Weaving the Web, by Tim Berners-Lee with Mark Fischetti (1999)

Review by Louis Eisenberg, 3-18-03

Despite what Rush Limbaugh may have told you, Al Gore did not invent the Internet, nor did he ever claim to. Neither did Tim Berners-Lee. Berners-Lee invented something much, much bigger. While Berners-Lee’s brainchild, the World Wide Web, merely harnessed the decentralized power of the already-established Internet, it is unquestionably the innovation that transformed the global network from a nerd sanctuary into a mainstream phenomenon and the focal point of the global society.

The subheading of the book reads, “The original design and ultimate destiny of the World Wide Web.” Indeed, this is really two separate volumes. First, Berners-Lee chronicles the Web’s journey from a vague notion in his mind in the 1980’s to its present-day ubiquity. While the tale is inspirational and offers a few larger lessons about perseverance and the secondary value of material wealth, it is more or less what you would expect: Tim develops brilliant new idea, other people resist, Tim tries harder and wins people over, movement grows, Tim expands on original idea, and so on. What this account really does is to prepare you for the second half of the book, which is Berners-Lee’s elaborate vision for the future of his creation. It’s reminiscent of the parts of the Bible where God ruminates the destiny of mankind, or perhaps when parents discuss their children. Since Berners-Lee unleashed it on the world, the Web has grown into something far bigger than him, of course; yet he maintains its stewardship as the director of the World Wide Web Consortium (W3C), the international coordinating body for its development.

We might wonder how Berners-Lee will face this new reality. Will he be the protective father, holding his child’s hand and warding off evil at every step? Or is he ready to let go and allow the Web to confront the world on its own? The former is impossible given the Web’s inherent decentralized nature. Yet Berners-Lee insists on a Web with some semblance of coordination. Without the W3C and other organizations to set standards for how people communicate with each other, the Web would have descended long ago into chaos. “It is important that we not be blind to the need for governance where centralization does exist,” he writes, “just because the general rule on the Internet is that decentralization makes central government unnecessary” (128).

The plans that Berners-Lee lays out dispel any fears we might have that he will be a reactionary old fogey who insists on preserving the Web in its original form; rather, he continues to embrace his role as an innovator in an evolving system. He understands that any attempt to impose a hierarchy on the Web would destroy the essence of its value. At the same time, he sees a desperate need for guidelines that will enable us to exploit the Web’s value in spite of that decentralization. The Web he foresees is more than just a web of information: it is a web of people, a web of privacy and trust, a web of knowledge and meaning, and ultimately, a web of peace and understanding.

Web of People

“The ultimate goal of the Web,” says Berners-Lee, “is to support and improve our weblike existence in the world” (123). Offline, society is already a vast network of links -- relationships both weak and strong -- between individuals. In many ways, he sees the Web as an extension of this network. As it is offline, the Web is -- and should be, according to Berners-Lee -- a scale-free network, with many less-connected nodes and few highly-connected nodes. But the online world is not a new or “virtual” reality; it is an augmentation of the reality we’ve always had. Instead of creating an entirely new set of nodes, the Web builds new and stronger links between the existing nodes.

At the same time, Berners-Lee fears that the rapid evolution of the Web has left society behind, “trying to catch up on ethical, legal, and social implications” (123). If people are to feel comfortable entering and fully participating in the meta-society of the Web, they will need certain assurances. Issues like information quality, privacy, and trust, says Berners-Lee, are “fundamental values in society, much misunderstood on the Web, and alas highly susceptible to exploitation by those who can find a way” (125). We expect our interactions with other people to conform to certain standards of honesty, integrity, and reliability. On the relatively new and uncertain medium of the Web, an individual user often has little at his disposal for determining how comfortable he should feel about the person or website with which he’s interacting. The old rules just don’t really apply. For the Web to realize its full connective potential, people have to be able to achieve the same kind of comfort that they experience in the offline world.

One element in that comfort is confidence in the quality and unbiased nature of the information that we receive. Berners-Lee worries about the vertical integration of Web-oriented firms. If both content and conduit fall under one company’s control, we can no longer trust that we’re receiving the best information; in the use of its distribution channel, the company faces a conflict of interest between serving its users and promoting its own content. For example, a browser program that continually directs its users toward certain websites is denying them a full, unconstrained view of the Web. He worries that such blurring of priorities could destroy the Web’s credibility. If Amazon only recommends the books that the publishers pay them to recommend, then what credence can we give to their recommendations? Imagine this type of misinformation becoming the prevailing trend throughout the Web.

Another important consideration is whether we can avoid seeing the information that we find objectionable -- and keep our children from seeing it. Berners-Lee adamantly opposes government intervention in this area. He emphasizes the distinction between an individual choosing how she filters content and the government imposing a filter on an individual without her consent. A major initiative by the W3C has been to oppose legislated censorship and to promote the feasibility of voluntary filtering solutions and rating systems. People make informed choices about what they view in the offline world; online should be no different.

Web of Privacy and Trust

The essential factors in our online interactions, says Berners-Lee, are privacy and trust. Privacy has been a concern for Web users from the outset. When companies first began selling products online, their transactions were often susceptible to credit card theft by devious hackers. The spread of encryption technology has mostly resolved that problem, but now consumers worry about the companies themselves: what kind of information are they collecting and to whom are they selling it? Berners-Lee offers an anecdote that is both amusing and scarily accurate:

A famous cartoon drawn early in the Internet’s life depicts two dogs sitting at a computer. One explains to the other, “The great thing about the Internet is no one knows you’re a dog.” It has been followed recently by another cartoon in which one dog has clicked to a page with a picture of dog food. Because of this, the server now does know it’s a dog. Pretty soon the server also knows it’s a dog that prefers a certain brand of dog food, elm trees, and Siamese cats (144-5).

While European governments are very protective of consumers’ rights, imposing strict limitations on how companies can share information, the U.S. government has generally been hands-off. While businesses have taken some small steps -- for example, in 1998 a number of major corporations and other organizations formed the Online Privacy Alliance -- no one who receives 30 “spam” emails per day would say that there isn’t substantial room for improvement. The W3C’s response is the Platform for Privacy Preferences Project (P3P), a new protocol that would enable a user’s web browser to negotiate automatically an acceptable privacy policy with each website that the user visits. This approach is a perfectly natural extension of Berners-Lee’s general philosophy. No one tells us who we can and can’t associate with; we deal with people and businesses of our choosing when the parties can find mutually acceptable terms for interacting.

In that same spirit, he envisions a “Web of Trust” that involves links of trust among people, businesses, and organizations. Currently the only major form of “trust” on the Web is the hierarchical Public Key Infrastructure, an industry-designed method for giving consumers confidence in e-commerce transactions. By contrast, the model for this web of trust is the existing network of email authentication/encryption called Pretty Good Privacy (PGP). A completely decentralized system, PGP allows users to assign values to other users indicating how much they trust them; by following the links between individuals, a user can calculate how much he should trust any other user, even if the two are complete strangers. This is a tangible application of the “six degrees of separation” property of our social network that people refer to so often. If we could extend a PGP-like system beyond email and apply it to the Web, users would be able to determine how much to trust the websites they visit -- and vice versa.

Web of Knowledge and Meaning

Perhaps the most interesting development that Berners-Lee proposes is the creation of a “Semantic Web.” The idea behind the Semantic Web is to employ machine-readable languages to store more information on webpages, enabling computers to perform much more sophisticated search and analysis tasks. XML (eXtensible Markup Language) and RDF (Resource Description Framework) allow people to define a semantics for the contents of their webpages. For example, if a spreadsheet of information is stored on a website using RDF, an automated “agent” can interpret that information without being tailored specifically for that site. A major problem with the Web currently is that it can be extremely difficult to find something, even when you know it’s out there. Search engines give us some hope of locating what we’re looking for, but they’re a pretty blunt instrument considering that the essence of their strategy is to find pages that contain the same set of words that appear in our query. If a sizable portion of the Web uses this approach, however, search engines will be able to respond to queries with intelligent strategies. If we then use knowledge representation to create inference rules that link concepts together -- for example, saying that “family name” and “surname” are the same, or that “car” is a type of “vehicle” -- computers will be able to use logic to solve problems that no one would dream of handing to a computer right now.

Again we see Berners-Lee’s philosophy in action. The beauty of the Semantic Web is that, because of its purely decentralized structure, no one set of interpretations or meanings will dominate. The Semantic Web learns to associate concepts based on the contributions of many independent sources. Writes Berners-Lee, “The Semantic Web will work when terms are generally agreed upon, when they are not, and most often in the real-life fractal [scale-free] mess of terms that have various degrees of acceptance” (187-8). Yet what makes something like the Semantic Web possible in the first place is the establishment of a standard syntax by an organization like the W3C. Given a bit of gentle guidance by its overseers, the Web can be turned loose and flourish.

Web of Peace and Understanding

In the end, Berners-Lee thinks the Web can even bring us peace and goodwill among men. Through the Web of Trust, the Semantic Web, and the other “webs” that he envisions as part of the ultimate World Wide Web, we will learn to work together and understand our differences in a whole new way. “A society that could advance with intercreativity [interaction where everyone can contribute information] and group intuition rather than conflict as the basic mechanism would be a major change,” he writes. In the exchange of ideas, the Web can become a vehicle for intelligent, rational debate, replacing what he sees as the far too common current practice of ad hominem attacks and angry disputes.

The most exciting part about all of Berners-Lee’s accomplishments is that they feel like such natural extensions of his way of seeing the world. He believes in freedom of expression, respect for differing viewpoints, and the building of consensus through decentralized networks instead of trees of authority. He only seeks uniformity when it involves a standard of communication that will improve efficiency; and even then, his goal is merely to suggest the standard, not to impose it on anyone. When you recognize the elegance of the philosophy driving his work, you understand that a creation like the Web could only come from someone with ideals as pure as his. Is he being unrealistically ambitious with his dream of a Web that brings world peace? Of course he is. But it’s hard not to take him seriously when you see what he’s accomplished already.