W3C home > Mailing lists > Public > www-rdf-logic@w3.org > March 2003

Re: RECAP: Why isn't FunctionalProperty a subClassOf owl:ObjectProperty?

From: Mike Dean <mdean@bbn.com>
Date: Wed, 05 Mar 2003 04:34:09 -0800
Message-Id: <200303051234.EAA20241@escher.SD.BBN.COM>
To: "Roger L. Costello" <costello@mitre.org>
Cc: www-rdf-logic@w3.org

> Just to make sure that I understand the discussion.  You are saying
> that:
>  
>    - SymmetricProperty
>    - TransitiveProperty
>    - FunctionalProperty
>
> are all subclasses of rdf:Property.  Is that what the conclusion is? 

I don't think so.  The initial driver for making
TransitiveProperty a subclass of ObjectProperty in DAML+OIL
(SymmetricProperty was added for OWL) was that literals
couldn't be subjects in RDF.  That hasn't changed.

Allowing datatypes to be InverseFunctionalProperties
(similar to database keys) in OWL Full muddies the situation
a bit, but OWL Full removes the disjointness of
ObjectProperties and DatatypeProperties, so continuing to
make TransitiveProperty and SymmetricProperty a subclass of
ObjectProperty isn't a problem at any of the OWL levels.

	Mike
Received on Wednesday, 5 March 2003 08:39:22 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:52:43 GMT