Tim Berners-Lee’s Boss Wrote “Vague, but Exciting.” The Web Started There.

In March 1989, Tim Berners-Lee handed his boss at CERN a proposal for managing information.

Mike Sendall wrote three words on it:

“Vague, but exciting.”

That was not approval for a global information revolution.

It was permission to keep experimenting.

From that opening, Berners-Lee built the first web server, the first browser-editor, and the basic technologies that allowed documents on different computers to link to one another across the Internet.

The problem was not “invent the Web”

CERN was a huge scientific organization. Researchers came from universities and laboratories around the world, used different computers, and produced information in many incompatible systems. The practical problem was simple to describe and hard to solve: how could people find and connect information without first learning where it lived or which software stored it?

Berners-Lee had already experimented with linked information in 1980 through a program called ENQUIRE. In 1989 he proposed combining hypertext with networked computers so that documents could refer to one another through links.

The proposal was not a detailed blueprint for today’s Web. That is why “vague” was a fair description.

But the direction was useful enough to test.

Then he built the smallest system that could prove the idea

By the end of 1990, Berners-Lee had a working web server and a browser-editor running on a NeXT computer at CERN. He also developed the foundations of HTML, HTTP and the addressing system that became URLs and URIs.

The first website explained the Web project itself. It told people what hypertext was, how to create a server, and how to find other servers as they appeared.

This mattered because the idea did not remain a slide deck.

It became a working loop:

A document could have an address.
A browser could request it.
The document could contain links to other addresses.
Other people could run servers of their own.
The system did not require one central database to contain everything.

Once those pieces worked together, other people could extend the network without asking Berners-Lee to redesign the whole system for them.

The Web grew because the architecture left room for other people

Many technologies before the Web supported hypertext or online information. Berners-Lee’s breakthrough was not the invention of every individual ingredient. It was a practical combination of networking, hypertext and simple common protocols that could cross organizational and machine boundaries.

CERN released Web software to researchers, and in April 1993 made the World Wide Web technology freely usable. That decision removed a major barrier to adoption.

The system could spread because nobody needed to buy permission from a central owner to create a website, write a browser, or run a server.

Later browsers made the Web far easier to use, and thousands of engineers and organizations transformed it into something much larger than the first CERN implementation.

But the original architecture had already created the crucial property: participation could be decentralized.

What Berners-Lee actually did well

He started with a real local problem. Scientists could not easily connect information across incompatible systems.
He reused existing ideas. Hypertext and the Internet already existed; he combined them instead of pretending every component had to be new.
He built a working prototype. Server, browser-editor and protocol evolved together.
He kept the core simple. The basic Web did not need to predict every future use before it could be useful.
He designed for other people to join. New servers and pages could appear without central coordination.
He supported open standards. The Web’s growth depended on interoperability rather than a single proprietary product.

A vague proposal can be useful if the next experiment is concrete

“Vague, but exciting” is often retold as a charming origin story. The more useful lesson is what happened after the comment.

Berners-Lee did not spend years trying to make the proposal sound inevitable.

He made the next version more concrete.

Then he built it.

Then other people could try it.

The early Web was incomplete, limited and small. That did not make it a failure. It made it testable.

When an idea is large enough to be vague, the next move is often not another hundred pages of prediction.

It is a small system that makes the idea real enough for other people to use.

Explore

Stay connected