Re: the precondition property in OWL-S 1.0

Suggestion:

In this case, the semantics should be deployed in the property itself, not in the range of the property.


Yuzhong Qu
  ----- Original Message ----- 
  From: Monika Solanki 
  To: Huhns, Michael 
  Cc: public-sws-ig@w3.org 
  Sent: Sunday, November 09, 2003 12:10 AM
  Subject: Re: the precondition property in OWL-S 1.0


  Hi Michael,

  The property that makes it special does not lie in the syntax but in the semantics. Precondition is a property which is required to be true, before the execution of the service. So there could be several formulae that could be classified as Conditions, however if any such formula is tagged with a qualifier that it is a precondition, it makes a difference in the interpretation of that Condition for the execution of the service.

  -Monika

  Huhns, Michael wrote:

Hi Monika,

If a Precondition is a special kind of Condition, then it must have some
property (or a restriction on some property) that makes it special.
What is this?  That is, given a logical formula that evaluates to true
or false, what property does it have that would enable you to determine
whether it is a Condition or Precondition?

Cheers,
Mike Huhns

-----Original Message-----
From: public-sws-ig-request@w3.org [mailto:public-sws-ig-request@w3.org]
On Behalf Of Monika Solanki
Sent: Saturday, November 08, 2003 5:15 AM
To: David Martin
Cc: public-sws-ig@w3.org
Subject: Re: the precondition property in OWL-S 1.0




David Martin wrote:

  
[Note: this thread is moving to public-sws-ig.  After this message,
replies should only be sent to public-sws-ig.]

Marta Sabou wrote:

    
 Monika, Drew,

 I completely agree with the proposed solution. This should
definitely be
implemented in the next release.
      
I also agree, except note that there are 2 proposals "on the table":

I think Monika suggested this:

    Process - hasPrecondition - Precondition
where Precondition is a subclass of Condition

whereas Drew seems to be suggesting this:

    Process - hasPrecondition - Condition

(with no Precondition class anywhere).
    

I think we should retain the Precondition class and the hasPrecondition 
property. This is because, although Precondition is effectively a 
Condition, however it is a "special" kind of Condition. In the process 
model, Condition is a general thing, which  is also  used for 
Conditional Effects and Conditional Outputs.

  
That reminds me - we still have an open issue about the class of an
effect (that is, the range of ceEffect).  Currently it's just "Thing",
    

  
which isn't very satisfying. Do people feel that it's OK to have 
Condition for this range, or do we need something distinct?

- David

    
 Cheers,

Marta


  Drew McDermott wrote:

      
  [Monika Solanki, in re DAML-S spec]
  Currently in the 1.0 version of the process model, we have the
following

    <owl:Class rdf:ID="Precondition" />
  - <owl:ObjectProperty rdf:ID="preCondition">
    <rdfs:domain rdf:resource="#Precondition" />
    <rdfs:range rdf:resource="#Condition" />
    </owl:ObjectProperty>

  Where, Condition is defined as,

  - <owl:Class rdf:ID="Condition">
    <rdfs:comment>This is a "place-holder" for now, which awaits
further
  work from the DAML/OWL community. An instance of Condition is a 
logical
  formula that evaluates to true or false. Eventually we expect this
        

  
to be
  defined elsewhere, as part of a OWL extension allowing for logical
  expressions.</rdfs:comment>
    </owl:Class>

  Somehow I am not able to grasp the utility of the property
  "preCondition".  Since in this model, we have IOPEs as Classes,
  therefore I believe all we need to do is make Precondition a
subclass of
  Condition.

   <owl:Class rdf:ID="PreCondition">
    <rdfs:subClassOf rdf:resource="#Condition" />
    </owl:Class>

  The property preCondition, introduces redundancy as it is ranging
over
  the class "Condition" anyways.

  Feedback appreciated in case I missed something.

You're right, as far as I can see.  The important property is 
hasPrecondition, which connects a Condition to a Process or Process 
step.  There is no reason for the class Precondition to exist, let 
alone a property preCondition linking a Precondition to a Condition 
(itself?).

-- 
                                            -- Drew McDermott
                                               Yale University CS 
Dept.
        

      

  


  -- 
  **>><<**>><<**>><<**>><<**>><<**>><<**>><<** 
  Monika Solanki
  Software Technology Research Laboratory(STRL)
  De Montfort University
  Hawthorn building, H00.18 
  The Gateway 
  Leicester LE1 9BH, UK 

  phone: +44 (0)116 250 6170 intern: 6170 
  email: monika@dmu.ac.uk 
  web: http://www.cse.dmu.ac.uk/~monika

  **>><<**>><<**>><<**>><<**>><<**>><<**>><<**