The issue: during enrichment--converting subject headings to FAST or adding additional FAST headings found through Classify--some terms may match different FAST terms in different FAST schemes with the same label (ex. the term "Manuscripts" found in both FAST FormGenre and Topical schemes). When minting this term as a local skos:concept under our own /terms namespace, how important is it to map this to one or the other terms instead of both? Should we have a single local term that does one or more of the following?:
1) belongs to multiple local schemes (ex. nyplterms:Topical, nyplterms:Form)
2) belongs to multiple FAST schemes
3) maps to both FAST terms
4) is typed as a subclass of skos:concept (ex. nypl:Topic, nypl:Form)
Questions to consider:
What other local schemes will we want to create?
For a given resource instance, will we ever need to know which controlled vocabulary an assigned term was originally drawn from, or do we consider all registry term assignments as enrichments?
How is prefLabel determined? Do we chose a priority vocabulary or do we apply our own capitalization and pluralization rules?
Other questions?
We decided to do 1) and 3) and add vocab source to altLabels.
todo:
define namespace and terms schemes and document (@saverkamp)
Replicates nypl-registry/serialization-terms#2
The issue: during enrichment--converting subject headings to FAST or adding additional FAST headings found through Classify--some terms may match different FAST terms in different FAST schemes with the same label (ex. the term "Manuscripts" found in both FAST FormGenre and Topical schemes). When minting this term as a local skos:concept under our own /terms namespace, how important is it to map this to one or the other terms instead of both? Should we have a single local term that does one or more of the following?: 1) belongs to multiple local schemes (ex. nyplterms:Topical, nyplterms:Form) 2) belongs to multiple FAST schemes 3) maps to both FAST terms 4) is typed as a subclass of skos:concept (ex. nypl:Topic, nypl:Form)
Questions to consider:
We decided to do 1) and 3) and add vocab source to altLabels.
todo: