Re: RDF and XML Schema types

Hi Manos,

It's important to note that anyURI is a lexical datatype
that defines a set of lexical representations which 
correspond to URIs. That does not mean that those lexical
representations, even if expressed in an RDF graph, are
the same as URIref nodes in the RDF graph which denote
resources.

So, in principle, there is no conflict between RDF and the
anyURI datatype. It simply depends on what you want to 
say in RDF.

If you want to specify a resource, use a URIref. E.g.

   <rdf:Description rdf:about="#something">
      <foo rdf:resource="http://foo.org/blargh"/>
   </rdf:Description>

If you want to specify a URI (the actual identifier string)
use a literal that is typed as anyURI. E.g.

   <rdf:Description rdf:about="#something">
      <bar rdf:datatype="&xsd;anyURI">http://foo.org/blargh</bar>
   </rdf:Description>

You might view the second case as a kind of quoting of
the URIref. Because it is quoted, it does not actually
denote a resource. Rather, the literal node denotes the URI
itself.

Does that help?

Cheers,

Patrick

[Patrick Stickler, Nokia/Finland, (+358 50) 483 9453, patrick.stickler@nokia.com]


----- Original Message ----- 
From: "ext m batsis" <mbatsis@netsmart.gr>
To: <www-rdf-interest@w3.org>
Sent: 23 September, 2002 16:54
Subject: RDF and XML Schema types


> 
> 
> Hi folks,
> 
> While building a 'simpleType' ontology for HumanMarkup (the base for a 
> tobe measurment toolkit), I encountered a possible conflict between the 
> RDF Schema and XML Schema class/type hierarchies.
> 
> I have been importing all subTypes of 
> http://www.w3.org/2001/XMLSchema-datatypes#anySimpleType as subclasses 
> of rdfs:Literal. My problem lies on the anyURI simple type, as it is 
> derived by restricting xs:anySimpleType, but should be under 
> rdfs:Resource per the RDF viewpoint.
> 
> So I'm looking for suggestions on how this should be handled. You can 
> find the post and the attached .rdf at [1]. Please direct additional 
> comments on this document at humanmarkup-comments [2].
> 
> [1] 
> http://lists.oasis-open.org/archives/humanmarkup-comment/200209/msg00073.html
> 
> 
> [2] humanmarkup-comment@lists.oasis-open.org
> 
> 
> Thanks,
> 
> Manos
>

Received on Monday, 23 September 2002 13:49:20 UTC