Re: Proposed reply for O rdf:type owl:Ontology optional?

Bijan actually had three comment messages, so here is a combined response.

peter




Thank you for your comments.  As they were received outside of the Last
Call review period they have not received the prompt attention they should
have.

On 18 September 2003 the WebOnt Working Group decided (see
http://lists.w3.org/Archives/Public/www-webont-wg/2003Sep/0216.html) to
make the O rdf:type owl:Ontology optional for anonymous ontologies.
Changes to this effect have been made in the editor's draft of OWL S&AS,
available at http://www-db.bell-labs.com/usr/pfps/owl/semantics/

As far as having multiple ontologies goes, this is not disallowed.  From
the definition at the end of Section 4.1 an OWL DL ontology in RDF graph
form can be the result of transforming multiple ontologies.   

Please reply to public-webont-comments@w3.org indicating whether you think
this is a satisfactory response to your commments.

Peter F. Patel-Schneider


From: Bijan Parsia <bparsia@isr.umd.edu>
Subject: O rdf:type owl:Ontology optional?
Date: Fri, 8 Aug 2003 13:36:20 -0400

> http://www.daml.org/2002/06/webont/owl-ref-proposed#OWLDocument,  
> section 2.1 reads:
> 	"An OWL document consists of optional ontology headers (generally at  
> most one).."
> 
> Yet  
> http://www-db.research.bell-labs.com/user/pfps/owl/semantics/ 
> mapping.html, section 4.1, the first two lines in the table require an  
> O rdf:type owl:Ontology triple. Shouldn't there be a few [opt]s  
> scattered in there? (I take that 4.1 of S&AS is the normative section  
> and would trump the reference.)
> 
> Tricky reading that table backwards, btw :)
> 
> Cheers,
> Bijan Parsia.

From: Bijan Parsia <bparsia@isis.unc.edu>
To: public comments <public-webont-comments@w3.org>
Date: Fri, 8 Aug 2003 14:25:37 -0400
X-Mailer: Apple Mail (2.552)
Resent-Message-Id: <200308081823.h78INkfq001898@frink.w3.org>
Resent-Sender: public-webont-comments-request@w3.org

> On Friday, August 8, 2003, at 01:36  PM, Bijan Parsia wrote:
> [snip prior comment]
> 
> Whoops, I forgot to add:
> 	What happens with multiple O rdf:type owl:Ontology s?
> 
> Reference says that there are "generally *at most* one" (emph added). I 
> see several reading of the mapping table reversed, including having to 
> generate two Ontology(...)s which differ only by their Annotation s.
> 
> Cheers,
> Bijan Parsia.

Subject: Fwd: O rdf:type Ontology repeatable? (was Re: O rdf:type owl:Ontology optional?)
From: Bijan Parsia <bparsia@isr.umd.edu>
To: public comments <public-webont-comments@w3.org>
Date: Fri, 8 Aug 2003 15:02:34 -0400
X-Mailer: Apple Mail (2.552)
Resent-Message-Id: <200308081900.h78J0ZMX014727@frink.w3.org>
Resent-Sender: public-webont-comments-request@w3.org

> Sorry, mail program picking on me.
> 
> Begin forwarded message:
> 
> > On Friday, August 8, 2003, at 02:25  PM, Bijan Parsia wrote:
> >
> >> On Friday, August 8, 2003, at 01:36  PM, Bijan Parsia wrote:
> >> [snip prior comment]
> >>
> >> Whoops, I forgot to add:
> >> 	What happens with multiple O rdf:type owl:Ontology s?
> >>
> >> Reference says that there are "generally *at most* one" (emph added). 
> >> I see several reading of the mapping table reversed, including having 
> >> to generate two Ontology(...)s which differ only by their Annotation 
> >> > s.
> >
> > Ok, I missed ontologyProperties, which require mulitple X rdf:type 
> > owl:Ontology s. Ok, but I don't quite see what prevents one from 
> > taking the subsequent rdf:type owl:Ontology s as input to the 
> > Ontology() productions. E.g., if the URIReference appears in the 
> > object position of some Annotation, is it barred from being the O of 
> > some Ontology(...)? What if it occurs as both subject and object? (I 
> > don't know *why* one would do that, but it certainly seems perfectly 
> > possible (e.g., with a foo:latestVersionOf ontologyProperty).
> >
> > Cheers,
> > Bijan Parsia.

Received on Friday, 19 September 2003 12:52:10 UTC