RE: Minutes of the May 8th Conference Call

[.2] http://lists.w3.org/Archives/Public/www-ws-desc/2003May/0020.html
<http://lists.w3.org/Archives/Public/www-ws-desc/2003May/0020.html> 

... NO IPR issue exists. Two new elements are proposed to be added,
(supports and requires) that may be used to indicate QoS. Phillippe
thinks it is more in the scope of WS-Policy assertions. Jonathan asks
what the wg should do with this document. Umit suggests we may cover it
when we discuss properties and features in the f2f. Jonathan will add it
to the agenda. 

 

 

Just to reaffirm the above remark... As far as I am concerned, there is
no IP issue with my organisation (University of Newcastle upon Tyne). I
think the same applies with Jim's company (Arjuna Technologies Ltd.) but
I'll let Jim confirm that.

 

The main idea behind the proposal for the "requires" and "supports"
elements was to give Web service architects the ability to advertise
non-functional requirements for ports in a binding (and perhaps
individual operations). By making this kind of information part of a Web
service's description (something that WS-Policy does not do as the spec
clearly specifies), consumers of Web services will have a much better
idea of what it is expected of them when communicating with particular
ports. It is imperative that such information is part of the Web
service's description and not a separate document. Having said that,
there is no reason why WS-Policy elements couldn't appear inside the
"requires" and "supports" elements, if makes sense semantically.

 

Furthermore, the information in these two elements can act as the glue
when combining Web services together (for example in workflows) and also
can give hints to WSDL processors about what kind of technologies are
required (you can imagine WSDL processors that use appropriate SOAP
actors depending on the information being advertised although, I guess,
this is not a primary concern for this discussion).

 

After looking at the Properties and Features document, I realise that
the aims are similar but the approaches slightly different. For example,
I don't believe that just a URI is sufficient to expose a capability
(what does it mean to refer to
http://schemas.xmlsoap.org/ws/2002/08/wstx?). However, I understand that
this discussion should be moved to the appropriate mailing list and I am
looking forward to it.

 

Regards,

--

Dr. Savas Parastatidis 

Chief Software Architect, North-East Regional e-Science Centre 

School of Computing Science, University of Newcastle upon Tyne, UK 

http://savas.parastatidis.name