the point is to investigate what has happened since people began to talk about the Internet, to look at the cultural and social factors, not as independent from but as in certain ways constituent of, all that goes with the emergence of “the Internet” on the world stage.

Cultural expectations and technological artefacts exist in an inextricably intertwined way. Technologies are perhaps best understood, not so much as agents in their own right, but as thought-objects for the collective enactment and exploration of hopes, desires and political visions.

the culture's enthusiastic embrace of a romantic vision of the Internet as an agent of change is what generated a flood of engagement and investment in the technology, and then shaped its character. The structure of feeling thus had a causal effect on subsequent developments.

the collapse of the Soviet Union and the return of the stock market to pre-1987 levels in the early 1990s, for example, gave new energy to individualist, pro-market, anti-government sentiments.

The argument, then, is that the conviction that Internet technology was inherently democratising, unpredictable and a vehicle for unique expression was not simply the triumph of a world view or a vision, but grew out of a set of specific historical accidents, a conjuncture of institutional blindspots, cultural traditions and political pressures, all pivoting around particular interplays of thought and feeling.

Before 1992, the Internet was a colloquial term for an experiment, and by 1996, it had become the global network of networks. Between 1992 and 1996, the meaning of “Internet” was transformed from being a quiet testbed known mostly to experts, to a global institution whose name seemed to be on everyone's lips and whose existence and importance was taken for granted

The danger here, as Williams pointed out, is that this basically base/superstructure-driven view of events rests on an idealist abstraction of both culture and the economy; in material life, one never occurs without the other (1978 Links to an external site.). These approaches leave difficult questions of causality and function to the level of narrative description, and offer little guidance on how to explain why some themes emerge at a given time and others do not, and how particular discourses function in (rather than merely reflect) society.

This essay will sketch a framework intended to provide more precision to understanding the specific relations between culture and the Internet, and argue that Raymond Williams’ old but still useful notion of a “structure of feeling” not only helps account for actions like Magaziner's, but also helps explain what we mean when we talk about the Internet itself. The goal, given the limited space, is to be illustrative rather than comprehensive.

This was a neoliberal moment: privatisation of the Internet seemed so obvious that it need not be even discussed. But it was also a cultural moment, reflecting style and attitude as much as ideology or policy positions. Both Magaziner and his audience knew that as a government bureaucrat, he was acting against type. We do not expect government officials to so willingly “get out of the room,” and generally when an official does not keep an eye on things, it is interpreted as an abdication of responsibility, not a heroic move.

In that case, businesses like AOL seem to be facing a double bind. They need the kind of dynamism that spontaneous cultural production and organization engender, yet they must avoid the alienating control structures that often have to be established to operate multibillion-dollar media conglomerates.

Community making, as an extension of family maintenance, falls under the influence of the same type of rhetoric that ‘‘pastoralized’’ women’s housework. American society continues to see volunteer work of the kind that generates and maintains communities (both on and offline) as market inalienable, as a noble and altruistic pursuit, even as companies like AOL commodify community.

In a sense, the case of AOL community leaders is a classic study of the process by which an occupation is born from unpaid work. At an early level of development, an occupation lacks the institutional and social recognition that helps the early ‘‘occupational pioneers’’ convince society that they are worthy of compensation.

The ephemeral nature of today’s jobs can hide labor in the context of leisure. In the case of the Internet, this labor is always in plain site (we see the wealth of information on the Web) yet those who do the work of generating and maintaining the Web remain hidden away under the rhetoric of volunteerism or hobby. Terranova tries to get at that ‘‘hiddenness’’, and explains it as a complex relationship between cultural production, or the social factory, and the technologies and methodologies of postindustrialism.

Taken collectively, the work of volunteers, content makers, website posters, and all others who add content to the World Wide Web constitute a ‘‘network of immaterial labor’’, comprised of a collective intelligence that is the self-organized, principle productive force of the digital economy. ‘‘Capital’s problem’’, Terranova states, ‘‘is how to extract as much value as possible’’,37 from this collective.

While there is little doubt that many did volunteer for the altruistic rewards, many came to AOL with other expectations. Some thought that volunteering would be a springboard to employment in a lucrative Internet company; others wanted to gain experience with computers; and still others simply wanted a price break on the hourly rates that were driving their service bill beyond their budgets. As their expectations failed them, the reorganization process positioned some of the volunteers to begin reassessing the meaning of volunteering and community at AOL. Whether as a means of revenge or as a means of empowering themselves against organizational and institutional forces that took from them work they valued, these few ex-volunteers chose to reconstitute themselves as employees and began viewing community as a commodity.

To this end, some volunteers were consistently promised paying jobs that were continuously moved out of reach.28 One particular guide, for example, was promised a job as a RAINMAN programmer and waited from January of 1998 until April of that year to get his first check.29 When it eventually came, it was for a lower pay-scale and only covered part of the time he had worked; the remainder of the time he had worked was still considered volunteered time.

The incident concerning the ‘‘Names and Initials’’ folder is of particular interest because it brought out much of the frustration the community leaders were feeling with management. Among other things, this incident was a disruption to the sense of community within the volunteer group and was emblematic of the process by which some community leaders began to see themselves as employees. One particular post put it in these terms: ‘‘Frankly I admit that there are times where I feel like we volunteers are now work in cubicles, where before it was an auditorium.’’25 Yet this was not the only outcome of the ‘‘Names and Initials’’ incident. What also surfaced was the declaration of a general fear among some community leaders that speaking out against changes would have repercussions. During the exchange over the folder, some of the community leaders confided that they normally did not talk about their feelings of demoralization as a result of the structural changes because of fears that they would be released, or because they generally found management inattentive to their concerns.26 The rift that was present at this juncture in the history of the CLO contrasts dramatically with the stories recounted by some community leaders of the early days of AOL.

Since, at that time, the volunteers were not clearly organized under any one internal division, AOL now realized that it had little control over how volunteers were representing AOL in the various public forums. As one manager put it, ‘‘There was this sudden light-bulb moment where they [management] said, ‘Oh, my God, we have thousands of people out there acting as our representative, and we don’t even know who they are’.’’

Still others logged on and became volunteers because they believed that AOL would provide them with the needed computer experience to be employable in the emerging tech-economy or even by AOL. Kelly Hallisey, a volunteer since the early 1990s, recalls explaining to her husband why she stayed online for such long hours, ‘‘We were having major arguments over it [staying on line] and I said, ‘You know [:::] this is the way things are going to go, I can see this turning into a really good paying job.’’

By the time AOL started to become an important player on the Internet, the hype surrounding computers and their business potential began to surface. Thus, many volunteers became involved with the Internet to acquire the increasingly valuable computer capital that they hoped would propel them to better lives.

Volunteers that maintain communities on the Internet have been around since the Internet’s early years; however, Netizen’s giving of their time and energy has its true roots in the hacker history that was an essential component of the formation of the Internet. The idea of freely giving up one’s time and knowledge is rooted in the academic, collaborative efforts that shaped the Internet as a project for the United States Defense Department’s Advanced Research Projects Agency (ARPA).

When it entered the Internet-service-provider business, AOL had approximately 75,000 users that it had attracted through its partnership with Apple, Tandy, and Commodore. By the early 1990s volunteers were an integral part of AOL’s community and did much work to establish content and help new members. In the early days of AOL, the company never specifically set out to create a volunteer organization but it welcomed the fact that the communities could maintain themselves through the work of volunteers.

Postindustrialism and the Internet have come together to draw value from cultural labor produced on the Internet; the case of AOL will show that management practices seeking to control the work process have helped define volunteers as workers.

The increase in member subscription was expected to be significant, and a wave of concern swept through the large remote-staff volunteer population, whose duties included monitoring electronic bulletin boards, hosting chat-rooms, enforcing the Terms of Service agreement (TOS), guiding AOL users through the online community, and even creating content using the AOL’s own program, RAINMAN (Remote Automated Information Manager), the text scripting language and the publishing tool that allows remote staffers to update and change content on AOL.

As other grounded analyses of the US technology sector have noted, employees in this industry adopt (and perpetuate) a “neoliberal subjectivity” (Marwick, 2013 Links to an external site.), wherein individual status-seeking and the drive to constantly iterate on one’s career is the norm. As a point of contrast, Tom’s case suggests that diagnosis and lesson-making, as the primary way that other employees related to technological death, are intrinsic to Silicon Valley’s neoliberal subject position.

Notably, no interviewee referred to the moment when servers shut off in 2018 as the platform’s passing–indeed, they rarely referred to it at all. By choosing these other moments to discuss in relation to Friendster’s demise, these interviews reveal the way that ‘platform death’ is not a bounded or uncontested event signaled by a final technical closure. Rather, death here is constructed heterogeneously, in that employees identified varying moments that signified Friendster’s end. To put it another way, this article does not reveal a final pronouncement about ‘what really killed’ Friendster through the voices of these employees, but instead reveals the social negotiation of technological death.

As these interviews attest, there is no one answer as to why Friendster died. For some employees, Friendster died because it could not solve its technical issues. For others, it died when they realized the platform would not have US market dominance. For many, its death was brought about through impasses in the product’s direction. For the remainder, Friendster had not necessarily died, because it lives on in platforms that persist today.

Yet many employees understood Friendster not as a failure, or even, to an extent, to be dead. Friendster, many noted, had been transmuted either into an instructive fable well known in Silicon Valley, or into the literal infrastructure of other existing platforms, through either feature sets or intellectual property. The notion that Friendster paved the way or laid the foundations for networks that have market dominance today was common. As former CEO Scott would say: We were like the Wright brothers, but never took off. We didn’t have that 36-second flight. Myspace was like the 36-second flight. Facebook is the flight that kept going.

While a number of former employees identified misguided visions as the reason for Friendster’s death, there was little consensus over which vision was the misguided one. Rather, many related these mistakes in product direction to a larger issue of not having a clear vision for the company in general, instead allowing it to list from one emphasis to the next. In turn, some, like Mel and Travis, spoke of the validity in maintaining the founder’s original vision for the platform throughout the company’s lifetime.

‘Wait–these Philippine users are our users. There’s a lot of them. Let’s figure out how we can serve them better.’ But I think it took a while for them to make the transition from having fallen from the top of the US to being like, ‘Okay, well maybe there are other things we can do.’

As employees noted, this user community did not align with the long-term vision that was held for Friendster, particularly by the board and other company leadership. Instead, the users in Southeast Asia were thoughts of largely as a “problem,” a “nuisance,” or an “annoyance,” and ultimately by some, understood to be Friendster’s cause of death.

Participants were recruited via professional online platform LinkedIn.

The language that platforms use to describe themselves, for instance, shapes platforms’ political standing and regulation (Gillespie, 2010 Links to an external site.), while the language that tech employees use to describe their conditions in turn influences social practices within those cultural spaces (English-Lueck, 2017 Links to an external site.; Neff, 2012 Links to an external site.). In other words, choosing “death” to describe these closures is itself descriptive of the cultural and economic landscape of the platform industry, and deserves analysis.

Even if a technology is no longer used for its designed purpose, these terms show how it continues to exist within a series of social and technical relations, how it continues to interact with labor, politics, identity, and materiality.

This article thus makes two central contributions: it argues that death, when viewed as a discourse, does not signal an analytical dead end. Rather, it acts as a useful heuristic for examining cultural theories of technological change in a mediascape saturated by platforms, and for critically parsing the often obfuscatory rhetoric of platform companies. Moreover, it shows the ways that death is fundamental to Silicon Valley mythmaking and ideology, both on an industry-wide level, as the limited academic literature on failure and Silicon Valley has noted (e.g. Draper, 2017 Links to an external site.), but also on an individual scale, as this case attests.

Given the platform’s unclear “time of death,” Friendster offers a case for understanding the death of a platform beyond the final technical termination brought about by powering off the site’s servers on its final day. As this article asserts, understanding the conceptual scope of platform death beyond this technical end is an important but understudied topic when it comes to recognizing the social power of platforms, and the cultural context that they operate in, today.

Much like Facebook today, Friendster users were encouraged to use their real names and images on their personal profiles, and could send messages, post photos, and eventually see others’ activity within the network via a centralized newsfeed. While built around users in the United States, Friendster had also become popular in Southeast Asia, including the Philippines, Malaysia, Indonesia, and Singapore, where its user base rapidly exceeded that in the United States within the site’s earliest years.

progenitor