Concepts, Not Words: The Model Already Exists, And Where The Graphs Diverge Is The Finding

The source of the concept map: a dictionary crossed with a thesaurus.

Why it is here

The founder asked for a concept layer that works like a dictionary and a thesaurus at once, with several centres of gravity rather than one. This is where that idea comes from, and it goes further than the layer built on this site: a published standard already models exactly this (one concept, one preferred label per language, alternative labels, broader and narrower relations), and comparing the graphs induced by two locales is computable. Where they diverge is not an error to fix. It is the finding.

What it says

• The word being reached for is concept, distinct from term or label.
• A published standard already models it: one concept, preferred and alternative labels, broader and narrower relations.
• Meaning survives translation because it was never stored in a word.
• A bad translation revealed that the English word was probably wrong first, so the layer is quality control on the original.
• Divergence should be surfaced rather than resolved.

Worth re-reading for. It diagnoses a translation failure as an English failure. That inversion is the most useful idea in the set for anyone editing this book.

Where this document lands on the site

The document is the peak. Below it hang the concepts measured in its text and the places on this site that rest on it — and the two are different kinds of link, drawn differently: a measured edge is dashed and carries its count, an authored edge is solid. The method is stated in full on the hub.

Read it