Re: [semanticweb] Ontology of interests

Hi again,

Am 25.07.2006 um 14:10 schrieb Danny Ayers:

> On 7/25/06, Knud Hinnerk Möller <knud.moeller@deri.org> wrote:
>
>> Yep, and that's why and additional foaf:topic statement seems to be
>> necessary.
>
> How much do you need to say about the topic object other than it
> exists?

Not much, as long as you give it a unique identifier (or use an  
existing one).

> Or maybe -
>
> _:person   foaf:interest _:document .
> _:document foaf:topic _:concept .
> _:concept rdf:type skos:Concept .

Ok, why not. But _concept should not be a blank node here, but  
identifiable in some way.

> er, remind me what the problem was again..?

The problem I see arises from using wikipedia URLs as identifiers for  
concepts:

* http://en.wikipedia.org/wiki/Resource_Description_Framework - does  
this identify the concept of RDF or the wikipedia page?
* in the foaf:interest example the wikipedia URL obviously identifies  
the page, as Damian pointed out, but then you'd still need to say  
what the foaf:topic of the page is, because if we have
    * Danny foaf:interest <http://en.wikipedia.org/wiki/ 
Resource_Description_Framework> .    and
    * Knud foaf:interest <http://www.w3.org/RDF/> .
* then how could anyone (without a human brain) know that we are both  
interested in RDF? This could only be inferred if we also say:
    * <http://en.wikipedia.org/wiki/Resource_Description_Framework>  
foaf:topic <nice.uri.for.RDF> .    and
    * <http://www.w3.org/RDF/> foaf:topic <nice.uri.for.RDF> .

I think that is the problem.

Cheers,
Knud
-------------------------------------------------
Knud Möller, MA
+353 - 91 - 495086
Digital Enterprise Research Institute
   National University of Ireland, Galway
Institiúid Taighde na Fiontraíochta Digití
   Ollscoil na hÉireann, Gaillimh