Sean said:
> That is perhaps true, but it doesn't necessarily have to be so. RDF
> Schema already has the "isDefinedBy" property as a means of linking a
> term to a schema that defines it (although I would prefer
> "isDefinedBy" to have sub properties defined such that some point to
> prose descriptions, and others point to serialized RDF). It is only a
> small step from there to define a property that relates a set of terms
> to each other in a "vocabulary". In fact, you could do this using
> rdf:type:-
>
> <http://purl.org/dc/elements/1.1/title> rdf:type :DC11Term .
> <http://purl.org/dc/elements/1.1/creator> rdf:type :DC11Term .
>
> There are probably other (better) ways of doing it. Just defining a
> "belongsToVocab" property is the initial solution that comes to mind.
> I think that the SWAG dictionary [1] uses the "xmlns-primary-prefix"
> term with a similar utility.
Yes! This is _exactly_ the sort of thing I was trying to get at. Thanks,
Sean, for making it clear!
I just wasn't sure whether it was strictly _necessary_ to make these extra
statements if it was permissible to infer namespace "affiliation" from the
URI of a property, but I think you're reinforcing my initial feeling that it
might be a good idea. And it becomes essential if we want to talk about
aggregates which do not correspond to the set of terms in one namespace
(which was my second question).
Thanks
Pete
|