- From: Jeen Broekstra <jeen.broekstra@aidministrator.nl>
- Date: Fri, 20 Sep 2002 15:41:56 +0200
- To: seth@robustai.net
- CC: m batsis <mbatsis@netsmart.gr>, www-rdf-interest@w3.org
Seth Russell wrote:
> Jeen Broekstra wrote:
>
>> By the way, most[1] inferencers over RDF Schema interpret the
>> domain/range property semantics as inference rules rather than
>> constraints, which basically means that whenever you introduce a
>> property p with a certain range R (p rdfs:range R), and you
>> introduce a instance a with that property (a p b), its value is
>> automatically derived to be of type R (b rdf:type R).
>
> But there is no way to define the property p *using only the RDF and
> RDFS vocabulary*, such that the RDF Model thory would entail {ex:b
> rdf:type ex:R.} from {ex:p rdfs:range ex:R. ex:a ex:p ex:b.}, is
> there ?
I'm afraid I'm not following. The example you give seems equivalent to
my own, and the entailment is a direct consequence of the RDF Model
Theory entailment rule rdfs3. I do not see where additional vocabulary
is introduced.
Maybe I do not understand what you mean with 'only the RDF and RDFS
vocabulary'?
Jeen
--
jeen.broekstra@aidministrator.nl
aidministrator nederland bv - http://www.aidministrator.nl/
julianaplein 14b, 3817 cs amersfoort, the netherlands
tel. +31-(0)33-4659987, fax. +31-(0)33-4659987
Received on Friday, 20 September 2002 09:42:53 UTC