The World Wide Web
A software engineer’s memo about the things CERN kept forgetting became the World Wide Web, and the move that made it universal was giving it away for nothing.

In March 1989 a thirty-three-year-old software engineer at CERN, the particle-physics laboratory on the French–Swiss border outside Geneva, gave his supervisor a memo about forgetting. Tim Berners-Lee had noticed that the laboratory kept losing what it knew. Several thousand people worked there, a typical stay lasted two years, and when someone left, the details of a detector or a piece of code often went with them. “Often, the information has been recorded, it just cannot be found,” he wrote. His remedy was a web of notes joined by links, in which any note could point at any other and the notes could live on whichever computers the laboratory happened to own. Mike Sendall read the proposal and wrote “Vague but exciting” on it. He approved no project. He did agree to buy Berners-Lee a NeXT, the black workstation sold by Steve Jobs’s second company, and suggested he use it to play with the idea.
The idea had an older root. During a six-month contract at CERN in 1980 he had written himself a program for tracking who worked on what and which part depended on which, and named it Enquire, after Enquire Within upon Everything, a Victorian household manual. Hypertext was already a familiar word by 1989. Ted Nelson had coined it in the 1960s, and Douglas Engelbart had clicked through linked documents in front of an auditorium of engineers in 1968. Berners-Lee proposed linking at the scale of the whole network, with links that crossed from one machine to another, no central index, and nobody’s permission required to add a page.
He began writing the code in October 1990, and by Christmas he had the three inventions that still define the Web. The first is the address, now called a URL, which names a method of fetching, a machine and a path: http://info.cern.ch/hypertext/WWW/TheProject.html tells a browser to speak HTTP to the computer called info.cern.ch and ask it for one particular file. The second is HTTP itself, the brief exchange in which a browser requests a page and a server sends it back. The third is HTML, the markup that lets a link sit inside the running text of a document. On the NeXT he wrote the first server and the first browser, which he called WorldWideWeb and which could edit pages as well as display them. The first website, a description of the project, ran on that cube in his office under a label in red ink: “This machine is a server. DO NOT POWER IT DOWN!!”
Here is the distinction to carry into any conversation. The internet and the Web are different things, and the internet is two decades older. The internet is the network, the descendant of the ARPANET, which had been passing packets between computers since 1969 and carried email, file transfers and remote logins long before anyone browsed. The Web is one application running over that network, a way of publishing documents and joining them with links. Late in 1993 it accounted for about one percent of internet traffic.
The WorldWideWeb (WWW) project aims to allow links to be made to any information anywhere. The address format includes an access method (=namespace), and for most name spaces a hostname and some sort of path.
We have a prototype hypertext editor for the NeXT, and a browser for line mode terminals which runs on almost anything. […]
If you’re interested in using the code, mail me. It’s very prototype, but available by anonymous FTP from info.cern.ch. It’s copyright CERN but free distribution and use is not normally a problem. […]
The WWW project was started to allow high energy physicists to share data, news, and documentation. We are very interested in spreading the web to other areas, and having gateway servers for other data. Collaborators welcome!
Tim Berners-Lee, “Re: Qualifiers on Hypertext links…,” posted to the newsgroup alt.hypertext, 6 August 1991.
That post, a reply to a question from an engineer at Digital Equipment, was the Web’s announcement to the world, and it went to a newsgroup for hypertext enthusiasts. The audience was small at first. The first server outside Europe went up at the Stanford Linear Accelerator Center on 12 December 1991, and volunteers wrote browsers for Unix machines: Erwise in Helsinki, Viola at O’Reilly, MIDAS at Stanford. The one that mattered came from the National Center for Supercomputing Applications at the University of Illinois, where Marc Andreessen and Eric Bina wrote Mosaic. Its first official release came on 21 April 1993, versions for the PC and the Macintosh followed that year, and because Mosaic showed pictures inside the page and installed without fuss, the Web became something people outside physics could use.
Nine days later CERN made the decision that settled the Web’s future. A statement dated 30 April 1993 put the WorldWideWeb software in the public domain, so that anyone could run a server, write a browser or build a business on the protocols without paying the laboratory or asking its leave. Robert Cailliau, the Belgian engineer who had co-written the formal project proposal with Berners-Lee in 1990, had spent some eighteen months pressing CERN’s directors to do it. A rival made his case for him. Gopher, a menu-driven system from the University of Minnesota, had spread quickly through universities since 1991, and in February 1993 the university announced licence fees for its server software. Developers worried the fees would spread, and many moved to the Web. By the end of 1993 there were more than 500 known web servers; by the end of 1994 there were 10,000, two thousand of them commercial, with some ten million users.
The design carried a compromise worth having a view about. Nelson wanted two-way links, so that a document always knew who pointed at it and no link could break. Berners-Lee’s links run one way. A page can point anywhere without asking, and the page it points at may vanish, which is why old corners of the Web are littered with dead links and why the error 404, Not Found, turned into a phrase everyone recognises. The breakage bought growth without coordination: nobody had to register a link, so nobody could refuse one. A system that guaranteed every link would have needed a central authority to keep the guarantee, and the Web would have grown at that authority’s pace.
Berners-Lee left CERN for the Massachusetts Institute of Technology in 1994 and founded the World Wide Web Consortium, which still writes the Web’s standards. He was knighted, and in 2017 he received the Turing Award, computing’s highest honour, for the browser, the protocols and the algorithms that let the Web scale. At the opening ceremony of the London Olympics in 2012 he sat at a computer in the middle of the stadium and typed four words that lit up the stands: “This is for everyone.” In September 2025 he used them as the title of his memoir. They describe the decision of April 1993 better than anything in the proposal of 1989.