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

Re: concerning lite, fast, large versions of OWL

From: Ian Horrocks <horrocks@cs.man.ac.uk>
Date: Sun, 20 Oct 2002 19:57:31 +0100
Message-ID: <15794.64667.697827.589097@merlin.horrocks.net>
To: "Lassila Ora (NRC/Boston)" <ora.lassila@nokia.com>
Cc: <www-webont-wg@w3.org>

On October 18, Lassila Ora (NRC/Boston) writes:
> 
> [second attempt, first one disappeared yesterday...]
> 
> Frank van Harmelen wrote:
> > Reasoners for OWL/RDF-style will be much harder to implement than reasoners
> > for OWL/FOL-style (complete reasoners would be impossible to implement if
> > OWL/RDF-style turns out to be an undecidable language, as it might well be)
> 
> Barring the possibility that RDF-style turns out to be undecidable, does
> anyone have any concrete (practical) estimates of what  "much harder to
> implement" really means? I am looking for a *compelling* pragmatic argument
> against the classes-as-instances feature.

To the best of my knowledge, there is no known direct algorithm for
such a language (i.e., OWL lite + "RDF-style"). The best you can do is
axiomatise and use an FOL prover. Reasoning in OWL lite would therefore
be much the same as reasoning in large OWL - and certainly much harder
than reasoning in fast OWL. There seems to be little justification for
the existence of OWL lite under these circumstances.

Ian
Received on Sunday, 20 October 2002 14:57:42 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:57:53 GMT