RE: DTTF: darkest africa

From: "Jeremy Carroll" <jjc@hplb.hpl.hp.com>
Subject: RE: DTTF: darkest africa
Date: Fri, 31 May 2002 11:56:00 +0100

> Peter:
> > But how does that prevent some other agent from using that
> > URIref?  Why is
> > owl:ABCDEF not (a QName that expands into) a valid URI that
> > can be used in
> > OWL?  I don't see any hint that the working group is going to
> > ``control''
> > the owl namespace to this extent.
> 
> 
> We have a duty to do so.

Huh?  Where did this duty come from?  When did the WG accept it?


> > But why make them [unusual constructions] contradictions?
> 
> I was trying to make them implementation defined behaviour.
> I could agree with your suggestion to do that syntactically.

That is not the thrust of what I would prefer at all.  I would prefer to
make them ``illegal'', i.e., any system that did anything with them
(besides signalling a syntax error) would be violating the OWL
specification.

> > Well, we disagree on the meaning of just about everything in
> > an ontology,
> > including things like
> >
> >   John a [intersectionOf A B] .
> >
> > so I don't see how ruling out misuses of OWL vocabularly will help the
> > situation too much.
> >
> 
> Come Peter, we at least agree that:
> 
> John a [intersectionOf A B] .
> 
> entails
> 
> John a B .

Yes, we agree on that entailment, but that is not sufficient to pin down
the meaning of intersection constructions.

> > If you don't want to consider triples that misuse the OWL
> > vocabularly, then
> > why not just make them syntactically invalid?  I would view
> > that as very
> > much clearer and cleaner.
> 
> We don't control the syntax, 

If we don't control the syntax, then how in the world are we supposed to
control owl:ABCDEF?

> but yes I think that would be plausible.
> OWL is defined over RDF graphs except ones that contain the specified
> triples.
> i.e. we do not define an OWL semantics for those graphs.
> That would even solve the layering problem in the sense that all RDFS
> entailments would then be OWL entailments (except those involving graphs
> which are not OWL graphs). 

I don't think that this is sufficient to solve the layering problem.

> The OWL model theory could have features like
> classes not in the domain of discourse even though this is a different
> philosophy from
> RDFS.

This would almost certainly involve violating the property you mention just
above.


> Jeremy

peter

Received on Friday, 31 May 2002 07:08:49 UTC