W3C home > Mailing lists > Public > www-webont-wg@w3.org > April 2002

Re: Dark triples motivation

From: Jonathan Borden <jonathan@openhealth.org>
Date: Wed, 17 Apr 2002 08:45:21 -0400
Message-ID: <3CBD6E61.7E972B1B@openhealth.org>
To: Jeremy Carroll <jjc@hplb.hpl.hp.com>
CC: www-webont-wg@w3.org
Jeremy Carroll wrote:

[snip]
.
> 
> Example:
> 
> The RDF document:
> 
> <rdf:Description rdf:about="#John">
>    <rdf:type rdf:resource="#Student">
>    <rdf:type rdf:resource="#Employee">
> </rdf:Description>
> 
> understood with rdf:type denoting set membership, would, under almost all
> set theories, entail #John being a member of the intersection of #Student
> and #Employee.
> In daml+oil this is said:
> 
> <rdf:Description rdf:about="#John">
>    <rdf:type>
>     <daml:class>
>       <daml:intersectionOf rdf:parseType="daml:collection">
>          <daml:class rdf:ID="Employee"/>
>          <daml:class rdf:ID="Student"/>
>      <daml:intersectionOf>
>     </daml:class>
>    </rdf:type>
> </rdf:Description>
> 
> If we leave the daml class expression:
> 
> <rdf:RDF>
>     <daml:class>
>       <daml:intersectionOf rdf:parseType="daml:collection">
>          <daml:class rdf:ID="Employee"/>
>          <daml:class rdf:ID="Student"/>
>      <daml:intersectionOf>
>     </daml:class>
> </rdf:RDF>
> 
> as unasserted then OWL can assign a meaning to this expression wholly
> independently of RDF.
> 
> Have I got it?

How does this particular example demonstrate a problem? Why should this
convince DanC? One of PeterPS's examples used a paradoxical class i.e. a
member of itself. Is there a similar class that does not depend on the
use of the 'Q's? e.g. would something like:

<ont:Class rdf:ID="foo">
	<ont:unionOf rdf:parseType="ont:collection">
		<ont:Class rdf:about="#foo"/>
		<ont:Class rdf:about="#bar"/>
	</ont:unionOf>
</ont:Class>

be a problem? or

<ont:Class rdf:ID="foo">
	<ont:disjointUnionOf rdf:parseType="ont:collection">
		<ont:Class rdf:about="#foo"/>
		<ont:Class rdf:about="#bar"/>
	</ont:disjointUnionOf>
</ont:Class>

I'm just guessing, but the example should demonstrate a paradox - right?

Jonathan
Received on Wednesday, 17 April 2002 08:50:09 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:57:49 GMT