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

Re: [owl-s] communication between web services

From: Massimo Paolucci <paolucci@cs.cmu.edu>
Date: Thu, 09 Sep 2004 10:42:48 -0400
Message-ID: <41406BE8.7080100@cs.cmu.edu>
To: jean-michel nougayrede <nougay_j@epita.fr>, public-sws-ig@w3.org

jean-michel nougayrede wrote:

>I'm working on web services that can communicate between each others without
>knowing the other ones.
>For example there is a web service A which sells a product P and then
>connect to another web service B in order to ship the product P. But the web
>service A doesn't know the web service B which can do the shipping. The web
>service A searches on the web in order to find one, and when it finds one,
>it uses it. But the problem is that the web service A doesn't know what are
>the functions to be called and with which arguments and in which order.
>When I read the white paper DAML-S it seems that it is possible to
>automatize the execution of web services but I don't understand the process.
>How can the web service A understand what function it has to call and which
>are the arguments on the web service B?
In a nutshell, in OWL-S (or DAML-S that preceded it) the execution of 
the Web service is controlled by a Process Model that describes what 
information service B needs to execute correctly.  In turn processes in 
the Process Model map into WSDL operations that can become remote 
function calls.
There are a bunch of papers that describe how this can be done,  you can 
find them at http://www.daml.org/services/owl-s/ or on the Web pages of 
the OWL-S coalition members, as well as in the proceedings of 
conferences such as the Semantic Web Conferences, World Wide Web 
Conference or the conference on Web services (ICWS).

>Must web services share the same language (in our case all the web services
>for the shipping have the same function and argument)?
OWL-S assumes that they share the same ontology, that is a set of terms 
and relations between them, and that they share a proof theory to make 
same derivations given the same knowledge.  In practice,  if service A 
sends to B a purchase order, than B should understand that it is an 
order to purchase exactly the things that A wants to buy.  Finally, the 
ontology should be written in OWL or one of its derivatives, such as 
SWRL (although with some sweet talk, and a few beers, I may be convinced 
that is more general than that.)

>Thanks a lot for your help.
I hope that helped.

--- Massimo

Received on Thursday, 9 September 2004 14:43:41 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:32:46 UTC