Closed andrea-perego closed 3 years ago
There was also a discussion around the range for dcat:theme
at: https://github.com/w3c/dxwg/issues/1153
I would be comfortable relaxing the ranges of theme and themeTaxonomy. But it is important to keep them as object-properties, since the object must be denoted by a URI - i.e. taken from a managed source somewhere.
All, there was an earlier discussion on this (see here), where I noted that dcat:theme
was explicitly minted for DCAT version 1 because it was felt that the range of dct:subject
was too broad and that it would be useful to have a property with a range restricted to skos:Concept
. Otherwise, dct:subject
could have been used.
Now, with the relaxation of the range of dct:subject
, which now also allows for textual values, I guess that dcat:theme
with relaxed range, but still requiring it to be an object property, can still be seen as a restriction on dct:subject
, effectively making it the same as the 'old' (pre-relaxation) dct:subject
. Right?
@dr-shorthair yes, it could work for our use cases if dcat:theme
is an object property
@makxdekkers I see your point. As I mentioned early in biomedical domain we have more owl
based ontologies and vocabularies. So if you still like to restrict range
of dcat:theme
property it would be nice to have list of values like skos:Concept
, owl:Class
.
@dr-shorthair
Why not relax range constraint for
dcat:theme
? In the biomedical domain we have more ontologies and often these ontologies haveowl:Class
typing for concepts.Originally posted by @rajaram5 in https://github.com/w3c/dxwg/issues/123#issuecomment-843290546