Re: Is it okay to use xsd:anyURI as the range for a owl:DatatypeProperty ?

On Mon, 2010-07-19 at 10:49 +0100, Axel Polleres wrote:
> Sure you can use xsd:anyURI as datatypes, but be aware that these URIs are strings representing URIs rather than rdf:resources.
> to make the differnce clear:
> 
> e.g. foaf:homepage data uses normally rdf:resourses, 
> 
>  :me foaf:homepage <http://polleres.net>
> which is something different from
>  :me foaf:homepage "http://polleres.net"^^xsd:anyURI

True; I think this is why Martin wants to use it.   There are a few
times people should use, because they want to talk about terms (rather
than the things identified by those terms) when they do not.  For
example, if you're talking about the status of terms in a vocabulary,
saying how stable they are, when some might well be sameAs each other,
you need to use strings (or anyURIs) instead of resources.

Anyway, on this subject, I'll point out that xsd:anyURI is not only on
the standard list of datatypes for OWL 2, it is also in the reduced
lists for each of the three profiles (QL, EL, and RL):
        http://www.w3.org/TR/owl2-profiles/#Entities
        http://www.w3.org/TR/owl2-profiles/#Entities_2
        http://www.w3.org/TR/owl2-profiles/#Entities_3
        
Finally, if you need to express a connection between a resource and a
string which identifies that resource, you can use the RIF, with the
iri-string predicate:
        http://www.w3.org/TR/rif-dtb/#pred:iri-string
        
(Axel happens to be editor of that spec.)   Also see func:iri-to-uri if
you need to actually use a URI.  RDF/OWL/RIF/SPARQL/etc normally use
IRIs, even when we say URI.

   -- Sandro



> HTH,
> Axel
> 
> On 19 Jul 2010, at 10:41, Martin Hepp wrote:
> 
> > Dear all:
> > 
> > Is it okay to use xsd:anyURI as the range for a owl:DatatypeProperty ?
> > 
> > Sometimes, it could be handy to define a property that can hold URIs 
> > that may either be conceptual elements (owl:Class etc.) or data 
> > instances. Using annotation properties is no option, since I want to 
> > make the property rdfs:subProperty of a more generic property.
> > 
> > So is the following okay?
> > 
> > foo:myProperty a owl:DatatypeProperty ;
> >         rdfs:subPropertyOf foo:anotherProperty ;
> >         rdfs:domain gr:ProductOrService ;
> >         rdfs:range xsd:anyURI .
> > 
> > I explicitly want to avoid standard reasoning on URI literals attached 
> > via this property.
> > 
> > Best wishes
> > Martin
> > 
> > --------------------------------------------------------
> > martin hepp
> > e-business & web science research group
> > universitaet der bundeswehr muenchen
> > 
> > e-mail:  hepp@ebusiness-unibw.org
> > phone:   +49-(0)89-6004-4217
> > fax:     +49-(0)89-6004-4620
> > www:     http://www.unibw.de/ebusiness/ (group)
> >           http://www.heppnetz.de/ (personal)
> > skype:   mfhepp
> > twitter: mfhepp
> > 
> > Check out GoodRelations for E-Commerce on the Web of Linked Data!
> > =================================================================
> > * Project Main Page: http://purl.org/goodrelations/
> > * Quickstart Guide for Developers: http://bit.ly/quickstart4gr
> > * Vocabulary Reference: http://purl.org/goodrelations/v1
> > * Developer's Wiki: http://www.ebusiness-unibw.org/wiki/GoodRelations
> > * Examples: http://bit.ly/cookbook4gr
> > * Presentations: http://bit.ly/grtalks
> > * Videos: http://bit.ly/grvideos
> > 
> > 
> 
> 
>

Received on Monday, 19 July 2010 11:34:16 UTC