The Internet Was Born in a Cube
35 years ago today, one guy at CERN uploaded a file. Almost nobody noticed.
If you were trying to design the least viral launch in history, this would be it.
No demo day. No investors. No press embargo. No product hunt page. Just a quiet post on a nerdy newsgroup explaining how to download a program that let you click words.
That program became the interface to modern life.
On February 26, 1991, Tim Berners-Lee posted a message to the alt.hypertext newsgroup. The subject line read: "WorldWideWeb: Summary."
The message explained how to download something he'd built: a program that let you click on highlighted words and jump to other documents. It worked on a NeXT computer, the sleek black cube Steve Jobs had designed during his exile from Apple. Berners-Lee had finished the first working version on Christmas Day, 1990, coding alone at CERN while everyone else was on holiday.
Almost nobody replied.
This is how the World Wide Web entered the world. Not with a press release. Not with a launch party. With a single message to a mailing list of hypertext enthusiasts, most of whom probably assumed it was just another academic curiosity.
Thirty-five years later, 5.5 billion people use what he built.
The Problem Was Boring
The origin story of the web is disappointingly mundane.
CERN in the late 1980s was a mess of incompatible systems. Physicists used VAX machines. Administrators used IBM mainframes. Everyone had their own file formats, their own documentation standards, their own silos. When researchers wanted to share information, they printed it out and walked it across the building.
Berners-Lee, a British contractor with a degree from Oxford and a reputation for building useful tools, saw a simple pattern: everything was connected in his head, but nothing was connected in the machines. He'd been obsessed with this problem since 1980, when he built a program called ENQUIRE that let him link notes together. It was named after a Victorian domestic encyclopedia called Enquire Within Upon Everything.
The idea was almost too obvious: what if documents could contain links to other documents? What if clicking a word took you somewhere else? What if "somewhere else" could be anywhere in the world?
In March 1989, he wrote a proposal titled "Information Management: A Proposal." His boss, Mike Sendall, scribbled a now-famous note on the cover page: "Vague but exciting."
It was the most consequential bureaucratic approval in history.
The Machine That Made It Possible
The NeXT computer was essential to what happened next.
Steve Jobs had been forced out of Apple in 1985 and founded NeXT to build the workstation of the future. The result was a perfect black cube: expensive ($6,500, equivalent to roughly $16,000 today), underpowered for its price, and a commercial disaster. NeXT sold only 50,000 units total.
But the software was extraordinary. NeXTSTEP included tools that let developers build complex applications in weeks instead of months. It had a built-in text system that handled hyperlinks natively. It was, in retrospect, exactly the development environment you'd need if you wanted to invent the web quickly and alone.
Berners-Lee later said he couldn't have built the first browser so fast on anything else. "I was lucky that CERN had some NeXT machines," he told an interviewer. The hardware was a commercial failure, but it became the launchpad for the most transformative software ever written.
There's something poetic here: the web was born on a machine built by a man in exile, designed for a future that hadn't arrived yet.
The Quiet Launch
The first web browser did everything. It let you read pages, sure, but it also let you write them. The same program was browser, editor, and server. Berners-Lee imagined a web where everyone contributed, where the line between reader and author barely existed.
That's not how it turned out.
When he posted to alt.hypertext on February 26, 1991, the message was clinical, almost apologetic:
"The WWW project merges the techniques of networked information and hypertext to make an easy but powerful global information system."
He explained the protocols (HTTP), the addressing scheme (URLs), the markup language (HTML). He noted that the software only ran on NeXT machines, which limited the audience to a few thousand people worldwide.
The response was muted. A few academics asked technical questions. Some pointed out that other hypertext systems already existed (HyperCard, Gopher, WAIS). The thread died after a handful of replies.
For the next two years, the web grew slowly. By the end of 1992, there were only 26 websites in the world. Most were at universities and research labs. The general public had no idea it existed.
The Decision That Changed Everything
In April 1993, CERN made an announcement that barely registered at the time but reshaped human civilization: they released the web's underlying code into the public domain.
No licensing fees. No royalties. No restrictions. Anyone could use it. Anyone could build on it.
This was not inevitable.
CERN could have patented the technology.
They could have licensed it to a single company.
Berners-Lee could have founded a startup and become the richest person in history.
Imagine paying a licensing fee to publish a website. Imagine Apple owning HTTP. Imagine Microsoft controlling URLs. The modern internet would look less like a public square and more like cable television.
(One estimate suggests the web would be worth over $30 trillion today if it had been monetized.)
Instead, they gave it away.
"Had CERN or I, for example, licensed the Web's technology and made a lot of money, I wouldn't have been any happier," Berners-Lee wrote later. "There was no business model. The value of the web came from its universality."
Three months after the announcement, Marc Andreessen and Eric Bina released Mosaic, the first graphical browser that ran on Windows and Mac. Within a year, traffic on the web exploded 341,634%. Andreessen went on to co-found Netscape, kickstart the dot-com boom, and become a venture capitalist worth $1.7 billion.
Berners-Lee became a professor.
The Fork in the Road
The web could have been proprietary. It could have been owned. It could have been controlled.
Every website, every app, every meme, every DAO, every protocol, every creator economy, every streaming service, every social network -- all of it traces back to a decision made in 1993 to give something away for free.
Now we're at another inflection point. The debates around web3, decentralization, and protocol ownership are fundamentally about this question: should the next layer of the internet be open, or owned? Should value accrue to protocols or to platforms? Should users control their data, or should they trade it for access?
Berners-Lee has opinions. He's spent the last decade working on Solid, a project designed to give people control over their own data. He's been critical of surveillance capitalism, platform monopolies, and the erosion of the web's original ideals. In 2019, he published a "Contract for the Web" calling on governments, companies, and citizens to protect online rights.
"I've become increasingly worried that the web I invented has been corrupted," he wrote in The Guardian. "The gap between online and offline life has been bridged in troubling ways."
The guy who gave away the most valuable technology in human history now spends his time trying to fix what others built on top of it.
The Sticker on the Machine
The original NeXT cube Tim Berners-Lee used to build the web now sits in the Science Museum in London. It's displayed in a glass case, next to a handwritten sign that was taped to it during its working years.
The sign reads: "This machine is a server. DO NOT POWER IT DOWN!!"
The exclamation points are original. Someone at CERN, probably Berners-Lee himself, was worried that a colleague might unplug it without realizing what it was doing.
That cube, with its desperate sticker, is the most valuable piece of computing hardware in history. Not because of what it was built to do, but because of what someone decided to do with it.
Thirty-five years ago today, one guy at CERN uploaded a file to a newsgroup.
He wasn't trying to build a trillion-dollar industry. He was trying to make it easier for physicists to share documents.
He invented the future almost by accident -- and then he refused to own it.
Almost nobody noticed.
Everyone lives inside it now.
Sources
Primary
- Berners-Lee, Tim. "Information Management: A Proposal." CERN, March 1989.
- Berners-Lee, Tim. "WorldWideWeb: Summary." alt.hypertext newsgroup, February 26, 1991.
- Berners-Lee, Tim. "Answers for Young People." W3C.
- Gillies, James, and Robert Cailliau. How the Web Was Born. Oxford University Press, 2000.
- CERN. "A Short History of the Web."
- Internet World Stats. "World Internet Users Statistics." February 2026.
- W3Techs. "Historical Yearly Trends in Web Technology Usage." 2026.
