W3C home > Mailing lists > Public > www-webont-wg@w3.org > November 2002

RE: MT for imports (was: Re: Imports Proposal)

From: Jerome Euzenat <Jerome.Euzenat@inrialpes.fr>
Date: Wed, 13 Nov 2002 18:37:35 +0100
Message-Id: <a05111b05b9f83b755d8f@[194.199.20.189]>
To: "Massimo Marchiori" <massimo@w3.org>, "pat hayes" <phayes@ai.uwf.edu>, "Jeff Heflin" <heflin@cse.lehigh.edu>
Cc: <www-webont-wg@w3.org>

Hello,

In his message (RE: MT for imports (was: Re: Imports Proposal)) of 13/11/2002,
Massimo Marchiori wrote:
>Yes, that's precisely the major unanswered issue I was pointing out in
>http://lists.w3.org/Archives/Public/www-webont-wg/2002Nov/0117.html
>(and previous emails)
>"owl entailment would depend on the timed web structure"
>and raised as very first objection at the first telecon we had on this ;)

The timed web structure applies to ontologies exactly like it applies 
to hyperlinks in html, xsl:include and xsl:imports in XSLT, etc.

So, my opinion that:
(1) we should not have to consider this, we should just consider the 
state of the web at that moment;
(2) we should deal with it "gracefully", i.e. try to not be bothered 
by failures. To that extend, I would suggest that if we fetch an 
ontology it is added to the current one, if we cannot (404 or else) 
it reduces to the empty set. This should not be that disturbing as 
long as when there is more ontology, there is only less models and 
thus more consequences (summary: the more 404, the less we can 
deduce, but we still can).

The point is that, if we constraint too much the semantics on the 
state of the web, we shall deduce nothing in a great number of cases. 
If we do not state the semantics of import (or allow to ignore it), 
then we might deduce too much (with regard to the designer's 
expectations) in a number of cases.

Of course, imports are important and having a semantics for these is 
part of the  job. But, concerning the trouble of stacking owl:import 
on RDF, I think this is a pity that RDF does not includes an 
rdf:import.

If I remember well, there is a proposed xml:include somewhere as well 
and I would have liked to see an owl:import based on an rdf:import 
relying in its xml form on xml:include.


-- 
  Jérôme Euzenat                  __
                                  /      /\
  INRIA Rhône-Alpes,            _/  _   _   _ _    _
                               /_) | ` / ) | \ \  /_)
  655, avenue de l'Europe,    (___/___(_/_/  / /_(_________________
  Montbonnot St Martin,       /        http://www.inrialpes.fr/exmo
  38334 Saint-Ismier cedex,  /          Jerome.Euzenat@inrialpes.fr
  France____________________/                Jerome.Euzenat@free.fr
Received on Wednesday, 13 November 2002 12:37:28 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:57:55 GMT