Re: Are there any plans to develop an OWL version of QUDT?

On 5/9/2014 11:53, Simon Spero wrote:
> On Thu, May 8, 2014 at 1:31 AM, Holger Knublauch 
> <holger@topquadrant.com <mailto:holger@topquadrant.com>> wrote:
>
>     Simon,
>
>     Yes in OWL, but in RDF this is perfectly valid, see
>
>     http://www.w3.org/TR/2014/REC-rdf11-concepts-20140225/#dfn-recognized-datatype-iris
>
>     "Semantic extensions of RDF might choose to recognize other
>     datatype IRIs..."
>
>
> Right.  Except you define the data type using OWL. Which means you 
> defined the  lexical space to be empty.

 From an OWL point of view, yes. RDF does not depend on OWL though.

> What happens when you have a literal in RDF whose string is not in its 
> lexical space.
>
> I will grant you that there are many legal RDF statements in the 
> vocabularies. Can you tell me what's going on with these ones?  If you 
> think they're fine, and perfectly consistent, feel free to just let 
> this thread die.

I agree on your specific examples (including the dateUnion literal from 
the previous mails which should have been xsd:date). I guess these have 
crept in during the evolution of the qudt models, and Ralph will 
certainly get this fixed for the next release.

I was however responding to your statements that QUDT is not even valid 
RDF, and further remarks on RDF versus OWL with regards to tooling. Most 
RDF-based tools will have no issues with the problematic triples that 
you highlighted.

Thanks,
Holger


>     vaem:namespace
>
>     rdf:type owl:DatatypeProperty ;
>
>     rdfs:label "namespace"^^xsd:string ;
>
>     rdfs:range xsd:anyURI ;
>
>     .
>
>     <http://qudt.org/1.1/schema/qudt>
>
>       rdf:type owl:Ontology ;
>
>       vaem:namespace "http://qudt.org/schema/qudt"^^xsd:string ;
>
>     .
>