W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > April 2003

Re: Denotation of owl:Class

From: Brian McBride <bwm@hplb.hpl.hp.com>
Date: Fri, 25 Apr 2003 14:52:46 +0100
Message-Id: <5.1.0.14.0.20030425144622.047468c0@localhost>
To: Mike Dean <mdean@bbn.com>
Cc: RDF Core <w3c-rdfcore-wg@w3.org>

At 06:03 25/04/2003 -0700, Mike Dean wrote:

[...]


>It may be helpful to think of this as owl:Class being a
>proper subclass of rdfs:Class in OWL DL and OWL Lite, and an
>"improper" (equivalent) subclass in OWL Full.  Recall that
>class equivalance can be expressed by mutual subClassOf
>relationships, so OWL DL and OWL Lite assert
>
>   owl:Class rdfs:subClassOf rdfs:Class
>
>while OWL Full adds
>
>   rdfs:Class rdfs:subClassOf owl:Class
>
>to make them equivalent.

Click, I think.

Its the "in owl lite" and "in owl full" language that confused me I think.

Can I think of this as:

   rdfs:Class rdfs:subClassOf owl:Class .

is true, but something that OWL DL reasoners don't know.

That fixes my "naive user assumption" problem with the denotation.

But it still leaves me wondering why its needed.  If its ok to feed only 
some of the semantics to a DL reasoner, why not stick to rdfs:Class but let 
it have only a limited understanding of Class?

Brian
Received on Friday, 25 April 2003 09:52:10 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:57:01 EDT