- From: Vladimir Alexiev via GitHub <noreply@w3.org>
- Date: Mon, 20 Oct 2025 23:00:51 +0000
- To: public-shacl@w3.org
VladimirAlexiev has just created a new issue for https://github.com/w3c/data-shapes: == Use case: SKOS profiles == Thesauri start simple but may grow complex. Consider the Getty Vocabularies, Nomenclature for museum cataloging , EuroVoc. Getty have: - different node types: Concepts but also Guide Terms and Facets . These are not used for indexing and are modeled as Thesaurus Arrays (as skos:Collection cannot be rooted under a concept) - different Broader varieties: generic, part-time, instantial (BTG, BTP, BTI) - SKOS-XL labels (nodes not mere literals) - Structured notes/definitions (nodes not mere literals). Unfortunately, SKOS doesn't say what class to use - provenance (sources and contributors) per concept, label, definition - sparse ordering (modeled with Collections) - change history, with some special events eg "changed parent from x to y" - for places (TGN) and agents (ULAN), dual nodes connected by foaf:focus and representing place/agent, with specific fields like bounding box or birth/death/bio. - Sophisticated language tags, with indication of scripts, transliteration method; and using custom subtags People often prefer simplicity so they may ask "give me plain SKOS", or any subset of the above. But that doesn't just mean "throw away triples". Often it means infer simpler representations, eg - the well-known prop paths to dumb-down XL labels to SKOS, eg `xl:prefLabel/xl:literalForm -> skos:prefLabel` - infer skos:broader from BTG, BTP, BTI (not all combinations are allowed) and skipping non-Concept nodes in the hierarchy - dumb-down lang tag to only the first part Such conversions are Please view or discuss this issue at https://github.com/w3c/data-shapes/issues/617 using your GitHub account -- Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config
Received on Monday, 20 October 2025 23:00:52 UTC