Documentation movement
The European effort, from the 1890s onward, to organize the whole of recorded knowledge into one retrievable system — the theoretical foundation modern information-science was built on information-science-wikipedia.
What happened
- 1895 — paul-otlet and Henri La Fontaine found the International Institute of Bibliography in Brussels. La Fontaine won the Nobel Peace Prize in 1913.
- Otlet designs the Universal Decimal Classification, an extension of Dewey that adds operators for combining subjects, so a document can be placed by a composed expression rather than a single bucket.
- The project's ambition was a card catalogue of every fact in every published work, queryable remotely. Otlet described links between documents, search, and retrieval at a distance long before the machinery existed to do any of it.
- Post-war, Suzanne Briet carries the tradition into its second generation.
- 1937 — the American Documentation Institute gives the field a US institutional home; it renames itself in 1968, by which point the word in use is information, not documentation (see asist).
Why it matters to this wiki
The documentation movement is the part of the corpus's ancestry that the corpus does not have. as-we-may-think (1945) is treated here as the founding document of mechanized personal knowledge, and Bush's associative-trails are read as the origin of linking as an alternative to hierarchy. Otlet was building a linked, remotely queryable knowledge base fifty years earlier, and from the opposite direction: not one thinker's private supplement to memory, but a shared public index maintained by an institution.
The two ancestries make different bets, and the difference is live in this corpus:
| Documentation movement | Memex lineage | |
|---|---|---|
| Scope | universal, public, institutional | personal, private, individual |
| Structure | a designed classification imposed in advance | associative links made by the user as they go |
| Maintained by | trained staff | the owner (and now the LLM) |
llm-wiki and gbrain sit oddly between them: personal in scope like the memex, schema-governed like a classification, and maintained by neither a librarian nor the owner. The corpus's recurring problem — who decides that two nodes are the same thing (see entity-resolution, orkg-predicate-redundancy) — is the classification problem the documentation movement had staff for.
Related
information-science · paul-otlet · asist · memex · as-we-may-think · ontology · tools-for-thought