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

Re: ISSUE-16 (entity annotations)

From: Michael Smith <msmith@clarkparsia.com>
Date: Wed, 26 Mar 2008 10:19:36 -0400
To: Boris Motik <boris.motik@comlab.ox.ac.uk>
Cc: public-owl-wg@w3.org
Message-Id: <1206541176.6984.37.camel@msmith-laptop-wired.int.clarkparsia.com>

Though tabled at the 2008-03-19 telecon, this issue was not revisited on
the list following the meeting.  Below I restate what I think were
concerns about the proposed resolution.

On Wed, 2008-01-30 at 19:36 +0000, Boris Motik wrote:

> <axiom> := <classAxiom> | ... | <annotation>
> Note that the terminal <annotation> would here mean "annotation axiom", not the meaning that it has in the existing spec. Then we
> could define <annotation> as follows:
> <annotation> := "Annotation" "(" <annotationTarget> <annotationURI> <annotationValue> ")"

Given this syntax, it would be possible to have an ontology consisting
of a single annotation axiom, e.g.

Annotation ( SubClassOf ( A B ) rdfs:comment "My SubClass Axiom" )

Concerns were raised on the question of whether such an ontology would
contain the annotationTarget axiom, in this case

SubClassOf( A B )

or whether the annotation axiom could exist without the target axiom.

Peter said he would object to Boris' proposal if the annotation axiom
could exist without the target axiom.  This could be addressing by
adding a non-structural restriction on annotation axioms.

On the other hand, I recall that a use case put forward for rich
annotations was the ability to have logical axioms exist within
annotations and not have logical consequences.  So, there may be *some*
reason to allow the axiom to be used in an annotation but not exist in
the ontology.

Hopefully, interested parties will comment.

Mike Smith

Clark & Parsia
Received on Wednesday, 26 March 2008 14:20:19 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:42:03 UTC