RE: Top three levels of Dewey Decimal Classification published as linked data

Hi Bernard, 

> > It is true that http://dewey.info/class/641/ is never mentioned in 
> > subject position, but it is in object position several times. So an 
> > agent might deduce the following about http://dewey.info/class/641/ 
> > from the response:
> >
> > <http://dewey.info/class/641/>
> >     dct:hasVersion <http://dewey.info/class/641/2009/08/about.en> ;
> >     dct:hasVersion <http://dewey.info/class/641/2009/08/about.de> ; 
> > ...
> >     dct:hasVersion <http://dewey.info/class/641/2009/08/about.es> .
> >   
> Indeed. But you can't infer from this the type of 
> <http://dewey.info/class/641/> AFAIK Unless some extension of 
> dcterms + SKOS would include some rule like
> (?y  a skos:Concept  AND  ?x   dct:hasVersion  ?y) => (?x  a 
> skos:Concept)

Agreed.

> > And this coincides with what we get if we use a SPARQL 
> query directly:
> >   
> Well, not exactly. How the SPARQL query retrieves the 
> following triple is a mystery to me.
> It figures I don't know which query is hiding behind DESCRIBE ...
> 
> <http://dewey.info/class/641/> rdf:type skos:Concept 
> 
>

SPARQL retrieves these triples because I cheated. They are in the store!
;-) I asserted that http://dewey.info/class/641/ is a skos:Concept. I
probably shouldn't have, because (for me) the question seems to be still
open whether it is a skos:Concept. But if not a skos:Concept, what is
it? One idea could be to make use of the FRBR/RDA vocabularies to model
the resource type (frbr:Work) and relationships (frbr:realization,
frbr:embodiment) of this frbr:Work to skos:Concepts.
 
Cheers
Michael

Received on Friday, 21 August 2009 21:28:20 UTC