Charting the Web's next transformation
Tim Berners-Lee assesses the future of his seminal invention and cautions about the dangers of not letting it remain an independent, impartial medium.
While a student at Oxford University in the mid-1970s, Tim Berners-Lee took a soldering iron and cobbled together a computer from TTL gates, an M6800 processor and an old television. Fifteen years later, demonstrating the same knack for invention, the software engineer radically changed the history of computing when he wrote the protocols that define the World Wide Web.
Since that momentous invention, Berners-Lee has led the World Wide Web Consortium (W3C), the Web's central standards body. As the consortium's director, Berners-Lee has guided the Web's underlying protocols as the Web has experienced its explosive, unanticipated growth.
Navigating a shifting and treacherous industrial-political landscape, Berners-Lee has held together working groups that combine the Web's most potent and competitive archrivals, including Microsoft, AOL Time Warner and its Netscape Communications unit, Sun Microsystems, and hundreds of others. He has steered the development of technologies ranging from the Web's lingua franca, HTML, to its data-structuring metalanguage XML and the formatting protocol CSS.
This week, as the United States marks the 10th anniversary of its first Web page, Berners-Lee is distributing some of his authority with the creation of a Technical Architecture Group (TAG) that will articulate principles behind the Web's architecture and clarify them when conflicts arise. The W3C is adamant that the group does not "replace" the director (who is a member of the newly formed group), but W3C observers call TAG a significant move toward the kind of democratization that many have seen as a hallmark of the Internet as a whole and of the Web in particular.
Born in London in 1955, Berners-Lee designed the first Web protocols when acting as a software engineering consultant at CERN, the European Particle Physics Laboratory in Geneva. He is the author of the 1999 book "Weaving the Web" about his seminal invention.
Berners-Lee recalled his early work on the Web and looked ahead to the Web's future in an e-mail interview with CNET News.com.
Q: Paul Kunz has said his early enthusiasm for the Web was a result of its ability to connect people to SLAC (Stanford Linear Accelerator Center) databases for research papers. What was your motivation in writing the application?
A: The aim was unification of all the many information systems, each of which did different useful things, but which did not interoperate. My role at CERN was not that of a physicist but that of a software engineer, so my personal needs were for tools to support collaborative software design. This included brainstorming spaces, records of decision, documentation, project management, and so on.
How long did it serve primarily this function?
For most people, it never was a collaborative space. The "killer app" at CERN which got it onto the desks was the phone book, served from a database on the mainframe. The next things were documentation and library systems.
What was the key event in widening the Web's use beyond scientific research?
People often ask for "key events" as though life was analyzable into such things. The Web actually spread, initially along three axes--in High Energy Physics, among NeXT users (who could try out the original application), and among people interested in hypertext applications. The explosion was amazingly steady--for an explosion. The load on the first Web server just grew a factor of 10 every year for three years.
The reason for this was that it was a grassroots syndrome. It happened because of a mass of small decisions made across the world: here a student deciding what project to do, there a few partially informed middle managers wondering about which technology to use, somewhere a sysop deciding to install something fun after a long night's work. A very significant factor was that the software was all (what we now call) open source. It spread fast, and could be improved fast--and it could be installed within government and large industry without having to go through a procurement process.
When did it really explode into the mainstream, and what was your response?
For me, it was a slow transition from the feeling of needing to push the bobsled to the feeling one had to jump in and steer. This happened between 1992 and 1994, when the World Wide Web Consortium started.
The Web was born in Switzerland, but today we're looking at the 10th anniversary of its spread to the U.S. What was special about the way the Web developed here? Was there something unique about the American environment that helped it flourish?
The Web was born in an international lab, and it grew internationally. It spread well in the U.S.A. because the Internet was already more widely deployed, and because the cultural block had a very large number of potential readers for each new Web page and potential authors for each new reader. The snowball effect of the Web's growth worked very well in the U.S.A.
Converse question: What about the American character rubbed off on the mainstream Web that you wish hadn't?
There isn't a "mainstream" Web. It isn't like that. There is no "party line." It is a decentralized system, where the existence of a bunch of commercial sites in one place doesn't affect or detract from the quite independent existence of some great free academic material in another. The fact that there is porn somewhere else does not affect those reading the poetry.Any regrets about the original Web protocol? If you could start over and do anything differently, what would it be? Two protocols, one for the academy and one for the rest of us?
Certainly not two protocols. The universality of the Web is key. You break it in half and either half has lost the essential power--to be able to link to anything.
If I had the chance to revisit it, I would have written into HTTP that a server undertakes to respect the privacy of the browsing person. And I think I would have found a way to do without the double-slash!
What do you predict for the Web's next 10 years?
The Semantic Web is very exciting, and now just starting off in the same grassroots mode as the Web did 10 years ago: All kinds of people making open-source software. In 10 years it will in turn have revolutionized the way we do business, collaborate and learn.
What were the four or five most important milestones in the Web's
development, beyond your invention of it and Paul Kunz's setting up the first
U.S. Web page?
One crucial step was, after 18 months of asking, CERN's signing of a document renouncing the rights to charge royalties for the technology. The U.S. government's decision to allow commercial use of the Net was very important. Pei Wei's decision to make "Viola," and later, Marc Andreessen's decision to make "Mosaic," provided Web browsers for more systems than the NeXT. Also, Nicola Pellow's "www" line-mode browser allowed people to get at the Web from older DOS, Unix and mainframe computers, until no one with a computer and a Net connection had an excuse not to have a browser.
( See http://www.w3.org/History.html for more milestones.)
What was the key difference between the Web and the computer networks that already existed?
The key is the UDI (now URI or URL) design. The concept was that whatever system of names and addresses was used for a set of information, from filenames to ISBN numbers, it could be coded up one way or other as a UDI--a Universal Document Identifier. This would mean that any conceivable information on the network could have a URI, and so a hypertext link could be made to it.
The HTTP specification specifies more about what happens with URIs starting with "http:" and HTML (now xHTML) specifies how Web pages are marked up, but the URI specification was and is the key.
Since then, W3C has constantly been working to make sure, even as we can do more and more amazing things in URI space, that it still remains a neutral space, accessible independent of the computer hardware you use, the software you bought, the network supplier you chose, the language you speak, any disabilities you might have, and your cultural background. Only when the medium remains impartial and universal can it remain a force for individual, regional and global understanding.
