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

Re: [owl-s] using disjointWith for IOPEs

From: Yuzhong Qu <yzqu@seu.edu.cn>
Date: Fri, 17 Oct 2003 22:50:21 +0800
Message-ID: <004101c394bd$fd7467a0$fc0b77ca@xobjects>
To: "Monika Solanki" <monika@dmu.ac.uk>
Cc: <www-ws@w3.org>
Sorry, I have no example to illustrate the utility of such a declaration. But I think
    
    The classification and partition (disjointedness of classes) are generally useful for reasoning, suppose that the partitioning is correct.   

Maybe the deisgner(s) of the OWL-S 1.0 can give us more details.


Yuzhong Qu 
  ----- Original Message ----- 
  From: Monika Solanki 
  To: yzqu@seu.edu.cn ; www-ws@w3.org 
  Sent: Friday, October 17, 2003 10:37 PM
  Subject: Re: [owl-s] using disjointWith for IOPEs


   
  Yuzhong Qu wrote:

    1. These constructs say that

        Input, ConditionalOutput, Precondition and ConditionalEffect  are four pairwise disjointed. classes.

        It's worthy of noticing that 

        "A is disjointed with B, C and D" doesn't imply  "B is disjointed with  C or  D".
  This is not what I was implying. There has to be some thing common between the classes A, B, C and D, which could lead to their instances being used interchangeably and therefore a disjoint is necessary. I can see that for Input and ConditionalOutput which are subclasses of Parameter. I do not see that for the disjoint declaration for Input and other classes.Maybe youy can explain?


    2.  The utility of such a declaration seems useless, but it may become useful in some situation.

  Maybe you could give an example?

  Thanks,

  Monika


    Yuzhong Qu
      ----- Original Message ----- 
      From: Monika Solanki 
      To: www-ws@w3.org 
      Sent: Thursday, October 16, 2003 7:46 PM
      Subject: [owl-s] using disjointWith for IOPEs 


       I have some issues about the use of "disjointWith" for the IOPEs classes with the structure in OWL-S 1.0.  

      IOPE structure:
      * class Parameter 
      - properties:
          - parameterType 
      - subclasses:
          - Input
          -ConditionalOutput
          -properties: coCondition (ranges Condition)
              - subclass: UnConditionalOutput (maxCard=0 on coCondition)
      *class Precondition
          - properties:
          - preCondition(ranges Condition)

      *class ConditionalEffect 
          - properties
          - ceCondition(ranges Condition)

      Monika: Over here there is another property that should be added, since it is defined in the Ontology, 

      ceEffect(ranges Condition - as defined in the Process.owl)

      David & All, Are we sure that the range of ceEffect is a Condition?, I am confused [This can be the subject of another email, however I am still adding it here ]

             - subclass: UnConditionalEffect (maxCard=0 on ceCondition)

      * IOPE's are disjoint


      The question that bothers me, is under what conditions should one declare a class as disjoint with another class. In OWL there is nothing that restricts the use of disjointWith i.e any class can be declared disjointWith any other, although they may not really related in anyway!!!!. However, I believe classes should be declared disjoint , whenever there are chances of them being used interchangeably otherwise it leads to redundant declarations. 

      In case of IOPEs, for 1.0, for e.g, we have Parameter as a superclass of both Input and ConditionalOutput. Precondition and ConditionalEffect are independent entities. 

      I am confused by the following

      - <rdf:Description rdf:about="#Input">
        <owl:disjointWith rdf:resource="#ConditionalOutput" /> 
        <owl:disjointWith rdf:resource="#Precondition" /> 
        <owl:disjointWith rdf:resource="#ConditionalEffect" /> 
        </rdf:Description>

      I can understand Input being disjointWith ConditionalOutput, however I am confused by the last two disjointWith declarations. I believe the semantics of P and CE make them anyways disjoint from I and O. What are the chances that their instances would be used interchangeably for Input, since they do not have anything in common: subclass, domain or  range. I do not understand, why they should be a part of the disjointWith declarations for Input. Had they all been subclass of a common class say for e.g "Parameter" then it would make sense. I agree that there is nothing that stops me from declaring them as disjointWith, however I wonder about the utility of such a declaration. 

      I have a similar problem with 

      - <rdf:Description rdf:about="#ConditionalOutput">
        <owl:disjointWith rdf:resource="#Precondition" /> 
        <owl:disjointWith rdf:resource="#ConditionalEffect" /> 
        </rdf:Description>

      - <rdf:Description rdf:about="#Precondition">
        <owl:disjointWith rdf:resource="#ConditionalEffect" /> 
        </rdf:Description>

      Maybe, I do not understand the semantics of disjointWith. Please help 

      Thanks,

      Monika






  -- 
  **>><<**>><<**>><<**>><<**>><<**>><<**>><<** 
  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

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


Received on Friday, 17 October 2003 10:49:54 GMT

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