W3C home > Mailing lists > Public > public-owl-dev@w3.org > October to December 2007

Re: [OWLWG-COMMENT] Punning and the "properties for classes" use case

From: Bijan Parsia <bparsia@cs.man.ac.uk>
Date: Mon, 5 Nov 2007 21:14:23 +0000
Message-Id: <675889A1-4952-4383-8018-83E5872DBBFD@cs.man.ac.uk>
Cc: Owl Dev <public-owl-dev@w3.org>
To: "Michael Schneider" <schneid@fzi.de>

On 5 Nov 2007, at 20:57, Michael Schneider wrote:

[snip]
> Ok, now we come to some core point: The use cases. When I wrote my  
> original
> mail, the only relevant usecase for "classes as instances" I could  
> imagine
> was to assign object properties to classes. But there are more,  
> right? What
> I am still missing is a document which lists and discusses some  
> relevant use
> cases for punning.

People want some syntactic freedom. The bad use is to get  
"meaningful" annotation properties (i.e., dc:date-modifed implies  
dc:date). The reason I regard this as bad is two fold: 1) it  
interprets classes into the same domain as the subject matter model  
and I think annotations are better for being more separated from the  
subject matter (subtle bugs because your classes end up being  
instances of Person are annoying to contemplate) and 2) I'd like to  
have more freedom in my logic of annotations.

Otherwise, it's pretty harmless. I'd prefer to pun annotations in to  
a disjoint domain.

> Particularly, I am looking for use cases where the
> missing sameAs-equivalentClass entailment does not hurt.

It never hurts. We've had it in pellet for years with no problems  
reported by any user :)

> Honestly, I have
> difficulties to imagine that this is really the case.

Ad esse, ad posse est. Frankly, I've grown ever more strongly against  
hilog semantics (= entailing iff) because it's really hard to  
understand (what happens with a disjunctive equality?), and the  
actual worthwhile examples are non-existent.

Most people don't even know about the possibility. I explained it to  
one owl full advocate and he said, "Oh, that's really a bug, isn't it?"

I regard relying on that entailment (esp. for alignment) as a total  
anti-pattern. Much better would be, after a merge, do query for all  
equalities involving individuals which pun classes, then review them  
manually to see which you actually think should push equivalences.  
Punning allows you to do this easily.

Mashups require some human intervention. I see no reason why ontology  
mashups wouldn't. Though it is good to be able to merge them and do  
*something* without getting a syntax error. That's what punning gives  
you.

> Nevertheless, the most
> relevant point is to first have such a use case document. And also  
> I want to
> have a technical specification of punning, because I have the  
> feeling that
> there is still no common agreement on what punning provides and  
> what not.

See the semantics document from OWL 1.1. See Boris's paper:
	<http://web.comlab.ox.ac.uk/oucl/work/boris.motik/publications/ 
motik07metamodeling-journal.pdf>

wherein it's called "contextual semantics" or pi-semantics.

There are multiple implementations. I don't know where you get your  
feeling from, but it seems rather off to me.

Cheers,
Bijan.
Received on Monday, 5 November 2007 21:12:57 GMT

This archive was generated by hypermail 2.3.1 : Wednesday, 27 March 2013 09:32:55 GMT