Re: Using DBpedia resources as skos:Concepts?

On 4 Nov 2009, at 08:37, Antoine Isaac <aisaac@few.vu.nl> wrote:

> Hi Richard,
>
> I've recently been caught in discussions on the fact that it was not  
> very good to have "universals" (in your case, concepts) mixed with  
> "particulars" (in your case, persons). But to tell the truth, in the  
> SKOS Recommendation there is no axioms stating that skos:Concept is  
> distinct from foaf:Person. So there is no contradiction. I'll let  
> you judge, maybe Dan and Pat can say more of it.

Sent fat-fingered from my iphone; excuse the lack of detail and  
citations...

My main concern with this idiom is that SKOS is designed for talking  
about the artifacts of library-like systems which rarely map onto tidy  
subclass hierarchies. However often these artifacts are 'clearly'  
individual things or category (class) terms. Since most SKOS  
properties are for talking about the characteristics of the artifact  
(admin-related dates, scope notes etc) not the 'thing in the world',  
and since many thesauri will overlap in scope especially around link- 
friendly entities like people and place, i see a tension.

An older rev of the SKOS spec had a nice example with Henry the 8th or  
similar, and urged a distinction between him and his lowercase-r- 
reification as a Concept in SKOS.

While this might seem longwinded, and in a sense all things described  
in rdf are done so from the perspective of some conceptualisation, I  
am keen that we make sure thesaurus etc admin metadata doesn't get  
garbled when two or more thesauri are overlaid in the same rdf graph.  
If we use direct equivalence too enthusiastically this seems a risk.

Cheers,

Dan

Received on Wednesday, 4 November 2009 08:16:42 UTC