W3C home > Mailing lists > Public > public-rule-workshop-discuss@w3.org > August 2005

1 - INTENT OF THE WG

From: Dieter Fensel <dieter.fensel@deri.org>
Date: Tue, 23 Aug 2005 16:59:17 +0200
Message-Id: <5.2.0.9.0.20050823164307.0510c598@mail1.uibk.ac.at>
To: public-rule-workshop-discuss@w3.org


Dear Jim,



indeed on the first glance there is a distinction between
a exchange syntax and a representation language for rules
on the web. However, I think we have the chance to define
a rule language for the web and even if we only define an
exchange language it needs to be done differently. Lets
take a look at the great success of OWL. If OWL was just
an exchange language then simply using First Order Logic
with equality would have done the job. Still you spent
much more effort? And I think that you were right in this!
OWL (Lite and DL) will soon replace most other Ontology
languages which is a great way to achieve interoperability.
Other language may survive for a while with slightly
different syntax and some exotic features but they will adopt
to the chosen paradigm and the way to express this. Well done [1]!

I think we have the chance to achieve the same for the rule
area with the Web Rule Language (WRL) if and only if we chose
expressivity and semantics properly. FOL with equality and multi
model semantics will NOT do this. We will neither define
a rule language for the web nor an language that makes
exchange of rules between different rule languages easy
since it does not define proper restrictions of expressivity
nor the common semantics based on a unique model.

In consequence, lets learn from the success of OWL and
let us add another success story to it.


         -- dieter


[1] A better layering of OWL-Lite and OWL-DL and inclusion
of HiLog features preventing you from OWL-Full would have served
an even better job but this is past for a while.
----------------------------------------------------------------
Dieter Fensel, http://www.deri.org/
Tel.: +43-512-5076485/8
Received on Tuesday, 23 August 2005 15:01:14 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:16:23 GMT