W3C home > Mailing lists > Public > public-owl-dev@w3.org > July to September 2006

Re: inconsistency in OWL docs re: annotation properties

From: David Martin <martin@AI.SRI.COM>
Date: Fri, 11 Aug 2006 16:59:55 -0700
Message-ID: <44DD19FB.7040400@ai.sri.com>
To: Bijan Parsia <bparsia@isr.umd.edu>
CC: public-owl-dev@w3.org

Thanks, Bijan; this is very helpful.

I should add that my question was based on a careless reading.  When I 
read that "The object of an annotation property must be either a data 
literal, a URI reference, or an individual" I was thinking that "URI 
reference" was talking about the anyURI datatype; that is, it had to be 
something like this:

<a:myAnnotationProperty 
rdf:datatype="&xsd;#anyURI">http://...</a:myAnnotationProperty>

Cheers,
David


Bijan Parsia wrote:
> On Aug 11, 2006, at 8:18 PM, David Martin wrote:
> 
>> In (1) it says an annotation property cannot have a class as a value; 
>> in (2) it says it can.  Could someone please clarify this?
> 
> It's not obvious. Semantically, it's harmless, since annotations are 
> just thrown away  (thus available to some other sort of tool).
> 
> Let's consider a simple example
> 
> :sheevah rdf:type owl:Thing.
> :sheevahClass rdf:type owl:Class.
> 
> :sheevah rdfs:seeAlso :sheevahClass.
> 
> 
> So, let's consider the reverse transformation to triples:
> 
> Namespace(rdf    = <http://www.w3.org/1999/02/22-rdf-syntax-ns#>)
> Namespace(xsd    = <http://www.w3.org/2001/XMLSchema#>)
> Namespace(rdfs    = <http://www.w3.org/2000/01/rdf-schema#>)
> Namespace(owl    = <http://www.w3.org/2002/07/owl#>)
> Namespace(a    = 
> <http://www.mindswap.org/dav/ontologies/bijan/2006/annotationProExample#>)
> 
> Ontology( 
> <http://www.mindswap.org/dav/ontologies/bijan/2006/annotationProExample>
> 
>  Class(a:sheevahClass partial)
>  Class(owl:Thing partial)
> 
>  AnnotationProperty(rdfs:seeAlso)
> 
>  Individual(a:sheevah
>   annotation(rdfs:seeAlso 
> http://www.mindswap.org/dav/ontologies/bijan/2006/annotationProExample#sheevahClass) 
> 
> 
>   type(owl:Thing)))
> 
> So, what would make this owl full? No class is treated as an individual 
> (that is, has at least one rdf:type triple with owl:Class and one with a 
> "regular" class). And the range of the seeAlso is a URI reference (which 
> happnes to name a class).
> 
> Both Pellet and the Wonderweb species validator pass this as OWL Lite. I 
> have an RDF/XML version at:
>     http://www.mindswap.org/dav/ontologies/bijan/2006/annotationPropExample
> 
> Pellet's validation:
>     <http://www.mindswap.org/cgi-bin/2003/pellet/pelletGet.cgi?inputFile=http%3A%2F%2Fwww.mindswap.org%2Fdav%2Fontologies%2Fbijan%2F2006%2FannotationPropExample&inputFormat=RDF%2FXML&inputString=&Species=true&classifyFormat=NONE&queryFile=&queryFormat=SPARQL&queryString=> 
> 
> 
> So the answer is that they are both right :)
> 
> Cheers,
> Bijan.
Received on Saturday, 12 August 2006 00:07:29 GMT

This archive was generated by hypermail 2.3.1 : Wednesday, 27 March 2013 09:32:53 GMT