because Facebook can only monetize content that users provide to it directly. Though Facebook engineers might take encouragement from the knowledge that three quarters of their users participate somewhere online, it is only the content generated by users on the Facebook platform that can fuel the Facebook content machine. Indeed, the BBC survey and its responses remind us that lurking and participation matter precisely because they are to do with what content users surrender to platforms and what content they keep to themselves or share with others off-platform: it’s not really participation if it hasn’t been captured and made potentially productive [8].

Part of Web 2.0’s solution to this dilemma emerges in Neil Perkin’s critique of the BBC study. Perkin notes that the study relied on “a very broad definition [of digital participation] encompassing all forms of activity from publishing a blog post to simply clicking a Like button” (Perkin, 2012). Platforms have heeded Nielsen’s advice to “make it easier to contribute” and “make participation a side effect” (Nielsen, 2006). These efforts have echoed research by Jenny Preece, Blair Nonnecke, and Dorine Andrews that enumerated “good strategies for encouraging lurkers to participate” in order to “enhance community experiences for everyone” [9]. The Preece group implies that participation is an inherently good and socially desirable act without any accounting for the economic incentives that drive platforms to decrease idle visits by users. By redefining participation to include discrete acts such as Liking, favoriting, starring, rating, friending, and following, platforms have found new techniques to extract content from users, stimulate repeated visits to the platform, and shape the inequality curve’s angle.

These easy forms of participation matter to social media platforms for two related reasons. First, they add incremental demographic and consumer preference data to the user profiles that platforms can transform into salable eyeballs for advertisers. Second, they stimulate the continuous production of content, which ensures the ongoing presence of an audience. While the importance of data collection has been well documented in the scholarly literature, the relationship between easy participation and the stimulus of user-generated content is less understood. However the story of this relationship is part of a larger story about social media and participation, rhetoric and lurking, structured and unstructured data. It is a story about a largely one-sided resource war that illuminates a key difference between Web 2.0 and older media forms: in addition to the scarcity of attention, participatory media are driven by a need to manage the scarcity of user engagement.



The blindspot

To see the full scope of this story, it is necessary to look back at the political economy of communication, the academic subfield that theorizes how media function as part of a system where money and goods are exchanged among different parties. Communication scholars drew upon the political economic tradition to make sense of advertising-supported legacy mass communication media: television in the first instance, but also radio, magazines, and newspapers. By examining this history, it becomes possible to contextualize the political economy of digital media and, by extension, Web 2.0 platforms. Among the subtle changes that Web 2.0 introduces in the media field are a shift in the conditions of the exchanges and the structural role of one irreplaceable party: the user. It is the modification of the user’s role in Web 2.0 that makes lurking matter to platforms and that positions lurking as a type of everyday, inertial resistance to capital.

Vincent Mosco provisionally defines political economy as “the study of the social relations, particularly the power relations that mutually constitute the production, distribution, and consumption of resources, including communication resources” [10]. In plain terms, political economy is a tool that allows us to make sense of how a system works: what is given to whom in exchange for what, and under which conditions? A political economy of communication looks past the content of media programming (specific TV shows, radio programs, newspaper stories, or Web sites) in order to understand how media work as cultural and economic institutions.

The classic formulation of the political economy of communication was first crystallized by Dallas Smythe in an essay written to address what he perceived to be shortcomings in 1970s Marxist thought: Marxist scholars of communication who engaged in ideology critique were distracted by content, and Smythe sought to refocus critical attention on mass communication’s “economic function for capital” [11].

Smythe asked what was for sale in the communication space. He responded that audiences were the commodity, that they were assembled and sold by media companies, and that advertisers purchased them with cash. The assembly process involved providing a “free lunch” to viewers, listeners, and readers in order to “recruit potential members of the audience and to maintain their loyal attention” [12].

The notion of the free lunch was borrowed from American journalist A.J. Liebling, who wrote derisively of newspaper publishers who regarded news content as “a costly and uneconomic frill” [13]. Media companies paid to produce or procure this free lunch, which took the form of news, sports, and entertainment programming on television and radio, and text and images in newspapers and magazines. In Smythe’s system, the fees associated with obtaining the free lunch amounted to the cost of doing business: content was a necessary input to attract viewers, listeners, and readers who could be packaged as an audience and sold to advertisers.

In the debates that followed the publication of Smythe’s essay, the free lunch typically garnered little attention. Graham Murdock, Smythe’s first respondent, later noted that it was the American flavor of Smythe’s analysis that provoked some of the differences between Smythe’s hardline political economic stance and Murdock’s insistence that ideology critique had a place (Murdock, 2014). For all that, Murdock said less about the free lunch than Smythe did, but a glance at Raymond Williams’ transcripts of television newscasts shows how different logics of advertising correlate with different free lunch spice sets [14].

Bill Livant stretched Smythe’s critique beyond advertising-supported media, asserting expansively that “in all sectors, all of the time, the audience commodity is being made. In all sectors it is being traded, in all sectors it is being measured” [15]. What is important here is that the parameters of programming and the economies of exchange vary from one medium to the next. Indeed, Livant would later elaborate on this claim, theorizing that “surplus watching time is [...] produced, although it may be sold in another medium at another time” [16], the implication being that all media seek to produce audiences for “that film star, that pop star, that personality” [17], regardless of whether the content is provided as a free lunch or is sold directly to the viewer, listener, or reader. In the case of ad-supported media, Livant and Sut Jhally conceived of the programming free lunch as a wage paid to the viewer in exchange for “extra watching” [18].

Eileen Meehan further complicated Smythe’s schematic by introducing an additional layer in the commodity production process. It is not audiences that media companies sell to advertisers, but ratings, which she pointedly described not as “reports of human behavior, but rather as products — as commodities shaped by business exigencies and corporate strategies” [19]. Television and radio ratings, and their analogues in print, amount to manufactured products that are engineered by ratings companies so that advertisers and media companies can “agree[] on a basic method for producing measures of productivity and quality” and “move to the real business at hand — the buying and selling of audiences” [20]. More than constructed representations of the pure number of viewers, listeners, or readers, ratings broke out different demographic categories (most prominently, age and gender) so that advertisers could better target their messages and media companies could mark off premium and down-market content. Rather than abstract notions of an audience, Meehan shows that the rationalization of audience measurement is a key component of the smooth operation of the political economy of communication.

The political economy rooted in Smythe’s formulation emerged from the so-called “blindspot debate” largely intact. Respondents like Murdock, Livant, Jhally, and Meehan added nuance, but the basic structure remained. With the emergence of digital media, and especially the participatory media of Web 2.0, the Smythe model has undergone further revisions in order to account for changing conditions of consumption, production, and measurement.



Digital political economy

In turning its attention to Web 2.0, political economy has explored the emergence of new types of information that advertisers and media companies can use to target messages. In addition, in the context of social media’s emphasis on user-generated content, a renewed debate broke out over the status of audience labor. In short, the scholarly response to participatory media is largely driven by the need to make sense of qualitatively and quantitatively new data flows and shifts in the consumption and production of media content.

One of the keys to platform success in Web 2.0 is the effective monetization of the data that users generate through their interactions with digital media. Writing about the text-heavy world of Google AdWords, the small bits of “sponsored content” that appears alongside search results, Micky Lee finds Web advertisers departing from the logic of their television, radio, and print forebears. “[A]dvertisers will not find the demographic information of all Google users alone valuable. Instead, advertisers will find information about keywords useful” [21]. This distinction has vast implications for the market in online advertising that spill over into the design of platforms, the business models of the companies that operate them, the experience of those who use them, and the relationship between those companies and those users.

Whereas legacy advertisers sought access to demographic groups, digital advertisers are able to place pitches for their products on screens of users who express interest in specific concepts. In addition to purchasing access to a certain age/gender combination, Web 2.0 allows advertisers to purchase access to words themselves. Lee constructs keywords, not demographically-coherent audiences, as the commodity that Google sells to its clients. As a corollary, she constructs search results as Google’s free lunch: rather than programming, Google organizes information and makes it available to curious users.

Strictly speaking, it is still the attention of a slice of audience that attracts advertisers. The shift from demographics to search terms represents a different proxy by which advertisers seek to identify the most promising leads for their products. As Siva Vaidhyanathan notes, “Google takes its money in small increments millions of times per day rather than by using the network TV model of taking millions of dollars a few times per day” [22].

In practice, social platforms — the Facebooks, Twitters, and Instagrams of the world — operate by packaging detailed demographic and consumer preference data. Moreover, the mechanics of the collection of these different types of user data undercut the role of the ratings companies that were so important to the political economies of legacy media. As one critic writes, “With the rise of commercial Internet platforms, audience ratings no longer need to be approximated, but permanent surveillance of user activities and user content allows the definition of precisely defined consumer groups with specific interests” [23]. Rather than sampling an audience, Web 2.0 companies are able to measure and report precisely what characteristics their audiences bear. Moreover, since “the appearance of a certain ad depends on the keyword that a user inputs” [24] or other features of the demographic and consumer preference profile that a platform has developed, a Web 2.0 platform can serve ads in a discriminating fashion: precision strikes rather than scattershot.

The fact that social media users spend, by his account, “billion[s of] hours” [25] feeding the Web 2.0 data collection apparatus leads Christian Fuchs to conclude that social media is built on the backs of users who go uncompensated for the work of creating content and generating salable data. Fuchs contests Jhally and Livant’s claim that the free lunch amounts to an audience wage: “Users are not paid on corporate social media [...], therefore they cannot generate money for buying food or other goods needed for survival” [26]. In short, since you cannot eat social media content, Fuchs argues that the labor Web users perform does not count as work and the free lunch does not count as pay.

Fuchs’s claims matter because they provide the context within which it becomes possible to make sense of Nielsen’s inequality curve, the measure of the proportion of users who do and do not participate on a platform. At stake are the parameters of the relationship between user and platform: what does each party expect of the other, what does it receive from the other, and how do platforms attempt to leverage the power inherent in their control of the flow of content and social relationships to modify user behaviors?

Fuchs’s claims have been contested in several different ways. First, Kit Hughes raises the fact that “non-monetary goals and needs [...] are met through [users’] labor” and stresses “the limits of a model of audience labor focused entirely on monetary exploitation” [27]. Hughes’s point is that users get something else from their participation, though the precise nature of that something else is “difficult to determine” [28]. The key is that the hows and whys of platform and user behavior may play out in economic, social, and cultural terms. As Hughes writes, “users may see themselves as operating primarily within an affective economy, while media corporations are interested in directly monetizeable behavior” [29]. The choice in responding to this objection is between assigning false consciousness to users or assuming that the economic explanation is necessary, but not sufficient, for understanding what people do with Web 2.0.

Another objection comes in the form of a question about the fundamental value of user-generated content. Brett Caraway, pushing back against the free labor thesis advanced by Tiziana Terranova early in the Internet era (Terranova, 2000), privileges the productive work of software engineers, interface designers, and other knowledge workers who build the platforms in the first place. The true value of Web 2.0 platforms is derived from knowledge work, not mindless status updates: “[U]ser-generated content is implicated in value creation only to the extent that by keeping down costs, it increases surplus value by making the imposition of more waged labor possible” [30].

hese two objections — over what social media users might gain besides monetary compensation and over the site of value in user-generated content — are part of broader cultural fissures in scholarly circles over what to do with Marxian thought as the moment of Capital’s composition recedes ever further into the past. These fissures were cleanly exposed in an exchange Fuchs engaged in with Adam Arvidsson and Elanor Colleoni. Fuchs, leaning on Marx’s labor theory of value, argues that user labor is the ultimate source of value for social platforms, and that the amount of value users generate is directly linked to the amount of time they spend on a platform. Acknowledging the contribution of tech-sector workers, Fuchs clearly ties platform profits to the exploitation of users: “The productive labor time that is exploited by capital, on the one hand, involves the labor time of the paid employees and, on the other hand, all of the time that is spent online by the users” [31]. Arvidsson and Colleoni counter that the labor theory of value no longer held in the moment of social media: “value is more related to the ability to create and reaffirm affective bonds” [32]. This is to say that value in social media is derived from the ability to provoke emotional responses in users. This position implicitly aligns Arvidsson and Colleoni with Caraway’s (2016) assertion that value is created by engineers and designers, because it is their labor that creates the possibility of provoking the sought-after affective responses. Fuchs (2012b) quickly retorted that Arvidsson and Colleoni’s analysis did not attend sufficiently to the exploitation of social media users. He also suggested that diminishing the role user labor plays in fueling platform profits amounts to turning a blind eye to widespread exploitation, not only of wealthy Western users, but also of the workers in developing countries who extract the materials needed to build the devices on which platforms are designed and run [33].

At their heart then, the questions raised by political economy’s encounter with participatory media are to do with locating the site of value production, and the question of lurking in Web 2.0 is to do with the relationship between platform health and non-participation. What role do users play in creating value for Web 2.0 platforms? Do platforms rely upon user labor to make ends meet? Do users receive just compensation in non-monetary form? Or is user-generated content just another distraction, another form of cultural production that diverts critical attention from the actual workings of the economic base? If users’ labor generates value for platforms, then lurking is a problem. On the contrary, if content generation is simply something users do to keep their hands from falling idle in between advertisements, then lurking matters far less. Lurking users still watch plenty extra, and the productive labor of knowledge workers processes that attention and converts it into revenues. Those who undertake the journey to solve political economy’s riddle must stop for lunch.



The blindspot of the blindspot

The free lunch is the blindspot debate’s blindspot. In the legacy media environment, there was little to discuss when surveying the pathways of exchange and tracing the course of money and power. The free lunch was simply “[t]he information, entertainment and ‘educational’ material transmitted to the audience” as an “inducement,” a “gift,” or a “bribe” [34]. Media companies needed to provide their audiences with a free lunch, so they would go out and buy it (from a production company or press agency) or they would cook it up themselves (in their own studios or newsrooms). In any case, the free lunch did not figure heavily in the traditional formulations of the political economy of communication: it was simply there. Indeed, one of Smythe’s motivations in writing the blindspot essay was to pull critical attention away from the ideological critique of content in favor of a focus on structural power arrangements.

But the structures of media have changed over the four decades since Smythe’s essay, and the nature of the free lunch has changed with them. Echoing and expanding on Lee’s assertion that search results were the free lunch Google offered its users, Marc Andrejevic argues that “the ‘free lunch’ of the digital era [...] is often not the content proper, but the organization of information” [35]. Just as media companies organize audiences into coherent wholes for advertisers, so, too, do they organize information into coherent wholes for users. Andrejevic dances around this position a bit in the lines that follow: “the content entices users,” “content ... continues to play an important role,” and most insightfully, platforms “entice[] [users] to create the very content that draws them to the site” ([36]. But his landing spot is clear: “The role [digital media companies] play is in helping to organize the information landscape, telling us where to find information, entertainment, products, and services, and allowing users to share information with one another” [37]. More concisely: “Organization, search, and retrieval are, in a sense, the new ‘content’” [38].

Though Andrejevic acknowledges the central role that users play in generating content, he sidelines this labor in at least two ways. First, Andrejevic clumps together search, e-commerce, and social sites. Even if it is true that all of these sites function by assembling an audience and selling it to advertisers, the data they collect from users along the way and the different uses to which different kinds of platforms put that data varies. If search’s primary attraction is the organization of content, then the data collected is related, as Lee explains, to users’ specific interests. The search box is, as Siva Vaidhyanathan has said, the place where users tell Google exactly what they want, and the results page is where Google meets that want alongside narrowly targeted advertisements (in Wilkinson and Vaidhyanathan, 2008). In the case of e-commerce sites, which are decorated with display and video advertisements, the free lunch is the possibility of shopping through the Internet. As with search, organization and access can be construed as the “free lunch,” but content matters: stores with unwanted products are quickly shuttered. Of course, in the e-commerce context, users can certainly spend money — and e-commerce platforms strive to provoke precisely that outcome — but the data collected during a user’s visit inform product placements and advertisements for alternative and additional products. In the case of Web 2.0 social platforms, however, the free lunch mutates a bit.

Andrejevic implies that the appeal of content pales in comparison to that of organization, that users are somehow just bedazzled by the admirable task of organization that Facebook has done with extant piles of information. This is akin to saying that people watched Hill Street Blues out of admiration for the balanced network schedule within which it sat. Not only does content still matter, but much of the effort that user experience teams expend is dedicated to coaxing users to generate the stickiest, sweetest concoctions that their peers can imagine. Without users composing tweets, maintaining Facebook Feeds, or pinning recipes, there would quite simply be nothing to organize in the social media space. Content — the “free lunch” — does not exist a priori, and social platforms depend on their ability to get users to generate large quantities of “good” content.



Making lunch

If broadcast television, terrestrial radio, and print periodicals (not to mention search and e-commerce sites) offer some kind of free lunch to their viewers, listeners, and readers, then social media platforms expect their users to make lunch for one another. As in a restaurant, the best lunch is the one that not only brings users in the door, but also entices them to come back often. While the “easy” forms of participation that Nielsen advocated can help in the immediate tasks of generating data points and keeping users looking, clicking, and tapping, it is the status updates, tweets, images, videos, sound files, and posts that constitute the content proper. This is the free lunch that brings users to social platforms and keeps them coming back. Yes, “networked sociality is the product” (Scholz, 2007), but it is also the inducement.

The “easy” forms of participation are necessary but insufficient: Liking, favoriting, starring, rating, friending, and following only go so far. These are highly structured data, quantified captures of affective expressions. Structured data fit readily into a database field. They exhibit rigidly-defined properties and yield to computation. We recognize structured data in forms like numbers, options on a pull-down list, or free-text entries that can be matched or added to lists of attributes. Age measured in years, genders chosen among limited options such as man or woman, and place names that can be connected to GIS resources: these are structured data.

The free lunch that users make for one another is deeply unstructured. Unstructured data do not yield readily to computation, but they do hold cultural value for humans. We recognize unstructured data in forms like sentences, ideas, images, speech, and songs. Computer science research has made great strides in mining and analyzing unstructured data, but producing it reliably still requires human labor. The content production function in social media is a matter of “harnessing human thought, precisely because it does not conform to machine logic” [39]. The role that social media platforms play is therefore that of algortihmized talent management: compelling and coaxing users to create the sort of content that will most effectively bring users in and convince them to come back often.

This is to say that the central question of content creation in Caraway’s objection — that the importance of user labor is oversold because it does not generate value, but at best serves to keep costs down — amounts largely to question of accounting. From users’ and platforms’ perspectives, user labor is exploited in order to boost media companies’ economic fortunes. Whether it contributes directly to the revenue side of the balance sheet or offsets other lines on the expense side is a distinction without a difference. Moreover, why must we choose between knowledge workers and users in identifying the source of value in social media? McDonald’s extracts value from its front-line workers, its kitchen staff, its supply-chain managers, its investment officers, and its customers, who advertise its products to their friends. Is it not possible — likely perhaps — that social media platforms extract surplus value from the labor of knowledge workers and users? Similarly, José van Dijck’s assertion that “the user’s role as data provider is infinitely more important than his role as content provider” [40] misses the point: both data and content matter to platforms.

The ultimate problem, then, is Nielsen’s pithy observation that “Most users don’t participate very much” (Nielsen, 2006). If platforms rely upon users to create content — to generate value directly, to hold down costs, or both — then the management task with which platforms must grapple amounts to the herding of so many cats. It is not only that non-money is a sucker, but also that it is hard to corral [41].



Anti-lurking ordnance

“Shape the inequality curve’s angle” is a polite way of saying, “make more users generate more content.” As Nielsen says, platforms are not going to break the curve, but they may be able to bend it. Can the 90-9-1 rule become an 80-15-5 rule? This is what platforms struggle to do in a battle of degrees.

In prosecuting this battle, platforms deploy a number of persuasive devices, some more insidious than others. ResearchGate, the social platform targeted at scientists, includes three factors in computing its proprietary ResearchGate Score: a “performance indicator calculated by an undisclosed algorithm [that] integrate[s] both bibliometrics and altmetrics by measuring researcher publications, questions asked and answered, and number of followers” [42]. Of course, for the inputs to the score to be counted, scholars must upload papers, ask and answer question on the platform, and amass followers inside the platform’s gates. As scholars begin pondering whether the ResearchGate Score ought to be included in tenure and promotion dossiers [43], the possibility of the Score gaining credibility in personnel decision-making processes brings the stakes into focus: participate or risk stalling professionally. In the context of ResearchGate, lurking amounts to a scholar pursuing research activities without working through the platform. The ResearchGate score is a quantitative measure of participation that drives qualitative content generation.

Much more familiar than the ResearchGate Score is the Facebook Like. Like is the quintessential example of “easy” participation. As Robert Gehl reports, marketing researchers found in the late 1980s and early ’90s that when consumers said they liked an advertisement, they were very likely to buy the promoted product — more likely, in fact, than any other advertisement trait tested in the study. The Like button thus acts as a tool that Facebook and its advertisers can use to quantify the affective impact that products have on potential customers. As Gehl put it, “You like, you buy,” and advertisers and content producers have made use of the Like in fine-tuning their hunt for customers or eyeballs, depending on their business model (Gehl, 2013).

The Like metric is crude, but it acts as an affect-laden supplement to the metrics that drive Web advertising: it’s at once the trace of a click and a more concentrated form of unique visitor. But it also acts as a form of anti-lurking ordnance. Likes are not only used to measure the effectiveness of product placement, and they do not simply accumulate in Facebook’s coffers. Instead, their totals are spread across the Web and integrated into each post on the Facebook news feed. This little number, flagged by a Facebook-blue thumbs-up icon, gives all users an easily-digestible indication of a post’s effectiveness. Like trains users to generate likeable content, and, because it acts as the quantifiable guarantor of effective Facebooking, it ensures that users care that the content they generate is likeable. In short, the omnipresence of Like counts ensures that users like Likeable things, that they like Liking them, and that they like it when others Like their stuff.

The presence of the number as a number impresses the measurability of impact and the quantification of quality. Other social platforms have their analogues, and in each case, the goal is to present users with an intellectual validation of participation. Indeed, in some implementations of Twitter’s Web interface, the numbers appear slightly heavier and darker than the Reply, Retweet, and Like buttons, as if to signal that the mere fact of a Tweet’s having been somehow touched by other users is more important than what they did with it. The metric signifies by its being a number rather than white space on the page, while the details — how big, how defined, how quickly — all fall into place a bit more slowly. The metric’s presentation in the interface encourages the internalization of the participatory logic of the free lunch.



Beyond the curve?

What, then, are we to make of lurking? It is a bit of human noise that disturbs the ever-expanding effort to rationalize the production and consumption of cultural products. It’s the remainder of human activity that fails to conform — deliberately or otherwise — to the capitalist logics that drive Web 2.0. If participation matters, then lurking does, too. Social media is a technology that transforms human attention, user affect, and networked sociality into resources. Lurking is a practice that preserves humanity: it is a bit of analog life that networked capital cannot digest.

Social media research can further nuance its efforts to decipher contemporary culture’s common space by attending to lurking with the same focus that it has devoted to participation. For example, scholars would do well to investigate the differentials of lurking across classes of people and along markers of social difference. Laurie Oullette and Julie Wilson proposed that online participation is heavily gendered, and they suggested that “women’s interactivity can [be] mobilized as a gendered requirement of neoliberal citizenship” [44]. The implication of their claims is that men’s participation asserts a right to contribute an idea and establish authority, while women’s participation folds into their social status as caretakers. Participation becomes a privilege for some users and an obligation for others, even if it all is absorbed by social platforms. Lurking, too, has a social geography. It is the privilege of those who have access to sufficient financial and social capital and do not rely upon the accumulated chits they can amass through online participation. It is a comfortable option for those who can muster enough social advantage through off-platform activity that they can hang in the margins of LinkedIn or Instagram. By the same token, lurking is the sentence meted out to those who are sidelined by ability, circumstance, or biology. These privileges and punishments can at times accrue according to personality and disposition, and at times according to identity, social status, and familiar categories like gender, race, and class, whose various inequalities new media have never really begun to eradicate. Plotting this geography is fertile ground for future research in new media studies.

Complicating this agenda is the fact that social platforms (which are the places where lurking can occur) are moving targets. Companies regularly tweak their platforms — though Ian Bogost cautions that the word tweak cloaks the fact that algorithms are mere business equipment, not “decrees from an omnipotent software-god” (Bogost, 2016). More fundamentally, there is emerging evidence that some platforms may be backing away from user generated content, or at least taking steps to minimize its role in their business models: “as a social network per se — a place where people go to connect with friends and acquaintances — Facebook may be just beginning to wane” (Oremus, 2016). Instead of relying on users to generate content, Facebook has adjusted its platform to emphasize polished, professionally produced video, audio, and text content. These adjustments represent a fundamental pivot from a home for networked sociality to “a personalized portal to the online world” (Oremus, 2016). At the end of the day, it’s still the eyeballs that matter to Facebook. In this case, participation becomes ornamental, and the forms of easy participation that today serve as gateways to increased participation may sink into platforms’ ever more sophisticated boxes of content-targeting tools. This shift, reminiscent of the portal wars of the late 1990s, hints at an endgame to the battle against lurking: if the participation inequality curve only bends so far, perhaps platforms will simply toss it aside and seek a direct line to users’ time, emotions, and attention. End of article


About the author

Scott Kushner is Assistant Professor of Communication Studies at the University of Rhode Island.
E-mail: scottkushner [at] uri [dot] edu



Ian Reyes and Justin Wyatt provided thoughtful comments on an early version of this essay. Gracious audiences at the University of Rhode Island, the 2015 Annual Conference of the Society for Cinema and Media Studies, and the Differential Mobilities conference at Concordia University offered helpful questions that improved parts of the argument. Two anonymous and generous reviewers provided critiques and suggestions.



1. Shirky, 2008, p. 65.

2. Jenkins, 2006, p. 3.

3. Jenkins, 2014, p. 272.

4. Andrejevic, 2002, p. 232.

5. Andrejevic, 2011, p. 285.

6. Whittaker, et al., 1998, p. 262.

7. Whittaker, et al., 1998, pp. 260, 262.

8. Cf., Bird, 2011, p. 504.

9. Preece, et al., 2004, p. 203.

10. Mosco, 2009, p. 2.

11. Smythe, 1977, p. 1.

12. Smythe, 1977, p. 5.

13. Liebling, 2009, p. 682.

14. Williams, 2003, pp. 101–120.

15. Livant, 1979, p. 96.

16. Livant, 1982, p. 214.

17. Livant, 1979, p. 96.

18. Jhally and Livant, 1986, pp. 136, 126.

19. Meehan, 1984, p. 221.

20. Meehan, 1984, p. 222.

21. Lee, 2011, p. 439.

22. Vaidhyanathan, 2011, pp. 27–28.

23. Fuchs, 2012a, p. 702.

24. Lee, 2011, p. 439.

25. Fuchs, 2012a, p. 716.

26. Fuchs, 2012a, p. 714.

27. Hughes, 2014, p. 650.

28. Ibid.

29. Ibid.

30. Caraway, 2016, p. 77.

31. Fuchs, 2010, p. 191.

32. Arvidsson and Colleoni, 2012, p. 136.

33. Fuchs, 2012b, p. 638.

34. Smythe, 1977, p. 5.

35. Andrejevic, 2014, p. 199.

36. Ibid.

37. Ibid.

38. Andrejevic, 2014, p. 200.

39. Kushner, 2013, p. 1,244.

40. Dijck, 2009, p. 49.

41. Cf., Benkler, quoted in Fox, 2007, p. 51.

42. Yu, et al., 2016, p. 1,002.

43. Ovadia, 2014, p. 167.

44. Oullette and Wilson, 2011, p. 549.



Mark Andrejevic, 2014. “‘Free lunch’ in the digital era: Organization is the new content,” In: Lee McGuigan and Vincent Manzerolle (editors). The audience commodity in a digital age: Revisiting a critical theory of commercial media. New York: Peter Lang, pp. 193–206.

Mark Andrejevic, 2011. “Surveillance and alienation in the online economy,” Surveillance & Society, volume 8, number 3, pp. 278–287, and at, accessed 31 May 2016.

Mark Andrejevic, 2002. “The work of being watched: Interactive media and the exploitation of self-disclosure,” Critical Studies in Mass Communication, volume 19, number 2, pp. 230–248.
doi:, accessed 31 May 2016.

Adam Arvidsson and Elanor Colleoni, 2012. “Value in informational capitalism and on the Internet,” Information Society, volume 28, number 3, pp. 135–150.
doi:, accessed 31 May 2016.

David A. Banks, 2015. “Lines of power: Availability to networks as a social phenomenon,” First Monday, volume 20, number 11, at, accessed 13 May 2016.
doi:, accessed 31 May 2016.

S. Elizabeth Bird, 2011. “Are we all produsers now? Convergence and media audience practices,” Cultural Studies, volume 25, numbers 4–5, pp. 502–516.
doi:, accessed 31 May 2016.

Ian Bogost, 2016. “Go tweak yourself, Facebook,” Atlantic (28 April), at, accessed 13 May 2016.

Jed R. Brubaker, Mike Annany, and Kate Crawford, 2016. “Departing glances: A sociotechnical account of ‘leaving’ Grindr,” New Media & Society, volume 18, number 3, pp. 373–390.
doi:, accessed 31 May 2016.

Brett Caraway, 2016. “Crisis of command: Theorizing value in new media,” Communication Theory, volume 26, number 1, pp. 64–81.
doi:, accessed 31 May 2016.

Nathalie Casemajor, Stéphane Couture, Mauricio Delfin, Matthew Goerzen, and Alessandro Delfanti, 2015. “Non-participation in digital media: Toward a framework of mediated political action,” Media, Culture & Society, volume 37, number 6, pp. 850–866.
doi:, accessed 31 May 2016.

Nick Couldry, 2011. “More sociology, more culture, more politics: Or, a modest proposal for ‘convergence’ studies,” Cultural Studies, volume 25, numbers 4–5, pp. 487–501.
doi:, accessed 31 May 2016.

Kate Crawford, 2011. “Listening, not lurking: The neglected form of participation,” In: Hajo Grief, Larissa Hjorth, Amparo Lasén, and Claire Lobet-Maris (editors). Cultures of participation: Media practices, politics and literacy. Berlin: Peter Lang, pp. 63–74.

Kate Crawford, 2009. “Following you: Disciplines of listening in social media,” Continuum, volume 23, number 4, pp. 525–535.
doi:, accessed 31 May 2016.

José van Dijck, 2009. “Users like you? Theorizing agency in user-generated content,” Media, Culture & Society, volume 31, number 1, pp. 41–58.
doi:, accessed 31 May 2016.

Justin Fox, 2007. “Getting rich off those who work for free,” Time, volume 169, number 9 (15 February), p. 51, and at,9171,1590440,00.html, accessed 31 May 2016.

Christian Fuchs, 2012a. “Dallas Smythe today — The audience commodity, the digital labour debate, Marxist political economy and critical theory. A prolegomena to a digital labour theory of value,” Triple-C, volume 10, number 2, pp. 692–740, and at, accessed 31 May 2016.

Christian Fuchs, 2012b. “With or without Marx? With or without capitalism? A rejoinder to Adam Arvidsson and Eleanor Colleoni,” Triple-C, volume 10, number 2, pp. 633–645, and at, accessed 31 May 2016.

Christian Fuchs, 2010. “Labor in informational capitalism and on the Internet,” Information Society, volume 26, number 3, pp. 179–196.
doi:, accessed 31 May 2016.

Robert W. Gehl, 2013. “A history of like,” New Inquiry (27 March) at, accessed 29 January 2016.

Robert W. Gehl, 2011. “The archive and the processor: The Internal logic of Web 2.0,” New Media & Society, volume 13, number 8, pp. 1,228–1,244.
doi:, accessed 31 May 2016.

Holly Goodier, 2012. “BBC online briefing Spring 2012: The participation choice,” BBC Internet Blog (4 May), at, accessed 29 January 2016.

Kit Hughes, 2014. “‘Work/place’ media: Locating laboring audiences,” Media, Culture & Society, volume 36, number 5, pp. 644–660.
doi:, accessed 31 May 2016.

Henry Jenkins, 2014. “Rethinking ‘rethinking convergence/culture’,” Cultural Studies, volume 28, number 2, pp. 267–297.
doi:, accessed 31 May 2016.

Henry Jenkins, 2006. Convergence culture: Where old and new media collide. New York: New York University Press.

Henry Jenkins, 1992. Textual poachers: Television fans & participatory culture. New York: Routledge.

Sut Jhally and Bill Livant, 1986. “Watching as working: The valorization of audience consciousness,” Journal of Communication, volume 36, number 3, pp. 124–143.
doi:, accessed 31 May 2016.

Bobbie Johnson, 2012. “Is the 1% rule dead? The BBC thinks so, but it’s wrong,” Gigaom (6 May) at, accessed 29 January 2016.

Scott Kushner, 2013. “The freelance translation machine: Algorithmic culture and the invisible industry,” New Media & Society, volume 15, number 8, pp. 1,241–1,258.
doi:, accessed 31 May 2016.

Scott Kushner 2011. “Virtually dead: Blogospheric absence and the ethics of networked reading,” Communication Review, volume 14, number 1, pp. 24–45.
doi:, accessed 31 May 2016.

Micky Lee, 2011. “Google ads and the blindspot debate,” Media, Culture & Society, volume 33, number 3, pp. 433–447.
doi:, accessed 31 May 2016.

Karen E.C. Levy, 2015. “The user as network,” First Monday, volume 20, number 11, at, accessed 13 May 2016.
doi:, accessed 31 May 2016.

A.J. Liebling, 2009. The sweet science and other writings. New York: Library of America.

Bill Livant, 1982. “Working at watching: A reply to Sut Jhally,” Canadian Journal of Political and Social Theory, volume 6, numbers 1–2, pp. 211–215, and at, accessed 31 May 2016.

Bill Livant, 1979. “The audience commodity: On the ‘blindspot’ debate,” Canadian Journal of Political and Social Theory, volume 3, number 1, pp. 91–106, and at, accessed 31 May 2016.

Eileen Meehan, 1984. “Ratings and the institutional approach: A third answer to the commodity question,” Critical Studies in Mass Communication, volume 1, number 4, pp. 216–225.
doi:, accessed 31 May 2016.

Vincent Mosco, 2009. The political economy of communication. Second edition. Los Angeles: Sage.

Graham Murdock, 2014. “Introduction to the blindspot revisited,” In: Lee McGuigan and Vincent Manzerolle (editors). The audience commodity in a digital age. New York: Peter Lang, pp. 55–57.

Jakob Nielsen, 2006. “The 90-9-1 rule for participation inequality in social media and online communities” (9 October), at, accessed 29 January 2016.

Blair Nonnecke and Jenny Preece, 2003. “Silent participants: Getting to know lurkers better,” In: Christopher Lueg and Danyel Fisher (editors). From Usenet to CoWebs: Interacting with social information spaces. London: Springer, pp. 110–132.
doi:, accessed 31 May 2016.

Tim O’Reilly, 2005. “What is Web 2.0: Design patterns and business models for the next generation of software” (30 September), at, accessed 31 May 2016.

Tim O’Reilly and John Battelle, 2009. “Web squared: Web 2.0 five years on,” at, accessed 29 January 2016.

Will Oremus, 2016. “Facebook isn’t the social network anymore,” Slate (24 April), at, accessed on 13 May 2016.

Laurie Oullette and Julie Wilson, 2011. “Women’s work: Affective labour and convergence culture,” Cultural Studies, volume 25, numbers 4–5, pp. 548–565.
doi:, accessed 31 May 2016.

Steven Ovadia, 2014. “ResearchGate and Academic social networks,” Behavioral & Social Sciences Librarian, volume 33, number 3, pp. 165–169.
doi:, accessed 31 May 2016.

Neil Perkin, 2012. “Is the 1, 9, 90 rule outdated?” Only Dead Fish (11 May), at, accessed 29 January 2016.

Laura Portwood-Stacer, 2013. “Media refusal and conspicuous non-consumption: The performative and political dimensions of Facebook abstention,” New Media & Society, volume 15, number 7, pp. 1,041–1,057.
doi:, accessed 31 May 2016.

Laura Portwood-Stacer, 2012. “Anti-consumption as tactical resistance: Anarchists, subculture, and activist strategy,” Journal of Consumer Culture, volume 12, number 1, pp. 87–105.
doi:, accessed 31 May 2016.

Jenny Preece, Blair Nonnecke, and Dorine Andrews, 2004. “The top five reasons for lurking: Improving community experiences for everyone,” Computers in Human Behavior, volume 20, number 2, pp. 201–223.
doi:, accessed 31 May 2016.

Eric Schmidt, Vittorio Colao, Sheryl Sandberg, Jim Hagemann Snabe, and Satya Nadella, 2015. “The future of the digital economy,” at, accessed 29 January 2016.

Trebor Scholz, 2007. “What the MySpace generation should know about working for free,” Re-Public, at, accessed 29 January 2016.

Clay Shirky, 2008. Here comes everybody: The power of organizing without organizations. New York: Penguin Press.

Dallas Smythe, 1977. “Communications: Blindspot of Western Marxism,” Canadian Journal of Political and Social Theory, volume 1, number 3, pp. 1–27, and at, accessed 31 May 2016.

Tiziana Terranova, 2000. “Free labor: Producing culture for the digital economy,” Social Text, volume 18, number 2, pp. 33–58.
doi:, accessed 31 May 2016.

Claes Thorén and Andreas Kitzmann, 2015. “Replicants, imposters, and the real deal: Issues of non-use and technology resistance in vintage and software instruments,” First Monday, volume 20, number 11, at, accessed 13 May 2016.
doi:, accessed 31 May 2016.

Siva Vaidhyanathan, 2011. The Googlization of everything (and why we should worry). Berkeley: University of California Press.

Steve Whittaker, Loren Terveen, Will Hill, and Lynne Cherny, 1998. “The dynamics of mass interaction,” CSCW ’98; Proceedings of the 1998 ACM Conference on Computer Supported Cooperative Work, pp. 257–264.
doi:, accessed 31 May 2016.

Will Wilkinson and Siva Vaidhyanathan, 2008. “Will Wilkinson and Siva Vaidhyanathan” (24 August), at, accessed 29 January 2016.

Raymond Williams, 2003. Television: Technology and cultural form. London: Routledge.

Min-Chen Yu, Yen-Chun Jim Wu, Wadee Alhalabi, Hao-Yun Kao, and Wen-Hsiung Wu, 2016. “ResearchGate: An effective altmetric indicator for active researchers?” Computers in Human Behavior, volume 55, part B, pp. 1,001–1,006.
doi:, accessed 31 May 2016.


Editorial history

Received 27 May 2016; accepted 28 May 2016.

Creative Commons License
This paper is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Read only: The persistence of lurking in Web 2.0
by Scott Kushner.
First Monday, Volume 21, Number 6 - 6 June 2016

A Great Cities Initiative of the University of Illinois at Chicago University Library.

© First Monday, 1995-2019. ISSN 1396-0466.