W3C home > Mailing lists > Public > public-ws-semann@w3.org > January 2007

Scenario for SAWSDL

From: Karthik Gomadam <karthik.gomadam@gmail.com>
Date: Tue, 30 Jan 2007 12:09:19 -0500
Message-ID: <718a6e090701300909g1f2d1b31k15b5eb8da2fafbea@mail.gmail.com>
To: public-ws-semann@w3.org

Scenario for SA-WSDL working group

* The categorization of semantics used in Web services has been
discussed in detail in the following paper[1]. In a nutshell
this comes to four types of semantics  - data,functional , non
functional and execution , that can be attached to a web service that
describes different aspects of the service.

1. The thinking behind the SA-WSDL approach is that the functional
sematics of the Web Services need to be clearly attached to the
interface description of the service. Hence the SA-WSDL spec focuses
towards clearly visible annotations inside the WSDL document. As far
as functional semantics are considered, this thinking is correct.

2. However there could be non-functional semantics that need to
attached with the service description. For example it could be an
ontological reference that encapsulates what kind of security and
reliability a service requestor need to have [ We need to have a toy
ontology that displays this clearly]. However the important fact is
that this meta data is not related to the functionality of the service
but just the quality of service. The recommended way of attaching such
non-functional semantics is by using WS-Policy as a vehicle.

The essence of this is that it should be possible to attach semantics
through WS-Policy of which have no direct impact on the functionality
of the service. These metadata will have to be incorporated into
WS-Policy rather than being added to the original model reference.
Hence a more correct restriction (if it must be added to WS-Policy
attachment specification) would be

"Policy Assertions on interfaces, (or WSDL 1.1portTypes), operations
and messages (the abstract WSDL descriptions) SHOULD NOT describe the
functional semantics of messages or the action performed bythe
operations."

Note the change in the word "functional" rather than formal.

1. http://lsdis.cs.uga.edu/library/download/SVSM03-ICWS-final.pdf


-- 
Thanks and Regards,

Karthik Gomadam
Research Assistant,
Kno.e.sis Center,
Department of Computer Science,
Joshi Research Center,
Wright State University,
Dayton,
OH 45435

Skype       : namelessNerd
URL         : http://lsdis.cs.uga.edu/~gomadam
Google      : Karthik Gomadam / Gomadam
Blog        : http://namelessnerd.blogspot.com
Received on Tuesday, 30 January 2007 19:04:28 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 17 April 2012 12:14:27 GMT