W3C home > Mailing lists > Public > www-webont-wg@w3.org > January 2003

Re: owl:Class in class expressions - substantive

From: Guus Schreiber <schreiber@swi.psy.uva.nl>
Date: Wed, 22 Jan 2003 16:15:49 +0100
Message-ID: <3E2EB5A5.6040603@swi.psy.uva.nl>
To: Jeremy Carroll <jjc@hplb.hpl.hp.com>
CC: www-webont-wg@w3.org

Jeremy Carroll wrote:
> 
> This msg proposes a substantive change in the mapping rules.
> I will make later comments of a more editorial/bug-fixing nature.
> 
> The focus is on class expressions e.g. a unionOf.
> 
> With the current mapping rules the following is an OWL DL document (modulo
> declarations etc)
> 
> DocA
> ====
> <owl:Class rdf:ID="u">
>    <owl:sameClassAs>
>      <rdf:Description>
>        <owl:unionOf rdf:parseType="Collection">
>            <owl:Class rdf:ID="a"/>
>            <owl:Class rdf:ID="b"/>
>        </owl:unionOf>
>       </rdf:Description>
>     </owl:sameClassAs>
> </owl:Class>
> 
> 
> The following, which I believe better follows standard DAML+OIL idiom, is
> not:
> 
> DocB
> ====
> <owl:Class rdf:ID="u">
>    <owl:sameClassAs>
>      <owl:Class>
>        <owl:unionOf rdf:parseType="Collection">
>            <owl:Class rdf:ID="a"/>
>            <owl:Class rdf:ID="b"/>
>        </owl:unionOf>
>       </owl:Class>
>     </owl:sameClassAs>
> </owl:Class>
> 

The DocB type of OWL examples are all over Guide and Ref, so yes, the 
mapping rules should cover them.

> The relevant mapping rule is:
> 
> unionOf(description1 … descriptionn)
> 
> =>
> 
> _:x owl:unionOf T(SEQ description1…descriptionn) .
> 
> 
> Three options are:
> A) leave us as
> B) change to
> 
> unionOf(description1 … descriptionn)
> 
> =>
> 
> _:x owl:unionOf T(SEQ description1…descriptionn) .
> _:x rdf:type owl:Class .
> 
> C) add above rule as an alternative

Options B & C are acceptable to me, with a preference for B as this 
seems the least work.

Guus

> 
> ==========
> Effect
> 
> A) DocA is OWL DL, DocB is OWL Full
> B) DocB is OWL DL, DocA is OWL Full
> C) both DocA and DocB are OWL DL
> 
> I argue that (B) has the additional advantage of being easier to articulate,
> e.g.:
>   "Within OWL Lite and OWL DL all nodes must have a type."
> 
> So I propose B, and similarly for the other class expression rules.
> (Issuette what type should a datarange have?)
> 
> Jeremy
> 
> 
> 
> 
> 
> 

-- 
A. Th. Schreiber, SWI, University of Amsterdam,
http://www.swi.psy.uva.nl/usr/Schreiber/home.html
Received on Wednesday, 22 January 2003 10:21:05 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:57:57 GMT