Ancient Web: Tim Berners-Lee’s 1989 Proposal Was About Losing Track of Work

Part of the Ancient Web Collection.

The document that helped launch the World Wide Web does not begin with plans for online shopping, viral videos, or a giant public encyclopedia. It begins with a complaint: people at CERN keep losing track of information as projects evolve.

Read Tim Berners-Lee's original proposal.

Dated March 1989 and May 1990, Information Management: A Proposal discusses research on particle accelerators and the increasing difficulty of following related work across different computers, teams, and documents.

Berners-Lee proposes a distributed hypertext system where a piece of information can point to related information without insisting that everything live in one central database. He compares conventional hierarchical filing with more flexible networks of relationships among people, programs, experiments, and documents.

Before it was called the Web

The introductory note on the HTML transcription says the idea was originally called Mesh. The name World Wide Web came later, as Berners-Lee wrote the first software in 1990.

The page also says it was manually converted from an original MacWord document and preserved without correcting errors, including misnumbered figures and unfinished references. Alternative versions are linked, including RTF and an HTML file generated in Microsoft Word in 1998.

That preservation note is almost as important as the proposal. It tells readers where this web page came from and what they are actually looking at: not the original word-processing file, but a later readable presentation of its text and diagrams.

The proposal's details are wonderfully modest. It argues that a useful first stage could be built on one machine with existing data, then extended across multiple machines. It mentions object-oriented programming techniques, compatibility with many kinds of computer systems, and the benefit of linked indexes.

The 1989 document did not yet describe the final Web as people know it today. This is an argument to CERN management about a practical problem, not a retrospective summary of an inevitable revolution.

A separate queued 1991 Usenet message documents Berners-Lee describing a working system to potential users. That later announcement has its own historical purpose; this page records the problem and proposed architecture before the public rollout.

Explore the original annotated proposal.

Part of the Ancient Web Collection.