W3C home > Mailing lists > Public > semantic-web@w3.org > September 2006

Re: Individuals with more than one rdf:type

From: Andrea Splendiani <andrea@pasteur.fr>
Date: Wed, 20 Sep 2006 14:18:34 +0200
Message-Id: <D1DC4A66-596B-48D6-B370-FA6EDF8D86EC@pasteur.fr>
Cc: semantic-web@w3.org
To: Steve Harris <S.W.Harris@ecs.soton.ac.uk>

Properties on classes are always allowed, but their semantics is not  
the same as in OO.
This applies to RDFS/OWL(any).
As far as i can see  you can export  the content in such an example  
in RDF, where the RDF dictionary comply to an OWL ontology. And you  
don't need OWL-Full, in fact I don't see how much you can gain from  
OWL in an ontology that reflects an OO model, that inform a  
distinction between classes and individuals quite naturally.
The are no problems in DL. Just one has to be careful not to assume a  
different semantics, or most likely, he will end up with a far less  
characterized ontology then expected (and maybe some unexpected  
inference due to Open World).

best,
Andrea



Il giorno 20/set/06, alle ore 10:41, Steve Harris ha scritto:

>
> On Tue, Sep 19, 2006 at 08:51:31 -0500, ben syverson wrote:
>>
>>
>> On Sep 19, 2006, at 7:46 PM, Richard Newman wrote:
>>
>>> RDF types are not object oriented classes. RDF instances are not
>>> objects. They do not inherit anything in the OO sense.
>>
>> Got it -- "inherit" is a dirty word. The issue I'm dealing with is
>> that the internal model of my application is indeed more closely
>> related to the OO class/object model, so I'm trying to figure out the
>> most useful way to publish that information as RDF documents and OWL
>> ontologies.
>>
>> For example, classes can have properties in likn, and I'm trying to
>> determine a good way to represent that in RDF/OWL. My current plan is
>> to simply copy out the properties of a class to its individuals when
>> serializing to RDF/XML. I don't know of a better way to represent,
>> for example, the concept "all individuals of the class CocaColaCans
>> have the property 'color' with the value Red."
>>
>> Any suggestions?
>
> Use RDFS or OWL Full instead? RDFS certainly allows properties on  
> classes,
> and I'm pretty sure OWL Full does too. It doesn't sound like youre  
> doing
> any complex reasoning, so the abilities offered by sticking to the  
> OWL DL
> subset is probably not relevent.
>
> - Steve
>
Received on Wednesday, 20 September 2006 12:18:43 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 07:41:53 UTC