W3C home > Mailing lists > Public > public-sws-ig@w3.org > June 2006

Re: OWLS: KIF-Condition vs SWRL-Condition

From: Bijan Parsia <bparsia@isr.umd.edu>
Date: Sun, 18 Jun 2006 14:05:18 +0100
Message-Id: <2F93EC00-3E65-470C-B4AE-E7117A298881@isr.umd.edu>
Cc: <public-sws-ig@w3.org>
To: "Gerhard Austaller" <gerhard.austaller@o2online.de>

On Jun 18, 2006, at 1:51 PM, Gerhard Austaller wrote:

> Hello
>
> In OWLS 1.2 there are several languages proposed to express  
> conditions e.g.
> SWRL and KIF. But are these two languages different in expressiveness?

Very much so. SWRL is a proper subset of first order logic while KIF  
is at least FOL (plus some funky things)>

> So
> couldn't it happen that some conditions that can be expressed with  
> KIF can
> not expressed with SWRL?

Yes, that's (part of) the point.

It's not great for interop, but since there aren't a lot of KIF or  
(full) SWRL reasoners about (in other words, you're best off sticking  
to tree like queries over OWL DL kbs), it's a bit of a moot point.  
KIF is in there explicitly mostly becasue one of the OWL-S developers  
had a need for KIF expressible conditions. Ideally, as other  
representation languages become popular (or just used by you) one can  
further parameterize OWL-S descriptions. But if you are going for  
interop/portability, I would stick with OWL-DL and tree like  
conjunctive queries of SWRL atoms (to be, one hopes, soon superceded  
by sparql queries).

Cheers,
Bijan.
Received on Sunday, 18 June 2006 13:05:07 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:32:54 UTC