W3C home > Mailing lists > Public > public-owl-wg@w3.org > February 2009

Re: Last call comment - conformance/datatype map

From: Bijan Parsia <bparsia@cs.man.ac.uk>
Date: Tue, 24 Feb 2009 15:52:51 +0000
Message-Id: <85916822-DA11-4B76-8009-691422198FD1@cs.man.ac.uk>
Cc: W3C OWL Working Group <public-owl-wg@w3.org>
To: Alan Ruttenberg <alanruttenberg@gmail.com>
On 24 Feb 2009, at 15:31, Alan Ruttenberg wrote:

> As I have stated in the past, our experiment with letting
> implementations define xml schema datatypes in OWL 1 was a failure.
> That point of view is nothing new

Sure, but it's not really the intention of the working group to do  
anything (more) about it. You presented it as if we had consensus  
about a new thing.

There's a big difference between encouraging implementors to  
implement something not well speced (OWL 1 sitch) and not encouraging  
them but not forbidding them.

And, really, the problem before was that we sanctioned them without  
any guidance. No one's proposing that now.

> - I've stated it in the past. I
> think we should be clearly setting expectations that standardizing OWL
> behavior in the W3C space is to be managed within the W3C process.

I don't think that we should set those expectations (qua working  
group). Nor do I think it should be part of our specs. That's a w3c  
corporate thing.

> We
> already have restrictions to this effect in Syntax.
> http://www.w3.org/2007/OWL/wiki/Syntax#IRIs_and_Namespaces
>
> There are a variety of ways of accomplishing this goal and I'm open to
> suggestions on how to go about doing this.

Uh, why isn't that sufficient? It already says that the XSD namespace  
is reserved.

"""All IRIs from the reserved vocabulary not listed in Table 3  
constitute the disallowed vocabulary of OWL 2 and must not be used in  
OWL 2 to name entities, ontologies, or ontology versions."""

I don't see what more that you want. I think anything that goes  
beyond that and tries to say more about implementations is likely to  
be a bad idea. Contrariwise, tis bit is totally unambiguous and seems  
to do the job.

Cheers,
Bijan.
Received on Tuesday, 24 February 2009 15:49:18 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 24 February 2009 15:49:18 GMT