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

Re: Proposal to close ISSUE-73 (was Re: My ACTION-64 (to propose closing ISSUE-29))

From: Bijan Parsia <bparsia@cs.man.ac.uk>
Date: Mon, 14 Jan 2008 14:14:41 +0000
Message-Id: <5B457A94-379A-48BB-87CE-79E0DD41CB33@cs.man.ac.uk>
Cc: "Web Ontology Language ((OWL)) Working Group WG" <public-owl-wg@w3.org>
To: Jeremy Carroll <jjc@hpl.hp.com>

On 14 Jan 2008, at 12:51, Jeremy Carroll wrote:

> I was a little surprised when an action to propose closing one  
> issue, result in a proposal to close another issue which has had  
> very little discussion.

I don't know why. Working on one prompted me to think of the other. I  
changed the subject. I wanted to propose closing this issue. Seems  
perfectly appropriate to me.

Or do you think I'm not entitled to propose closing issues?

> I do not find Bijan's message as compelling on the main points

You didn't address *any* of my technical points.

> which are:
>
> - the Web context is not one that can be given a finite bound. Any  
> ontology that tries to do so is limiting its interoperability with  
> other ontologies.
[snip]

I'm sorry but this is silly. If I say that C subclass of not D, I'm  
"limiting my interoperability" with other ontologies that say that C  
subclassof D in exactly the same way. If I say that a *class* C is  
finite (i.e., equiv to {:a, :b :c} (where a != b != c) then I "limit  
my interoperability" with an ontology that says that C is a different  
finite (i.e., equiv to {:a, :b, :c, :d} where !a= b!= c).

This isn't interoperability, it's disagreement. Disagreement has to  
be allowed on the web, right? You want to eliminate a point of  
disagreement. This elimination has unknown technical effects, breaks  
backwards compatibility, and is not supported by implementors who  
would be affected by it. It's a non-starter.

Since it breaks (or has unknown effects) on recent modularity work  
(which allows people to extract and combine parts where they *do*  
agree), this requirement *hurts* interoperability.

So, again, this motivation is entirely ideological. You have not  
considered and seem unwilling to consider the technical impact in any  
detail (though you blithely pronounce on it).

Prima facie this is not a topic that needs a lot of discussion. To  
the degree it needs more, the burden is on you. At this point, it  
seem safe to close it. If you come up with new information, it can be  
reconsidered.

Cheers,
Bijan.
Received on Monday, 14 January 2008 14:12:52 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 14 January 2008 14:12:53 GMT