- From: Markus Krötzsch <markus.kroetzsch@comlab.ox.ac.uk>
- Date: Wed, 25 May 2011 21:13:49 +0100
- To: Michael Schneider <schneid@fzi.de>
- CC: antoine.zimmermann@insa-lyon.fr, W3C SWIG Mailing-List <semantic-web@w3.org>
On 25/05/11 12:47, Michael Schneider wrote: > Hi! > > On Thu, 19 May 2011 Markus Krötzsch wrote: > > >> What Markus says here I guess is that, in spite of the limitations of > >> the punning mechanism, a full-fledged OWL 2 DL reasoners will likely > >> infer more things than *currently existing* incomplete OWL Full > >> reasoners. > > > > Right. > > Not right! See my yesterday's mail: > Please be careful. The sentence is "What Markus says here I guess is that ..." and this is certainly right. Further careful reading will reveal that there is no claim made that any OWL DL tool will derive strictly more consequences than any OWL Full tool. I think this should be obvious. But a very light-weight RDF entailement-rule reasoner may miss other entailments that OWL (direct and RDF-based) semantics would give you, so it might well be that in some situations the DL reasoner is "more complete" for the RDF-based semantics of your input than any RDF-based semantics tool (I am aware of the subtle differences that the semantics have in both ways, but this might not be the main issue for a pragmatic decision). As you rightly say, using owl:sameAs to align classes is not a good approach in direct semantics, since it would not entail owl:equivalentClasses. I already argued for this position, but deliberately selected a more complicated case in my email since the owl:sameAs issue would be relatively easy to work around (by [additionally] using equivalentClasses in modelling right away, or by a simple "syntactic" completion before passing things to a direct semantics tool). Inferred sameAs would not allow such a treatment. I think the paper on OWL 2 metamodelling that I pointed to gives quite a good idea of what is possible in direct semantics. Regarding the rest of your email: the discussion there seems to grow into an argument to defend the usefulness of OWL 2 Full in general. I do not think that I have taken a position in this respect, although you read that into my statements. I have no intention to even argue against any of what you said, so let us just put this topic aside. Best regards, Markus -- Dr. Markus Krötzsch Oxford University Computing Laboratory Room 306, Parks Road, Oxford, OX1 3QD, UK +44 (0)1865 283529 http://korrekt.org/
Received on Wednesday, 25 May 2011 20:14:19 UTC