W3C home > Mailing lists > Public > public-owl-wg@w3.org > March 2008

RE: Question about annotation of and to axioms

From: Boris Motik <boris.motik@comlab.ox.ac.uk>
Date: Thu, 13 Mar 2008 17:44:23 -0000
To: "'Alan Ruttenberg'" <alanruttenberg@gmail.com>, "'Web Ontology Language \(\(OWL\)\) Working Group WG'" <public-owl-wg@w3.org>
Message-ID: <003001c88531$dfe49390$d012a8c0@wolf>

Hello,

No, we can't have axioms as targets of annotations. This is highly related to ISSUE-16, and is one of the undesirable features of
OWL 1.1 that we might want to change. (Just as a reminder, ISSUE-16 actually complains about the fact that, in OWL 1.1, annotations
on axioms are parts of axioms themselves.)

We might fix ISSUE-16 and address your concern by making annotations on axioms external to the axiom. Here is what the change would
look like:

- We delete the annotation from all axioms.

- We add a new type of axiom, called AxiomAnnotation. The grammar would be like this:

axiomAnnotation := 'AxiomAnnotation' '(' annotation { annotation } axiom ')'

The changes to the diagrams, the RDF serialization, and the XML syntax should be straightforward. I believe that such a solution
would actually be much nicer from an architectural point of view. Thus, I would support this change. If everyone's OK with this, I
could change the spec in no time.

Regards,

	Boris

> -----Original Message-----
> From: public-owl-wg-request@w3.org [mailto:public-owl-wg-request@w3.org] On Behalf Of Alan Ruttenberg
> Sent: 13 March 2008 17:34
> To: Web Ontology Language ((OWL)) Working Group WG
> Subject: Question about annotation of and to axioms
> 
> 
> We can attach annotations to axioms in the current proposal. Can we
> also have Axioms as the target of annotation? The question arises in
> the context of a proposed epistemic relations for the (OBO) relation
> ontology - evidential relationships that say, for instance, some
> document supports some axom. The inverse of the relation might be
> used if axioms can't be the objects of annotations, in this case, but
> I'm still curious as to the answer.
> 
> -Alan
Received on Thursday, 13 March 2008 17:45:46 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 13 March 2008 17:45:47 GMT