Summary
Berners-Lee’s memo starts from a concrete institutional problem: CERN was losing information as people rotated through, and neither hierarchical trees (CERNDOC, file systems) nor keyword search could model how the organisation actually worked — “a multiply connected web whose interconnections evolve with time.” His answer was a linked information system: nodes for people, projects, concepts, and documents, joined by typed links, generalizing his earlier Enquire program. The design requirements read as a founding charter — remote access across networks, heterogeneity across systems, private links over public information, gateways into existing data, and above all non-centralisation: systems must link together “without requiring any central control or coordination.” The proposed architecture separates display software from storage behind a defined interface — the client/server split the Web still runs on.
Why it matters
This is the Web’s point of origin, and its most striking property is that decentralization was a stated requirement, not an accident — no permission needed to publish or link. The Independent Internet argues that this founding property eroded through the Web 2.0 era and must be deliberately rebuilt; see decentralization and decentralization across the lineage. The proposal also anticipated machine-readable meaning: typed links (“depends on”, “made”, “refers to”) that software could analyze — an idea Berners-Lee returned to in the Semantic Web, and one the LLM-maintained wiki realizes in a new form: a linked knowledge system that grows and evolves with the organisation it describes.
Part of the AP0110.ORG source library. See the wiki overview.
Full source text: read the original document.