Re: OWL-S process preconditions

On Jun 18, 2004, at 12:09 PM, Dónal Murtagh wrote:

> On Jun 18, 2004, at 10:00 AM, Dónal Murtagh wrote:
>
>>> Greetings,
>>>
>>> My understanding is that OWL-S does not currently mandate any 
>>> language
>>> for
>>> expressing preconditions, although the current frontrunner is SWRL.
>>
>> OWL-S 1.1 will, by default, make use of conjunctions of SWRL (like)
>> atoms. It will allow for other preconditions languages (like KIF).
>
> I didn't realise there is an OWL-S 1.1, the latest release listed at
> http://www.daml.org/services/owl-s/ is 1.0.

1.0 is the latest release. Hence the use of future tense :) 1.1 is due 
any day now and I would have included the URL for the draft release but 
there are big letters at the top of the page:

	 Please note: this is a DRAFT release, and HAS NOT YET BEEN ANNOUNCED. 
Please do not distribute the URL of this page.

So, I won't.

>  Any idea where I can find
> information about this new version?

A bit of simple and obvious URL hacking, OTOH, will take you to the 
promised land.
[snip]
>> There is a presumption that there is a knowledge base -- e.g., a
>> planner's "state of the world" -- against which preconditions are
>> evaluated and effects applied. That kb could do infogathering on
>> demand, I suppose.
>
> Sure, but my question is really about what is initially in the KB, is 
> it
> just that information which the invoking agent has explicitly supplied 
> (to
> the process)?

That's application dependent. The OWL-S API, I think, let's you supply 
whatever information you'd like.

[snip]
> OK, so a precondition should be expressed as a conjunction of SWRL 
> atoms,
> rather than a SWRL rule (with an antecedent and consequent). By the 
> way, are
> there any examples available of SWRL being used to express process
> preconditions (or indeed any other OWL-S conditional expressions).

I believe the standard examples are being augmented with example 
preconditions, etc. Our lab should have some of the standard planning 
domains ported to OWL-S by release time.

Cheers,
Bijan Parsia.

Received on Friday, 18 June 2004 12:28:38 UTC