W3C home > Mailing lists > Public > www-ws@w3.org > September 2003

Re: what is the "semantic core" of DAML-S?

From: Bijan Parsia <bparsia@isr.umd.edu>
Date: Sat, 20 Sep 2003 17:59:27 -0400
To: www-ws@w3.org
Message-Id: <B44DBB64-EBB5-11D7-8C02-0003939E0B44@isr.umd.edu>

On Saturday, September 20, 2003, at 05:28  PM, Drew McDermott wrote:

>    [Joachim Peer]
>> a) surely everything can be represented somehow by DL ontologies, and
>> DAML-S does just that.
>
>    [Bijan Parsia]
>    Uh. That seems false to me, unless you mean, "some representation 
> can
>    be encoded in" a la how DRS encodes more complex logical formulae in
>    plain RDF. Not the most useful sense of "able to represent", IMHO.
>
> It seems false to me, too.

And to push it even further, since there are tons of very inexpressive 
DLs, it seems true that for any given DL there will be all sorts of 
"canonically" ontological facts it can't express.

>   I think what has happened is that people
> have gotten used to expressing some "ontological" facts in DLs, and
> think of stuff that can't be so expressed as requiring something else,
> like "rules."
[snip]

I see no evidence of this coming out of the DL community itself.

>   The conclusion that everything "ontological" can be
> expressed in DLs then becomes a tautology.
[snip]

Pesonally, I don't know if there's ANY sensible characterization of 
"ontology", particularly one that supports useful characterization of 
what expressivity is needed.

(PSL uses quite a bit of first order logic, well beyond, I believe, any 
of the decidable fragments. Plus, many ontology types lean toward 
non-monotonic inheritence, which is often difficult to combine with 
standard expressive DLs, like OWL.)

Cheers,
Bijan Parsia.
Received on Saturday, 20 September 2003 17:56:21 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 3 July 2007 12:25:44 GMT