W3C home > Mailing lists > Public > public-sws-ig@w3.org > November 2004

Re: Relation between OWL and OWL-S

From: Drew McDermott <drew.mcdermott@yale.edu>
Date: Wed, 24 Nov 2004 14:24:25 +0000 (GMT)
To: public-sws-ig@w3.org
Message-ID: <E1CWy4n-0004DA-Ag@frink.w3.org>


> [Daniela CLARO]
> 
> If I have a web service called bank, what will I use to discover this
> service?
> -My answer: I will use the description in the profile and also the input > and
> output parameters
> 
> But if I have another service called bank, so how will I make the differe> nce
> between them? Suppose that the first one refers to economical bank and th> e
> second one is a wood bank for sitting on.
> 
> In OWL is clear that we can make this difference between the hierarchical
> view proposed, so the economical bank will be localised under economical
> tranche and wood bank will be localised under wood section like my model
> below:
> 
> 		     ...
> 		 _____|_____
> 		|           | 
>   	  economical     wood
>             |           |
>            bank        bank
> 
> Using that model in OWL we can make this distinction, and how can I do th> e
> same thing to discover a service in OWLS.
> 
> If I have a service called bank, how can I make this distinction?

It's hard to answer your question, because it mushes together several
distinct questions.  Let me try to separate them --

Q1: How does one avoid the ambiguity between different senses of the
word "bank" in the innards of ontologies?  The answer would be
namespaces.  (By the way, the banks you sit on are only banks in
Italian; in English they are "benches," which comes from the Italian,
but conveniently changed spelling, thus eliminating the confusion.  To
compensate, the earthen walls of rivers are called "banks," so if we
have a web service selling tours of river margins, the confusion is
restored.)

Q2: How does one deal with the ambiguity of words in HCI for web
services?  Good question, also asked by Charlie Abela.

Q3: Must we do without Owl mechanisms such as inheritance when we use
Owl-S?  The answer is No!  Owl-S is an _application_ of Owl, and
doesn't replace a hair on its pointy little head.

Q4: Is there a problem with multiple inheritance if an agent is both
a service _and_ an "economical entity"?  I don't know, but I don't see
why there would be a problem.

Does any of these alternative questions come close to what you
intended?  

-- 

                                         -- Drew McDermott
                                            Yale University
                                            Computer Science Department
Received on Wednesday, 24 November 2004 14:24:57 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 16 March 2008 00:10:58 GMT