W3C home > Mailing lists > Public > www-rdf-logic@w3.org > June 2001

Re: DAML-S expressiveness challenge #1

From: David Martin <martin@ai.sri.com>
Date: Fri, 15 Jun 2001 10:38:34 -0700
Message-ID: <3B2A481A.D2800F8F@ai.sri.com>
To: Jim Hendler <jhendler@darpa.mil>
CC: www-rdf-logic@w3.org


Jim Hendler wrote:
> 
> >          DAML-S expressiveness challenge #1 -
> >
> >Is it possible (or is it likely to be possible in some future release of
> >DAML+OIL or DAML-L) to express the following constraint/restriction:
> >
> >    "The value of an instance of property P1, in some particular
> >context/scope/situation, must be the same as the value of an instance
> >of property P2 in that same context/scope/situation."
> >
> >What do I mean by context/scope/situation?  Well, that's up for
> >discussion, but to get us started, let's just say I mean "namespace".
> 
> David-
>   I'm confused - seems to me you are confusing language and processing
> of the language.  DAML-S seems to me to be a model of declarative
> specification - prescribing how something "binds" seems to me to be a
> specification as to how you process that knowledge.  Seems to me that
> if DAML-S wants to add this sort of processing semantics (which I
> don't object to in principle), it is separate from the DAML language
> per se -- I.e. you need a DAML-S processing spec -- see the XML
> Process spec work at W3 for an example

Jim -

Thanks for the comments, and for the reference to the XML Process spec.

To clarify, I was asking if a certain sort of constraint/restriction
might eventually be expressible in DAML+OIL or in DAML-L, when it
arrives.  I agree that what I'm asking for *might* be outside the scope
of both DAML+OIL and DAML-L.  If it is outside the scope of DAML+OIL and
DAML-L, then yes, that points towards additional mechanisms/semantics
that we, the DAML-S folks, would want to add.  But we certainly don't
want to re-invent the ability to express something that may soon be
expressible in DAML-L (in particular), and thus we thought it wise to
initiate some discussion about whether the sort of
constraint/restriction I described might be envisioned by the folks
working on DAML-L.

Further, it's by no means *obvious* that the expressiveness I'm seeking
goes beyond the boundaries of DAML+OIL or DAML-L (including their future
possible evolutionary paths).  Surely, expressing something having to do
with "bindings" and "scope" doesn't automatically take that expression
out of the realm of a declarative language.

Notice that I phrased my expressiveness requirement in a very general,
declarative way; that is, the ability to say something like this:

"The value of an instance of property P1, in some particular
context/scope/situation, must be the same as the value of an instance
of property P2 in that same context/scope/situation."

Is there something about the above that necessarily takes it outside the
realm of, say, DAML-L?

Regards,

- David
Received on Friday, 15 June 2001 13:41:13 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:52:40 GMT