What was Web 2.0? Versions and the politics of Internet history

This presentation, given at the Oxford Internet Institute May 4 2011, is a reduced version of a paper of the same title that includes substantially more examples and all appropriate references. Please refer to this paper for a full account.

What was Web 2.0? [ full paper]


In 2008, the journal Fibreculture published an issue entitled “After Convergence” exploring questions of human, social and technological connectivity within a world where computer networks had led to the convergence of formerly disparate cultural practices. In describing the several contributions to the issue, the editors wrote:

we have asked not only what makes ‘2.0’ distinct from ‘what came before’ but also how it will be understood in the future. We ask this question not least because we are somewhat alarmed by visions of proliferating version control as 2.0 merges with 3.0 and 4.0 looms on the horizon (Bassett et al.)

My paper, in general terms, takes its lead from this critical interest in things ‘2.0’, focusing specifically upon Web 2.0. I will outline the way in which the emergence of Web 2.0 brought to the web the discourse of versions. A history created in versions, a particular form of an object’s history, required that, as well as Web 2.0, there had to be as Web 1.0 and of course presupposed the emergence of Web 3.0. The articulation of one depended, explicitly or implicitly, on the others.

Is Web 2.0 dead?

Web 3.0 labels many trends in the development of the web that might presage a ‘new’ time involving such ideas as the Semantic Web, both systemically and in specific application; new investment opportunities; and even new scholarly critiques and theories. So, perhaps it is time to ask: “Le Web 2.0 est-il mort?” (Lequien). Far and wide across the web, the phrase Web 3.0 yields a vast array of returns from search engines whether they reference marketing slogans, political commentary, technical discussions or techno-evangelist opinion.

Yet the discourse of Web 3.0 bears an uncanny resemblance to the rise of Web 2.0: different in time, not substance, and marked by the same jumble of competing, but inherently irreconcilable, differences of perspective and purpose as people position themselves, their technologies, and their ideals in relation to what has come before and what might come in future. In such circumstances, the real questions to ask, then, are:how did the web come to have versions in the first place; what is the discursive process by which these versions come to make sense; and what is revealed by analyzing this history of versions?

Web 2.0 and Web 1.0 – continuity or change

Around the start of 2006, Web 2.0 became the principal way to describe the then-current web rather than being a term which looked towards an as-yet unreached future. Yet several businesses and web services thought exemplary of, or essential to, Web 2.0 date from much earlier times, as do the technologies on which they rely. Examples include blogging (Blogger and Livejournal), distribute payment services (Paypal), crowd-sourced, user-generated content (Wikipedia), social networking (Sixdegrees), and algothrimic search and associated marketing (Google).

Equally, behaviours and sensibilities which have for several years regularly been discussed in terms of Web 2.0 pre-date its origin and extend back beyond the web itself. It has been claimed that:

… the essential difference between Web 1.0 and Web 2.0 is that content creators were few in Web 1.0 with the vast majority of users simply acting as consumers of content, while any participant can be a content creator in Web 2.0 and numerous technological aids have been created to maximize the potential for content creation. (Cormode and Krishnamurthy).

The visibility of such behaviours in earlier Internet times suggest it would be less clear that Web 2.0 created this change but merely promoted it. Examples from before the web include USENET, bulletin boards, email lists, chat environments, and MUDs, all demonstrated the founding basis of socio-technological networking: people wished to share information, create content and work with others in doing so. The World Wide Web did not mean an end to these earlier forms, but enabled a rapid increase in their utility and visibility; there was also significant website creation and curation by individual users and the concept of the ‘webring’ emerged to enable networks of users and authors to emerge.

These examples stand in contradiction to the current ‘history’ of Web 2.0, which generally views the technologies, businesses and social formations of the past several years as initiating and making possible an online world of participatory, user-generated and open content and communication. Of course, the point is not that this current history is wrong: histories are not wrong, per se, but contingent on the circumstances and purposes of their creation and circulation and current contestations are fought out through, rather than by reference to, historical accounts.

Juxtaposing examples which suppose a continuity of development online, with a dominant history of Web 2.0 as radical transformation allows us to look more deeply at the cultural complexity of the notion of ‘version’. The idea of versions is becoming a cultural commonplace of today’s world because of the rise of computing: the conventions of software development are coming to play a critical role in the consumption of goods and services dependent on digital technology. While it is not new for goods and services to be promoted and sold on the basis that they replace what came before (for example, a new model of car or washing machine), it is definitely the case that, in computing culture, consumers are now receptive to the idea of purchasing something which they know will be replaced in a short period of time by a new version and willingly enter into this transaction, becoming part of the development cycle as much as being its recipients. Consumers accept, if reluctantly, that the digital products they are not entirely finished, and will be regularly ‘patched’.

These points are not merely a comment on current software marketing: they reveal the semiotic work that the discourse of versions can do. Versions allow products to claim to be new, but not threateningly so, because they also sustain continuity and promise an easy transition from that which came before. At the same time, versions allow new products to derogate earlier incarnations of themselves as limited or otherwise demanding of replacement. Through this duality of appeal, a discourse of versions both reassures consumers that they were right to buy the product first time around, but must now, of course, consume again and not rely on what they had already bought. Ultimately, the fetish object for consumer gratification becomes the process of upgrading, rather than what is possessed after that act.

Perhaps, then, technology might legitimize the move to a new, second version of the web while also sustaining continuity? Code within Web 2.0 is more sophisticated and enables developers and users to do far more; and in many different ways. Yet many Web 2.0 sites do not demonstrate technological sophistication, nor rely on innovations in code. Further the importance for the computing industry of shifting the way we think of the Internet from channel (media discourse) to platform (computing discourse) cannot be understated as a rationale, and implies technological change is consequential to a more fundamental commercial re-orientation. Thus, as Berry has argued, the insistence on technology as a discriminator between Web 2.0 and things other, is more of a demand for what ought to be, rather than an objective description of actual change. Further if technology is to authorize the legitimacy of claims for a transition to Web 2.0, then necessarily technology is presumed to determine or at least substantially control the consequences and meanings of that change.

Not all explanations or discussions of the transition to Web 2.0 relied on appeals to technology, however. We can contrast contemporary commentators Schauer and Hinchcliffe who argued respectively that new behaviours emerged because of technological development, or that the traditional behaviours became influential because of the large number of regular Internet users which the web created. In both cases however, these authors but demonstrate a fundamental tension within the language of Web 2.0: the need to manage the transition between versions, to explain the 2.0 which both ‘breaks’ with the past, and also connects to it. This tension is not easily resolvable because, in truth, the tension is what gives ‘versions’ their semiotic cogency.

Yet, without an articulation of change and continuity between ‘then’ and ‘now’ there could be no rationale for Web 2.0 and thus, even as this form of the web was proposed as novel, it had to be presented as less than novel. Throughout the texts of Web 2.0, simple dichotomies of new and old are presented hand-in-glove with the assertion of a contradictory, more developmental path from earlier times, often within a few short sentences of such assertions. While popular advice might be that “The definition of Web 1.0 completely depends upon the definition of Web 2.0.” (Strickland), in fact the existence of Web 2.0 depends utterly on Web 1.0. Without it, the absences and failures which Web 2.0 solves would not be knowable.

Web 2.0 and Web 0.0 – realignment to ideals

There are other ways in which the creation of Web 2.0 came to define the particular sense we have of the history of the web with strong claims that Web 2.0 returned to the origins of the web, and indeed the Internet more generally, realigning everyday technology and social practice with the ideals which had first given birth to the web. In other words, Web 2.0 was not a continuation from Web 1.0 so much as a ‘reset and restart’ returning the web to its alpha version 0. This alpha version, according to Berners-Lee, was:

a common information space in which we communicate by sharing information [and] … dependent on the Web being so generally used that it became a realistic mirror (or in fact the primary embodiment) of the ways in which we work and play and socialize… we could then use computers to help us analyse it, make sense of what we are doing, where we individually fit in, and how we can better work together

In this respect, the web represented ideals of social practices through network connectivity inspired by the prototypical cultural forms of the Internet As Dean puts it “Web 2.0 designates nonetheless the surprising truth of computer-mediated interactions: the return of the human”.

Within this particular conception of Web 2.0 the web needed a restart because Web 1.0 represented the failure of 1990s business. Not appreciating the web’s ‘true’ origins and seeking only commercial gain, business had imposed upon the web ideas and expectations drawn from the traditional media. Not only had most businesses involved in Web 1.0 mismanaged their own affairs pursuing the illusory goal of media convergence, but in doing so had threatened the potential of the web to transform the world. The dotcom crash showed that the reality of how and why people used the Internet was not what business had thought and thus proved the original ideals of open communication, sharing and so on were not only good, but true: “the Internet was literally given back to the people” (Raffl et al.).

Yet the emergence of Web 2.0 saw a return to speculative behaviour and commercial exploitation of the common information space:

The 2005 Web 2.0 conference reminded me of Internet trade shows during the Bubble, full of prowling VCs looking for the next hot startup. There was that same odd atmosphere created by a large number of people determined not to miss out. Miss out on what? They didn’t know. Whatever was going to happen—whatever Web 2.0 turned out to be (Graham).

Thus if Web 2.0 was a return to an earlier time, before Web 1.0, it would be marked by the same political economy as the 1990s, as part of informational capitalism and with competing forces vying to constitute the web as that particular fusion of technology and capital necessary to their commercial interests. Thus Web 2.0 could not completely ‘reset’ the web’s development, for it was intrinsically part of the competition within capital over the ways in which to appropriate the value of consumers’ attention, labour and tastes.

The particular nature of the wrong direction of the web is best understood by a specific analysis to the problem of ‘design’ and its relationship to the technologies through which design came to dominate people’s internet-mediated interactions. The importance of design is evident in the brief article of DiNucci which, in 1999, first coined the term Web 2.0, and in the opposed views of good design  of visual designers (e.g. David Siegel) and HCI experts (e.g. Jakob Nielson). Exemplifying the way the web disrupted conventions and refused easy definition of norms and standards, the debate sums up the primary techno-capitalist challenge of the 1990s: how might  media and computing combine, whether within one site, or within corporations set on paths of convergence. Yet this debate shows how Web 1.0 had diverged from the origins of the Internet, which were resolutely outside of the media, forming a space of information and collaboration that specifically did not draw on classic media forms, tropes or models. By 1995, from the corporate media perspective, rather than being a novelty suited to computer enthusiasts, the “WWW was seen principally as something you watched, rather than something you built a small corner of yourself” (Roscoe) and the source of this maturity was the imposition of media models to explain its future significance.

Hagel at the time stated that “In many respects, Web 2.0 represents a return to origins of Internet”, portraying Web 1.0 as radical discontinuity from the ‘the web’ which might have existed.  Web 2.0 could then be proposed as a further discontinuity which would undo Web 1.0. However Web 2.0 could never fully return to that time as if there had not been this misdirection. The need for the version (rather than a return to just ‘the web’) stemmed from the fact that commercialization could not be undone, and a new form and approach was required. Furthermore, a vast ‘audience’ that had developed, the users whom Berners-Lee’s idealistic vision was to serve. Their needs and expectations, courtesy of Web 1.0, were not what had be assumed in that vision. Just as Web 1.0 could be said to have failed because it did not understand the cultures of use of the Internet, so too, there was no way to reset the web without now accounting for the new cultures of use that had emerged in the 1990s.

The discourse of versions

The emergence of Web 3.0 in recent years was inevitable once Web 2.0 started to be used. It is just too easy for technology evangelists to slip into the language of versions to help communicate their messages. But, it is not just superficial talk. A move from discussing ‘the web’ to discussing Web 2.0 creates the foundations for a teleology of development legitimising what had come before, was current, and what was still to come. The emergence of one version, to replace another, ineluctably requires there to be yet another version, still to come, which in time will become current and, eventually itself be replaced. How the web came to have a history shows us that, in building this meaningful narrative, a discourse of versions works in three ways.

First, the discourse enables a return to origins, to create the legitimacy for current moves that, far from being developed from the previous version, in fact realign with a trajectory of development originally intended. The recent past is placed to one side and ‘normal’ progress is resumed. Web 1.0 becomes the repressed other, only visible because it explains the contradiction of the move to Web 2.0 from the alpha version zero. In this repression, certain key features of the Internet in society (excess value appropriation; the complex relationship of the Internet and media; and contestations within informational capitalism) are strategically obscured.

Second, a discourse of versions enables a different movement from the past into the present, whereby the recent past is normalised as creating the pre-conditions for what has now emerged. The past is overturned by incorporating it within the self of the present, not repressing Web 1.0 but adding “1” to it. Here, version zero is repressed, since the latest iteration only references that which immediately preceded it and thus helps explain away contradictions which might produce a critique of the latest version.

Third, versions create the conditions for knowing and anticipating the future in an orderly manner, managing what is to come as astutely as what has been, positioning those who control the specific meaning of each revision as the authorities on what ought to be, based on their success in modifying current reality. Web 2.0 is a part-completed project, the model for Web 3.0: thus, the reasons for current failings and problems can be safely ignored because solutions are just one step away, to the next version. The discourse of versions enables ‘erasure’ the current version, even as it speaks it, locating attention towards the version next-to-come. The perfectibility of the Internet, and along with it, the whole technocratic project that it signifies, is reassured.


The dominant, popular history of the web is told through versions; these versions. provide the semiotic sites at which critical debates about financial, technological and regulatory issues can be played out, in a fight to define the future, through control over the meaning of the past and the referential present. For that reason there is no single, stable accounting of the versions and what they mean: the web is not software engineering, where versions represent agreed and defined iterations of the design and coding process. Yet, the origin of versions in engineering is apt: versions create order, control and mastery over a process that might otherwise become impossibly flawed in the absence of a consensus about the history of the application or product. A history of the web told in versions is all about the way that people seek to influence the direction of future development to suit their ideals, profits, or personal ambitions but only insofar as this historical account becomes the basis for collective, or shared, understanding.

What of the other sorts of histories to which we should pay attention? Shared history of versions of the web, where periods are defined, originators and pioneers identified, and generalisations made, occludes the private, personal histories of Internet use that tell of the individual experience of connectivity and which reveal a very different kind of relationship between technology and individuals. I will conclude with two examples.

Consider the case of Justin Hall, as described by Walker:

Hall’s narration of his life online began in January 1994, with a simple homepage, and extended into a detailed hypertextual version of his life told in traditional node and link HTML. When weblogging software began accessible, Hall started using it, and posted almost daily fragments in this decade-long autobiographical project until early 2005. At this point, Hall posted a video where he discussed the problems of publicly narrating one’s life at the same time as relating to the people in one’s life, and ceased his personal blogging.

This history (and there are many more like it) stands in stark contrast to the idea that Web 1.0 was a time of static, commercially oriented content produced for a mass audience and that this inappropriate form of the web was heroically undone by the expertise (whether business or technical) of the Web 2.0 revolution, enabling people to lead social lives online. Hall can be characterized as Web 2.0 before this term existed; and returned to Web 1.0 during the time of Web 2.0.

Second, the web is effectively designed by the preferences, behaviours and interests of its users and not by software engineers (or indeed communications designers). As Millard and Ross found:

… the relationship between Web 2.0 and those original [hypertext pioneers’] visions is more complex than [expected]: many of the aspirations of the hypertext community have been fulfilled in Web 2.0, but as a collection of diverse applications, interoperating on top of a common Web platform (rather than as one engineered hypertext system)… The Web 2.0 model is heterogeneous, ad-hoc, evolutionary rather than designed…

This example suggests an entirely different history of the web, consisting in the innumerable and largely invisible minute acts by all the users of the web which, in their effects, become a crowd-sourced history and, in the end, visible only in its effects or in the recollection of the place within the crowd which any individual occupied at a given time.

Web 2.0 engendered a history of the Internet (its technologies, peoples, businesses and politics) that both depends on and asserts the primacy of the discourse of versions as the correct way to tell this history. This effective claim that the only legitimate way in which web histories can be told is with due deference to the technical language of the originating discipline is, ultimately, the most profound consequence of Web 2.0. Perhaps, if now we are asking “Is Web 2.0 dead?” we might more positively ask: what other ways might we explore the histories of the web such that users’ agency in their own historicity is more fully realized.