W3C home > Mailing lists > Public > www-webont-wg@w3.org > January 2003

Re: Review of Reference

From: Peter F. Patel-Schneider <pfps@research.bell-labs.com>
Date: Fri, 03 Jan 2003 10:15:11 -0500 (EST)
Message-Id: <20030103.101511.76119651.pfps@research.bell-labs.com>
To: heflin@cse.lehigh.edu
Cc: hendler@cs.umd.edu, www-webont-wg@w3.org

From: Jeff Heflin <heflin@cse.lehigh.edu>
Subject: Re: Review of Reference
Date: Fri, 03 Jan 2003 09:55:55 -0500

> 
> Jim Hendler wrote:
> > 
> > >
> > >- Property elements, last para. "A property is a binary relation that
> > >may or may not be defined in the ontology." Did the WG ever discuss
> > >whether or not it was legal to reference a class or property that wasn't
> > >defined in the ontology? I certainly would not support this. It makes it
> > >even easier for errors to creep into ontologies.
> > 
> > I must be missing something, we had long discussion about reference
> > to things outside current ontology during the imports debate in
> > particular.  I can certainly say in document A that something is a
> > B:foo (foo a class) where B is a different document and no imports is
> > specified.  Do you mean something different by "not defined in the
> > ontology"?
> 
> My apologies; in my haste to make the review deadline, I was a little
> sloppy in writing my comments. The referenced paragraph seems to me to
> be saying that the property may not be defined in ANY ontology (because
> it then says if it isn't defined, it is assumed to have no constraints).
> This is what I was wondering if the group discussed, that is, can a
> legal OWL document reference B:foo when B does not have a class or
> property named foo? In any case, the paragraph needs to be reworded to
> make it clear what it means. It does not take into account that
> properties may defined in other ontologies, so it is not clear when the
> property should be assumed to have no constraints.

Yes, certainly.  In OWL Full, as in (my view of) RDF, any URI reference is
fair game.  The following situation is completely acceptable in OWL/Full.
(I'm using triples so as to not get into any problems with RDF/XML.

http://x.ex/foo
	<http://x.ex/foo#bar> rdfs:comment "This is the only allowable URI reference in this name space." .

http://y.ex/w
	<http://y.ex/w#x> <http://x.ex/foo#bbb> <http://y.ex/w#y> .


This is not, however, correct OWL/DL.  To make it so would require adding
something like
	<http://y.ex/w#x> rdf:type owl:Thing .
	<http://y.ex/w#y> rdf:type owl:Thing .
	<http://x.ex/foo#bbb> rdf:type owl:ObjectProperty .
to
	http://y.ex/w

peter
Received on Friday, 3 January 2003 10:15:21 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:57:56 GMT