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

Re: "OWL 2 IRI vocabulary" considered harmful

From: Peter F. Patel-Schneider <pfps@research.bell-labs.com>
Date: Fri, 10 Apr 2009 12:21:38 -0400 (EDT)
Message-Id: <20090410.122138.09191780.pfps@research.bell-labs.com>
To: schneid@fzi.de
Cc: public-owl-wg@w3.org
From: "Michael Schneider" <schneid@fzi.de>
Subject: RE: "OWL 2 IRI vocabulary" considered harmful (was Re: Occurrences of "OWL 2 Full" in our documents)
Date: Fri, 10 Apr 2009 17:57:52 +0200

> Peter F. Patel-Schneider wrote:
> 
>>> But, after all, isn't the most distinguishing aspect of the
>>RDF/RDFS/OWL 1/2 vocabularies
>>> that they consist of URI/IRIs? They don't have much to do with RDF
>>graphs. So why not
>>> simply call them "IRI vocabularies"? This seems to at least avoid
>>confusion with our
>>> other syntaxes.
>>
>>I don't think that "IRI vocabulary" works at all.  After all, names of
>>classes, etc., in OWL 2 are IRIs, so "IRI vocabulary" is even more
>>confusing than just "vocabulary".  The most important aspect of the
>>vocabulary that is given special meaning in the RDF-based semantics is
>>that it is vocabulary in the RDF graph encoding for OWL 2 that is given
>>meaning by the RDF-based semantics, so "RDF" should be somewhere in the
>>term itself.
> 
> Then what about... "OWL 2 RDF-Based Vocabulary"? :-)
> 
> (The names for the different concepts are already so ugly, it doesn't matter
> to introduce yet another ugly name. But this one would at least perfectly
> match the chosen naming convention so far, and wouldn't have any of the
> different problems we have identified, I think.) 
> 
> Michael

Fine by me.

peter
Received on Friday, 10 April 2009 16:23:14 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 10 April 2009 16:23:14 GMT