W3C home > Mailing lists > Public > public-rif-wg@w3.org > November 2007

Introduction qua OWLWG liaison

From: Bijan Parsia <bparsia@cs.man.ac.uk>
Date: Wed, 7 Nov 2007 08:02:01 +0000
Message-Id: <A7DA48B2-F148-4267-9F18-6BD3944323DE@cs.man.ac.uk>
To: RIF WG <public-rif-wg@w3.org>

At the last OWLWG telecon, I was appointed the OWLWG liason to the RIF:
	<http://www.w3.org/2007/OWL/wiki/Teleconference.2007.10.31/Minutes>
(search for "Bijan with RIF").

  I also am part of the OWLED task force on DL Safe rules:
	<http://code.google.com/p/owl1-1/wiki/SafeRules>

I also am involved with both OWL+Rules reasoner implementations  
(e.g., Pellet:
	<http://clarkparsia.com/weblog/category/semweb/rules/swrl/>)
and editor implementations (historically, Swoop, and currently  
Protege4 derivatives). I know the people working on the Protege3.x  
series SWRL tab and similar efforts.

Obviously, y'all know me from my occasional email contributions to  
this list. As liaison, I'm happy to convey any questions or concerns  
from the RIF to the OWLWG, as well as evangelize RIF-OWL synergies  
more generally. Obviously, the direct concern of the OWLWG is on the  
OWL compatibility front. There seem to me to be three areas of  
possible interest to the OWLWG:

	1) Round tripping horn or other fragments.
If you look at the tractable fragments document of OWL1.1:
	<http://www.webont.org/owl/1.1/tractable.html>
it turns out that most (all?) of the fragments are *practically* horn  
reducible. That is, it is a reasonable implementation to translate  
them fairly naively to horn rules and run them on a more or less  
standard Datalog engine. (SHIQ can, practically, be reduced to  
disjunctive datalog, as shown by KAON2, but that seems to be a  
different kettle of fish).

Aside from this implementation fact, one can imagine wanting to round  
trip different aspects of the OWL syntax through a RIF based rules  
implementation.

	2) Extending OWL with Rules (DL Safe, Weakly DL safe, hex  
predicates, etc.)

	3) Data predicates and builtins.

Less directly of interest to the WG per se, I think, but of  
considerable interest to me and I assume other participants is:

	A) Extending (various) Rule Systems with OWL

(Of course, some of these are just variants of 2 and, indeed, 1, but  
some might not be. So if there are hooks or other things the OWLWG  
could add to support, e.g., Production rule system's calling OWL  
reasoners, I'm sure the group would be happy to consider them.)

My personal goal qua liaison is to avoid anything that might cause  
either group's schedule to slip.

I do not intend to participate as a full member of the group (having  
OWLWG is already nigh more than I can handle) and, unfortunately, U.  
Manchester does not have other resources for that. So, I'll be focus  
pretty exclusively on the relationship to OWL. To that end, I'm happy  
to answer questions, convey questions, try to recruit reviewers and  
implementors, etc. I'll also happily call into the odd telecon to  
facilitate progress, esp. if we can arrange the agenda to cover as  
much of any OWL topics in as few telecons as possible.

Cheers,
Bijan.
Received on Wednesday, 7 November 2007 08:02:25 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:33:43 GMT