Re: [Proposal][SKOS-Core] skos:denotes

* Dave Reynolds <der@hplb.hpl.hp.com> [2004-09-29 16:13+0100]
> 
> The RDF resource of type foaf:Person is already a concept or at least an 
> abstract thing, it's not the real Al. So you have a concept denoting an 
> abstract thing. Sorry but I don't follow this. This is, no doubt, due to my 
> lack of sufficient grounding in philosophy but I think you'll need more 
> explaination if us poor technical types are supposed to be able to grok it.

SKOS is one layer of abstraction further removed from reality than 
basic RDF, the proposed skos:denotes (or 'conceptualizes', which I
originally called it) is intended as a way of bridging these two
representational styles. RDF/OWL itself is good at saying things about
individuals and classes of individual, but it is also quite good 
(eg. SKOS) at carrying other representational formats. So the goal here 
is to find ways to avoid these different options breaking the 'network
effect' of distributed, merge-able RDF data.

If several parties use SKOS and they have a concept in their
SKOS-expressed thesauri that stand for the person Alistair Miles, while
other parties simply write RDF statements about Alistair directly, we'd
imho benefit if we had some conventions for figuring out that they were 
talking about the same thing. But they're not "the same thing" in the
conventional RDF/OWL sense, since the class Person and the class
skos:Concept are presumably disjoint. I agree that the RDF layer itself
is an abstraction too; people aren't actually inside the RDF graph itself. 
But we don't conventionally labour that point when choosing RDF property names. 
Eg. we write xyz:age, xyz:weight rather than
xyz:age_encoded_in_a_string_of_a_person_represented_in_an_rdf_graph.

Maybe skos:represents would work better as a name?

Dan

Received on Wednesday, 29 September 2004 15:33:50 UTC