CERN on March 13 celebrates the 20th anniversary of a proposal entitled, “Information Management: A Proposal,” by Tim Berners-Lee, which would become the blueprint for the World Wide Web.
Twenty years ago this month, a software consultant named Tim Berners-Lee at the European Organization for Nuclear Research (better known as CERN) hatched a plan for an open computer network to keep track of research at the particle physics laboratory in the suburbs of Geneva, Switzerland. Berners-Lee’s modestly titled “Information Management: A Proposal,” which he submitted to get a CERN grant, would become the blueprint for the World Wide Web.
The Web was not an overnight success. In fact, it took nearly two years before Berners-Lee—with help from CERN computer scientist Robert Cailliau and others—on Christmas Day 1990 set up the first successful communication between a Web browser and server via the Internet. This demonstration was followed by several more years of tireless lobbying by Berners-Lee, now 53, to convince professors, students, programmers and Internet enthusiasts to create more Web browsers and servers that would soon forever change the world of human communication.
On Friday March 13, Berners-Lee, Cailliau and other Web pioneers will gather at CERN to celebrate the 20th anniversary of that original proposal. To get the inside story on how the Web came to be, not to mention the man behind the idea, SciAm.com spoke with Scientific American editor Mark Fischetti, who in 1999 collaborated with Berners-Lee to write Weaving the Web: The Past, Present and Future of the World Wide Web by its Inventor, a seminal work that analyzed and commemorated Berners-Lee’s achievement a decade after the Web’s birth.
[An edited transcript of the interview follows.]
Why was the Web invented at CERN?
Tim Berners-Lee was a software consultant at CERN in the 1980s when he began writing Tangle, an application to help him keep track of CERN’s many scientists, projects and incompatible computers. Thousands of researchers would travel to CERN, do their experiments using their own computers (which they brought with them), and then go home to crunch the data. It was a major pain at CERN to accommodate the many incompatible computers, which also had to work with the CERN mainframe that actually ran the mammoth particle accelerators. Tim was responsible for helping everything and everyone work together. He thought it would be a whole lot simpler if the computers could swap their information directly, even though, at that time, computers didn’t communicate with one another.
March 2009 marks 20 years since Tim Berners-Lee first proposed a project that would become the World Wide Web. What inspired the larger vision?
He made the proposal to CERN management in March 1989 for funding and an official okay to use some of his time to work on this project. But in thinking about solving the incompatibility problem, he realized that it would be even more cool if the scientists, after they went back to their labs, could still share their data. They might even be able to run some of their experiments at CERN over a network from wherever they were located, if the distant CERN computers could talk over the Internet. The Internet itself is just a set of wires and a protocol for sending information over those wires. The Web would be an application that ran on the Internet. It just so happens that the Web turned out to be the killer app of all time. (Other Internet applications already existed, including File Transfer Protocol, or FTP, and e-mail.)
What were the key innovations that formed the Web? Who created them?
The three main innovations are HTTP (hypertext transfer protocol); URLs (universal resource locators, which Tim originally referred to as URIs, for universal resource indicators); and HTML (hypertext markup language). HTTP allows you to click on a link and be brought to that document or Web page. URLs serve as an address for finding that document or page. And HTML gives you the ability to put links in documents and pages so they connect. Tim created all three of these pieces of software code from October to December of 1990.
What’s the best analogy for explaining how the Web works?
Tim likens it to a market economy: anyone can trade with anyone else without having to go to a physical market square to do it. The traders just need to know the rules. The hardest thing for people to grasp about the Web is that it has no center; any computer (or node, in mathematical terms) can link to any other computer directly, without having to go through a central connection point. They just need to know the rules for communicating.
Berners-Lee accessed the first Web page, on the first Web server, using the first Web browser on Christmas Day 1990. Why did it take until 1993 before the public became aware of the creation?
Once Tim and Robert Cailliau established that the Web worked, they wanted to spread the word. After getting CERN to buy in, Tim spent 1991 flying around the world meeting with people who were interested in hypertext and the Internet and linking to create Web browsers to access what was a growing repository of information on Tim’s CERN computer. He also encouraged enthusiasts to start their own servers. From there, listservs helped spread the word; so did university computer science programs, which saw the coding of browsers and servers as a great way to get students to experiment. (One of the best known of these projects was headed by the University of Illinois’s Marc Andreessen, who would later transform his creation into the Netscape Web browser.) Tim began to get concerned, though, about universities and companies like Microsoft creating their own networks that might compete with the Web, or charging for content, which would violate his core principle: that everyone should be able to communicate freely with everyone else. To stop this from happening, he got management at CERN to release all of his source code under a general license so that any programmer anywhere could use it for free. He thought that if the whole world was building the Web together, no one company could take control of it.
What caused the Web to finally take off?
Tim designed the Web to be a social medium, first, rather than a technical one—a system that would connect people through their computers, and the grassroots building [of the Web] took off because of that. However, the general public didn’t really enter that picture until the mid-1990s, when companies like Netscape and AOL [America Online] commercialized browsers. These companies would snail mail free CDs with their browser software so people would get on the Web, hoping that once they got there, they would discover services the companies offered for a fee, such as e-mail.
Why did Berners-Lee abruptly leave CERN to begin the World Wide Web Consortium at the Massachusetts Institute of Technology in 1994, just as the Web began to rapidly expand?
At that point, the Web was clearly becoming a juggernaut, and commercial forces did indeed threaten those core principles. CERN was not in the business of overseeing Internet systems or applications—it existed to do high-energy physics experiments. Tim couldn’t be the caretaker and stay there, so he moved on to M.I.T.’s Laboratory for Computer Science, which became the host for a new World Wide Web Consortium, where Tim has been ever since.
What has most surprised him about the Web’s evolution?
What surprised Tim most is that for years people were so much more interested in simply browsing for and reading content rather than in creating it. His very first browser—WorldWideWeb—was actually both a browser and an editor. It let you write your own pages, post them online, and edit pages posted by others. But the commercial browsers didn’t offer editing capabilities. This frustrated him for a number of years. The whole point of the Web, to him, was not to just see information but to publish it, too. This didn’t really happen until blogs emerged, followed by sites like Facebook, where people can easily post content.
What does the future hold for the Web, given that the openness that Berners-Lee built into it is continually exploited by miscreants?
It’s hard to implement controls on the Web—because it was created in the ethos of the Internet—in that it’s totally open. But for Tim, confronting issues like privacy and protection of intellectual property is not a matter of a technical fix. First, you need a social fix. If the Web is open to good people, it’s open to bad people, too. The way you deal with security and other problems on the Web is the same way you deal with it in society: You need laws and social conventions that guide people’s behavior. Once those are developed, then the technical ways to implement them can be created.
Facts about web’s creation
First program by Tim Berners-Lee that attempted to link bits of data:
—Enquire, 1980, for Berners-Lee’s personal use as a software consultant at CERN; he later left and the code was lost
—Tangle, 1984, when Berners-Lee returned, to help him keep track of CERN’s many scientists, projects and incompatible computers
Early names for the Web:
—Information Mesh, Mine of Information, The Information Mine (But Berners-Lee thought the acronym, TIM, was too egocentric!)
Computer the Web code was written on, and Web browser was designed on:
—NeXT, by NeXT, Inc., founded by Steve Jobs, who had started Apple Computer earlier and returned to it later
Programming language used:
Time taken to write the code:
First Web browser:
—Called WorldWideWeb; it could edit Web pages as well as access them; it worked only on the NeXT platform
First server address:
—nxoc01.cern.ch (NeXT, Online Controls, 1), with an alias of info.cern.ch
First full demonstration:
—Christmas Day 1990, operating over the Internet from Berners-Lee’s NeXT machine to the NeXT computer of his office partner and now Web co-developer, Robert Cailliau
Content of first Web page:
—The CERN phone directory
First U.S. Web server:
—April 1991, hosted by the Stanford University Linear Accelerator lab
Hits (pages viewed) on the info.cern.ch server:
August 1991: 100 a day
August 1992: 1,000 a day
August 1993: 10,000 a day
First Web browsers:
—WorldWideWeb, December 1990, for the NeXT platform, by Berners-Lee
—Erwise, April 1992, for Unix, by students at Helsinki University of Technology
—Viola, May 1992, for Unix, by student Pei Wei at the University of California, Berkeley
—Samba, summer 1992, for Macintosh, by Robert Cailliau at CERN, finished by intern Nicola Pellow
Notable early servers that showed the Web’s complex capabilities:
—1992, virtual museum of objects in the Vatican, by programmer Frans van Hoesel
—1992, virtual geographic maps, with pan and zoom, by Steve Putz at Xerox PARC
Courtesy: Scientific American
Berners-Lee returns to CERN to reminisce on the Web’s past and focus on its future
Computer scientists, engineers and journalists converged on the CERN particle physics lab in the suburbs of Geneva, Switzerland, today to pay homage to a piece of paper—several pieces of paper, actually—that together form Tim Berners-Lee’s March 1989 proposal that would come to be the blueprint for the World Wide Web.
Berners-Lee, the one-time CERN software consultant who went on to invent the Web and found the World Wide Web Consortium (W3C) at the Massachusetts Institute of Technology (M.I.T.), began his keynote today commemorating the 20th anniversary of his proposal with a copy of his now-famous document in hand. “I wrote it 20 years ago, 20 years ago nothing happened,” he said, referring to the seven months the proposal languished on his supervisor’s desk before in September that year he was given money to buy some computers and pursue his idea. (For more coverage of the Web’s 20th anniversary, see Scientific American.com‘s in-depth report.)
The Web came to life on Christmas day 1990 and grew exponentially from that moment on. One of the reasons for the its phenomenal success was Berners-Lee’s insistence that there be one Web for everyone to use, regardless of the type of computer, software program or documents they were using. For years, he says he was concerned that the original Web would split into many specialized webs, such as one for academia and another for businesses. But in the end his vision prevailed. “Universality,” he said, “that was the rule, and it worked.”
In a speech that touched on the past but also emphasized the Web’s future (including its potential benefits as well as dangers), Berners-Lee pointed out that there are 100 billion Web pages today, roughly the same number of neurons in the human brain. The difference, he added, is that the number of pages grows as the Web ages, whereas the number of nerve cells shrinks as we get on in years.
“One of the dangers of celebrating anything is having people look back and [focus on] what we did,” he said today. “But the rate of creative new design on the Web is getting faster and faster. The Web is not done; it’s just the tip of the iceberg. I’m convinced that the things that are going to happen will rock the boat even more.”
One of the W3C’s priorities is promoting access to the Web via mobile devices, phones in particular. “There are more browsers on phones than on laptops, by a long shot,” Berners-Lee said. Another major priority is maintaining the universality that he had in mind when he envisioned the Web. “Eighty percent of people can’t access the Web,” he said, because, among other reasons, they can’t get a connection or the Web pages they access are written in a language they can’t read.
As the Web grows, so do concerns about the confidentiality of private info on it. The main worry is that a hacker will access and use personal information such as credit card, banking account or Social Security numbers. One possible way to avoid this, says Berners-Lee, is to specify how the data may be used. “As the data is moved around,” he said, “the appropriate use is tagged along with it.” The tag, or set of instructions accompanying the data, would prevent it from being misused.
This suggestion is part of Berners-Lee’s vision for a “Semantic Web” that would be easier to surf than the Web is today. In the Semantic Web, search engines would focus more on finding the information you’re looking for, rather than simply locating Web pages that might contain that information. One way to do this, he said, would be to change the way new data is added to the Web so that it can be immediately linked to other data, making it easier to find. W3C is working on a way to do this through its Linked Open Data project, a key component of the Semantic Web.
An example of how the Semantic Web would work, which the W3C has posted to its Web site: you could populate your Web-based appointment calendar (such as those offered by Google or Yahoo) with appointments as well as with other dated information to which you have access (such as bank statements and digital photos).
Image of CERN’s Globe building, where Tim Berners-Lee and others celebrated on March 13, by Jim Shank via Flickr
Courtesy: Scientific American