What it is and the problem it solves
The World Wide Web is a global, decentralised information system for sharing content over the Internet. It solves document storage, update, discovery, and distribution problems in large, dynamic organisations like CERN. It was not designed for mass public use, but for collaborative science.
How it works
It connects hypertext to TCP and DNS. It uses HTTP for client–server communication, URLs for global addressing, and HTML for document structure. Links are unconstrained, cross-computer, and can point to mutable data or trigger programs. It runs on independent computers, supports simultaneous multi-user access, and was built to handle graphics, speech, and video—not just text.
What works
The first implementation worked end-to-end by December 1990: a browser-editor (WorldWideWeb), a server (CERN httpd), HTTP, URLs, and HTML. It supported cross-computer hyperlinks, simultaneous multi-user access, and multimedia. It ran on NeXTSTEP. It published the first website on 20 December 1990. It connected to the Internet via TCP and DNS as intended.
What does not
It did not solve discoverability at scale. It offered no native search, indexing, or ranking. It assumed users would manually follow links or navigate known addresses—no mechanism for finding unknown resources. It provided no access control, authentication, or persistence guarantees. Links broke when files moved or servers went offline.
What it changes
It changed how organisations share documents across administrative boundaries. It replaced centralised, hierarchical, or keyword-tagged systems with a decentralised, link-based model. It shifted responsibility for linking from system designers to authors. It enabled direct machine-to-machine data referencing—but only if both ends implemented the protocols.
Is it worth your time
Yes—if you work with information systems, interoperability, or decentralised publishing. It solved a concrete CERN document management problem by rejecting hierarchies and keyword tagging. Its royalty-free release in 1993 made it operable at scale. But its success depended on adoption choices outside Berners-Lee’s control: browsers, servers, and gateways had to be built and shared voluntarily.
