W3C home > Mailing lists > Public > public-owl-wg@w3.org > January 2009

Re: LC Comment: Names for axioms

From: Bijan Parsia <bparsia@cs.man.ac.uk>
Date: Mon, 26 Jan 2009 14:33:37 +0000
Message-Id: <C5CB44BE-6C3A-4EC0-A3B3-511C8C07FE9A@cs.man.ac.uk>
To: W3C OWL Working Group <public-owl-wg@w3.org>

On 25 Jan 2009, at 20:15, Michael Schneider wrote:

> Bijan Parsia wrote:
>
>> The immediate overall goal is to support so
>> called "stand off" annotations on axioms...that is, axiom annotations
>> that are not a structural part of an axiom they annotate. Such
>> annotations could be stored in separate documents (for example).
>>
>> The simplest feature that would support this would be the ability to
>> name axioms with a URI. This ability is, I think, present in OWL  
>> Full.
>
> Hi! For my interest and understanding: What do you exactly mean by  
> giving a name to an axiom in OWL Full? Do you mean to use a URI  
> instead of a bNode in the RDF encoding of an axiom annotation,

Yes.

> such as in (adapting the example from Section 2.3.1 of the RDF  
> Mapping):
>
>   s p o
>   :spoName rdf:type owl:Axiom
>   :spoName owl:subject s
>   :spoName owl:predicate p
>   :spoName owl:object o
>   TANN(annotation_1, :spoName)
>   ...
>   TANN(annotation_m, :spoName)
>
> ? This would indeed be allowed in Full (though currently not  
> reverse-mappable AFAICT; would there be any technical problems with  
> allowing this?).

I have no problem with this, except that it has more impact on our  
specs than a designated annotation property (i.e., one would have to  
revise the XML, Manchester, and functional syntaxes as well). There  
might be the need for additional constraints to prevent "accidental"  
axiom merges and to consider some issues about axiom/class/etc. punning.

Cheers,
Bijan.
Received on Monday, 26 January 2009 14:30:11 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 26 January 2009 14:30:19 GMT