W3C home > Mailing lists > Public > public-owl-dev@w3.org > July to September 2009

Re: Semantics of owl:unionOf vs subclass ...

From: Bijan Parsia <bparsia@cs.man.ac.uk>
Date: Fri, 10 Jul 2009 16:37:45 +0100
Message-Id: <AAABFE41-5600-4946-BF80-5941BE29941F@cs.man.ac.uk>
Cc: Brandon Ibach <bibach@earthlink.net>, public-owl-dev@w3.org
To: Ruth Dhanaraj <ruthdhan@gmail.com>
On 10 Jul 2009, at 04:08, Ruth Dhanaraj wrote:

> There may be more than one property with domain C.... isn't it useful
> to have it named for future use?

Sometimes yes and sometimes no. My only point in bringing it up is to  
show that you don't *need* to define a new concept in order to  
express Union semantics.

If you want to be as close as possible to having multiple domain  
axioms interpreted disjunctively, then using the class expression is  
better.

That is, right now

1)	p domain A
2)	p domain B
entails
3)	p domain IntersectionOf(A B)

So the closest way to change this to union semantics is to delete 1  
and 2 and add:

4) 	p domain UnionOf(A B)

Now, if you are happy naming that expression...go ahead. You might  
also consider adding the expression version (named or otherwise) to  
the first case, i.e., replacing 1 and 2 with

5) 	p domain C
6) 	C equivalent IntersectionOf(A B)

These modeling choices have slightly different marginal advantages.

> Plus, the resulting info can be
> expressed in plain RDF (without OWL)...

You mean that the domain axiom "dumbs down" better? Yes, I agree. The  
cost is an additional concept that will always be "visible" in your  
interface. Sometimes that's good, sometimes it's not.

Cheers,
Bijan.
Received on Friday, 10 July 2009 15:45:23 GMT

This archive was generated by hypermail 2.3.1 : Wednesday, 27 March 2013 09:32:57 GMT