The Web as Information Paradigm

The following essay was my term paper for Physics/Philosophy 419: Space, Time, & Matter, taught at the University of Illinois Urbana-Champaign by Professor Philip Phillips. I took the course as a senior in Spring 2021.

As one of the few significant essays I've written (I graduated with a B.S. in Computer Science, so we didn't write in English all that often), there will likely be bugs. Any spelling or grammatical errors, stylistic inconsistencies, or citation issues are not intentional and I would love for you to make me aware of them.

In addition, I wrote this essay without having completed Weaving the Web by Sir Tim Berners-Lee. After reading firsthand his ideas about the web's inception and evolving goals, I would probably have reworded the Normal Science section in particular, and minor other sentences of the essay in general. Either way, I wanted this blog to contain the term paper in its original form.

Logistics and reflections aside, this essay argues that the World Wide Web, as we know it, fits within the Kuhnian framework for a paradigm. The implication is that a new paradigm will, ultimately, supplant the Web as our shared medium of digital information exchange.

Introduction

Thomas Kuhn induced a paradigm shift with his seminal book, The Structure of Scientific Revolutions. In particular, this shift occurred in the ways we collectively view the process of discovery: the manner in which new ideas are conceived and subsequently refined. Kuhn’s work was focused on the natural sciences, specifically physics. However, the reach of Kuhn’s work extends to many other domains. Scholars in mathematics and linguistics, for example, have employed Kuhn’s framework to understand the development of their own fields. In addition, the notion of a “paradigm shift” permeates popular culture, particularly among entrepreneurs and corporations purporting to offer groundbreaking new products.

One technology of world-changing consequence was the World Wide Web (WWW, or just, the Web). Invented and implemented by Sir Tim Berners-Lee in the early 1990s, the Web is the mechanism by which we access information online. It is a service, much like email or Usenet, that sits on top of the Internet. Many people forget that the Web was an invented technology that followed the internet, and that it has an idiosyncratic history worthy of study. What, though, is the best framework by which to analyze the technical development of the Web? What about its broader impact on technology and society? Many people, and especially technologists, view the Web as an unchanging and amorphous medium for information. But the Web is not without its problems, many of which are becoming more pronounced in our increasingly-digitized era.

Like Newtonian or Einsteinian mechanics in physics, the World Wide Web, and particularly its utility to humans, can be understood as a paradigm. By analyzing the Web’s structure from this perspective, I will argue that we can collectively escape from its limitations -- and, consequently, reimagine the possibilities for information exchange.

The Structure of the Web

What follows in this section is a fitting of the Web -- its development and its impacts -- into the Kuhnian philosophy of science. There is certainly not a one-to-one correspondence, but nevertheless, many of Kuhn’s most famous ideas -- namely, the pre-paradigm period and schools, normal science, anomaly, and crisis -- apply surprisingly well. The main implication of understanding the Web from this perspective, as we will detail in Section III, is a reframing: the Web, although the first mainstream mechanism for digital information access, does not necessarily prefigure our information future.

Pre-Paradigm Period and Schools

In the decades leading up to the Web’s inception, there was increasing awareness about the opportunities in information science presented by computers. Moore’s Law, formulated in 1965, stated that the number of transistors in an integrated circuit would double roughly every two years; in other words, computing power and storage capacity was exponentially increasing. However, there was a commensurate number of ideas about what the tools for information access ought to look like. I will discuss two of the main projects pegged, in different ways, as alternatives to the World Wide Web: Project Xanadu and the Gopher Project.

Arguably the underlying technology of the Web is known as hypertext: text that embeds references (hyperlinks) to other text. The idea for hypertext, though, actually comes from Ted Nelson, the founder of Project Xanadu. Nelson saw physical paper as a suffocating, dimension-bound, and single-sequence object. Computers, on the other hand, could be used to parallelize reading/writing and clearly display the interconnections between documents. Although the World Wide Web relies on Nelson’s concept of a hyperlink, it leaves out the visible bridges between linked documents, and there is certainly no built-in notion of parallelism. Project Xanadu, despite being officially founded in 1960, did not produce a working prototype until 1998 -- several years after the Web’s founding and subsequent growth. To this day, Nelson continues work on “The Original Hypertext Project”, and even states on his project’s homepage: “The computer world is not just technicality and razzle-dazzle. It is a continual war over software politics and paradigms. With ideas which are still radical, WE FIGHT ON” (Project Xanadu).

The Gopher Project, born at the University of Minnesota and bearing the same name as its mascot, was a much more practical Web alternative than Project Xanadu. The Gopher system had a number of key features, some of which even influenced the Web: file-like hierarchical management, more rigid document structure (in contrast to the relative free-formness of HTML in the Web), and file system tools like search engines (Kirscht). One of its most important elements is simplicity: because it doesn’t have focus on page styling or commercialization, information is presented in a more straightforward way than most modern websites. Despite its benefits, the popularity of the Mosaic web browser, which targeted web technologies, and Minnesota’s introduction of licensing fees for the Gopher software made the Web a more attractive option for most computer users. Still, it is important to note that some technology hobbyists still make scant use of the Gopher project today.

In sum, before the widespread adoption of the Web in the mid-1990s, the ecosystem for digital information tooling was vibrant and diverse. This system closely resembles, in my view, the pre-paradigm eras for many of the major natural sciences. As we all know, the Web won out, and most technologists, especially those younger, have little knowledge of the historical development of the platform -- the Web -- in which most software is built on top of. Still, there is an additional similarity between the Web and natural science pre-paradigm eras that cannot be ignored. Ted Nelson continues to fight for the “radical” future that he envisioned back in 1960. Now in his 80s, Nelson bears striking similarities to the older natural scientists who, Kuhn writes, never fully accept a new paradigm.

Normal Science

The original goal of the Web was to connect people and information. In the decades following the Web’s inception, many efforts have been undertaken to advance this goal. These developments are analogous to normal science, in that they match facts with and aim to refine the existing paradigm (namely, the Web). To illustrate this, I will trace a brief history of the Web post-1990, touching particularly on the popular -- and unofficial -- terms “Web 1.0”, “Web 2.0”, and “Web 3.0.”

“Web 1.0” was the retronym applied to the first era of the World Wide Web. Arguably, this version of the Web was the most simple: webpages were static, content was provided from a server’s filesystem (as opposed to a database), and users were passive consumers of websites.

Seeing the Web as burgeoning technology, developers from the University of Illinois sought to make information on the Web more accessible. The result was Netscape, the company, and Netscape Navigator: a web browser with a focus on user-friendly design. Until 1994, when Netscape released its browser, early web-surfing applications were notoriously dull, devoid of multimedia and color. Navigator, on the other hand, supported video clips, sound, forms, and bookmarks -- this diverse set of features made it the standout among early browsers and, at large, greatly accelerated the Web’s growth (Blitz).

As an agent of the Web’s larger mission, Navigator was highly successful. In particular, it seemed to move the Web much closer to actualizing its goal of being accessible: without an elegant and featureful web browser, it is likely that the World Wide Web wouldn’t have been widely adopted by non-technologists. Although Navigator was superseded by Internet Explorer and, later, Google, it had a decidedly positive impact on the Web’s accessibility. In the Kuhnian sense, Navigator represents a perfect example of matching facts -- modern technologies -- with theory -- the Web’s goals.

The term “Web 2.0” was coined in 1999 by Darcy DiNucci, and tacitly defined the “Web 1.0” just discussed. This new era in the Web’s history was marked by tremendous advances in interactivity; that is, the ability of users to create and consume web content dynamically. Some of the features of Web 2.0 include user-generated content; APIs to automated usage of websites by, for example, other applications; and rich user experiences enabled by new web technologies like JavaScript, Ajax, and Adobe Flash.

The essential attribute of Web 2.0, and arguably that which we most associate with the Web now, is user participation. Facebook, despite its modern controversies, embodies this characteristic better than nearly any other piece of software. As noted above, Web 1.0 was designed to facilitate user consumption of online information -- websites, web browsers, and web technologies were all tailored to satisfy this need. Sites like Facebook, however, reimagined the possibilities of the Web, and specifically, the user’s relationship with information. They saw the real potential of the Web as a connector: not just of people to information or information to information, via websites and hyperlinks, respectively, but of people to people.

Facebook, and related services, represent an articulation of the Web’s paradigm. The minds behind this wave of software extended what the Web was capable of, technologically and ideologically, but they did not altogether supplant the Web (which would have constituted a paradigm shift). Web 2.0, in my view, is aptly named even if it is an informal designation: nothing about the concept of the Web fundamentally changed.

Anomaly

As Web 2.0 progressed, a fundamental problem was identified: to take full advantage of the Web, users had to give up their data. Moreover, their data was often owned by several different service providers; a typical user may, for instance, give substantial personal information to Facebook, Google, and LinkedIn. From the perspective of the Web as a paradigm, data privacy was a veritable anomaly.

The specific features of “Web 3.0” are not as clearly defined as its 1.0 and 2.0 predecessors, in large part because it doesn’t exist yet. However, Tim Berners-Lee, the Web’s inventor, has been working to resolve the aforementioned anomaly and consequently bring about a new web. As he puts it, Web 3.0 will be defined by people having complete ownership over their data, rather than corporations. He plans to get us there via the Solid project and Inrupt, the company leading its development.

To understand Solid, we must briefly touch on a core technology of the Web: the client-server model, wherein a client (user) sends and receives data from the server (company, like Facebook). Typically, when a user creates an account, their data will be permanently stored “server-side”. As we’ve discussed, this is especially problematic when you are interacting with multiple servers, since each additional copy of your data further impinges on your privacy. This model also makes it more likely that your data could be stolen, since any of these servers could theoretically be breached.

The idea behind the Solid project is a decentralized data store, called a Pod. As described on Solid’s website, Pods are like “secure personal web servers for data.” Through their Pod, people can specify which applications can access their data, and can grant or revoke access at any time. The business model for Inrupt, Berners-Lee’s company, is to manage user Pods by simplifying the sign-up process and abstracting the technical details of running a Pod. Though there is the option to self-host -- that is, to set up one’s own Pod on a personal computer or cloud server -- this option is likely infeasible for any non-technical person (Solid Project).

Crisis

In my view, Solid and Inrupt is Berners-Lee’s attempt to resolve the privacy anomaly. This makes sense, because Berners-Lee is unable to envision alternatives to the Web itself. He is making, more or less, an ad-hoc modification to the paradigm -- namely, that a user’s data should not be provided and should instead be licensed. However, Solid and Inrupt, as currently designed, is likely to fail in this goal. By requiring users to store their data in a company-hosted Pod, user privacy is still at risk, albeit on less fronts. More pernicious, though, is the fact that even in Web 3.0, corporations will still run applications to which users will connect for service. Web usage will therefore continue to involve fundamentally-unwanted third-parties. Undoubtedly, the Web and its community are in crisis.

The Resolution of Crisis

As Kuhn describes in Structure, a key sign of crisis is when the field’s eminent authorities begin to take notice of an anomaly (Kuhn, 67). Berners-Lee’s work on Solid has shown this to already be the case: he is desperately trying to resolve the Web’s inherent data privacy issues. However, as crisis goes, his work is likely to, at the very best, become a band-aid fix -- patchwork solutions can only go so far.

What our information age needs is a new paradigm, a new theory of information access and exchange. And there is at least one, that I can identify, which seems to have massive potential: Urbit. In its own words, Urbit is a “peer-to-peer internet being built from scratch to be more private, secure, and durable than the current internet.” Let’s analyze what exactly that means against the backdrop of the modern web (Urbit).

First, some terminology for the following paragraphs: Urbit is the name of the project, for which we just gave a high-level definition. In addition, “an Urbit” is an informal way to reference a personal server connected to the Urbit network. This will make more sense in a moment.

Key features of the Urbit stack:

To many technologists, the ideas of Urbit seem radical. And indeed, for these people, they are. As Kuhn would put it, the ideas propounded by the Urbit project are incommensurable with those of our web-dominated world. But the unique nature of paradigms is that they will ultimately be replaced. Urbit exists as one such replacement, but there will certainly be others.

Conclusion

Conceiving of the Web as a paradigm would seem to violate the very notion of technological advance. After all, many would say we live in a world whose technology has built upon its predecessors in a perfectly explainable, linear fashion. However, I contend that technological advance is unrelated to the central question of how we, as humans, use our tools. Technology has paradigms in the same way that natural science has paradigms: instead of trying to explain nature, as the sciences do, technology instead tries to improve humankind’s relationship to nature (and, by extension, to one another). When our tools are no longer serving our needs, we replace them.

Shifting away from the Web paradigm is not to abandon the Web: core technologies like hyperlinks, multi-way communication, search, and browsers will continue to influence future tools and ideas for information exchange, much as they have done in the Urbit project. Rather, this shift is about digital agency: how we recapture it, and how we preserve it.

Bibliography