search

semanticarts / gist

Semantic Arts gist upper enterprise ontology
Creative Commons Attribution 4.0 International
153 stars 18 forks source link

Add range of `xsd:string` to `gist:description`? #1093

Open dylan-sa opened 3 months ago

dylan-sa commented 3 months ago

The datatype property gist:description does not currently have a range. xsd:string seems appropriate.

Jamie-SA commented 3 months ago

What about rdf:langString?

dylan-sa commented 3 months ago

rdf:langString makes sense to me since things could have descriptions in multiple languages. I don't think I've used it before, though--I wonder if it is generally supported by tooling/triplestores in the way xsd:string is?

rjyounes commented 2 months ago

I think we should hold off on rdf:langString until we make a deliberate effort to support multi-lingual or non-English implementations (an issue we have raised before and decided we are not ready for). There are plenty of other datatype properties that would need to be updated as well, including gist:name, gist:containedText, etc. I'm in favor of adding range xsd:string.

Jamie-SA commented 2 months ago

What is the purpose of setting a range on a datatype property?

uscholdm commented 2 months ago

What is the purpose of setting a range on a datatype property?

  • inference?
  • guidance?
  • other?

I don't think inference is a help here, but it is a signal of what the object should be.

dylan-sa commented 2 months ago

@Jamie-SA -- It's a good question. Guidance seems like the main thing, as Michael suggests.

There is some value in inference, too, but more from a validation perspective. For example, these triples would generate an inconsistency:

gist:sequence rdfs:range xsd:integer .
ex:_OrderedMember_1 gist:sequence "1"^^xsd:string .