RE: Wiki page on Goals

> > Specifying creatorFoo (e.g. creatorLabel) as rdfs:subPropertyOf
> > rdfs:label (or skos:prefLabel) should make it clear this property is
> a
> > label for the creator rather than the creator itself.

I retract the rdfs:subPropertyOf rdfs:label/skos:prefLabel suggestion.
It doesn't align:

http://www.w3.org/TR/rdf-schema/#ch_label

I agree your "shortcut property" rationalization makes more sense as a
justification for an owl:DatatypeProperty. I think it's an important use
case because there are many times where someone knows the name of
something but can't coin a de-referenceable URI.

> > ex:shakespeare an x-dcterms:Agent ;
> > 	rdfs:label "William Shakespeare" . # skos:prefLabel should be
> > even better
> 
> Or foaf:name?

foaf:name and skos:pref/altLabels are all rdfs:subClassOf rdfs:label, so
sure. Presumably this is the common denominator that would correspond to
the creatorLabel datatype property.

> > You could recommend skos:inScheme on the ex:shakespeare resource to
> > indicate "authority control". This suggests it might be tempting to
> use
> > skos:Concept as the range for all ObjectProperties to minimize the
> > ontological commitment. This would allow others to use foaf:focus to
> > model reality as they see fit.
> 
> Or one could remain silent on the range and allow skos:Concept
> (harmlessly) to be inferred from the use of foaf:focus.

It worries me that foaf:focus doesn't have an owl:inverseOf to point in
that direction. This will make inferencing harder in mixed datasets
where some sources reference a modeled thing and others reference a
corresponding concept.

Jeff

> Thanks for the suggestion!  I'd be interested to hear if others
> on this list would also find such a property to be useful.
> 
> Tom
> 
> --
> Thomas Baker <tbaker@tbaker.de>
>

Received on Saturday, 4 December 2010 22:27:39 UTC