Re: possible semantic bugs concerning domain and range

From: "Jonathan Borden" <jonathan@openhealth.org>
Subject: Re: possible semantic bugs concerning domain and range
Date: Mon, 23 Sep 2002 20:47:12 -0400

> Peter F. Patel-Schneider wrote:
> >
> > > Peter F. Patel-Schneider wrote:
> > > >
> > > > RDFS already allows multiple ranges, just as does OWL, and gives them
> > > > intersection semantics, just as does OWL.
> > > >
> > >
> > > Hrmmph ... this seems counterintuitive, I'd assumed that RDFS applied
> > > union-of semantics to multiple ranges.
> >
> > This was changed a long time ago.
> >
> 
> I see. Well I've just burnt a few vacuum tube in my frontal lobes trying to
> figure out if either of these options (conjunctive/disjunctive) are
> monotonic, and I'll just give up now and go back to drinking my glass of
> wine and thinking about easy stuff like brain surgery...
> 
> Jonathan

The disjunctive one is non-monotonic, as can be easily seen.

The following RDFS graph

	ex:foo rdfs:range ex:bar .
	ex:a ex:foo ex:b .

entails

	ex:b rdf:type ex:bar .

However a disjunctive reading of rdfs:range means that

	ex:foo rdfs:range ex:bar .
	ex:foo rdfs:range ex:baa .
	ex:a ex:foo ex:b .

does not entail

	ex:b rdf:type ex:bar .


peter