Web service in WSDL vs. "service" in OWL-S

Carine,

I just checked relevant documentations on W3C Web site and found out
that the definition of "Web service" defined by W3C is different from
the same terminology defined by OWL-S.

In Web Services Architecture (http://www.w3.org/TR/ws-arch/) W3C defined
that "A Web service is a software system designed to support
interoperable machine-to-machine interaction over a network. It has an
interface described in a machine-processable format (specifically WSDL)"
- a service refers to a WSDL based agent.

In OWL-S: Semantic Markup for Web Services
(http://www.w3.org/Submission/OWL-S/), however, OWL-S people said "By
'service' we mean Web sites that do not merely provide static
information but allow one to effect some action or change in the world,
such as the sale of a product or the control of a physical device." - a
service is a Web site.

Such difference can clarify many vague terminologies used in OWL-S, as
in fact, a Web-based service provider may actually a WSDL-based service
requester. For this reason, the semantics in OWL-S may actually have NO
relationship with WSDL-based service but rather, are something about
Web-site related events.

If W3C and this SWS-IG try to define service semantics for WSDL-based
service, other than Web-site based service, people have to re-examine
the suitability of OWL-S for SWS because OWL-S targets at a wrong object
(Web site) other than Web service defined by W3C. 

While I have been insisting on WSDL based Web service semantics, many
people could not understand why I keep debating and challenging OWL-S,
etc. By now, I have to ask you, Carine, whether you and W3C really care
about WSDL-based Web services or Web-site related Web services as
suggested by OWL-S? If you tell me a definite answer that you and W3C
are interested in Web-site related Web services, I will not debate with
OWL-S people in this IG again as we are talking about different things.

The current status of SWS (WSDL based service) is NOT optimistic - if
such SWS aims at enabling dynamic (WSDL based service) service
discovery, matchmaking, composition and invocation, then we can see
that:

1. there is NO real service registry yet, so how can SWS people find the
service?
2. there is NO formal service semantics defined, so how can SWS people
do matchmaking?
3. if service composition (integration/aggregation/mediation) is not
based on assumption, how can SWS people work on this stage before they
develop a real registry and service semantics?
4. dynamic invocation is impossible as SWS keep developing services that
do the same thing but have different APIs.

Regards,

Xuan

Received on Tuesday, 1 August 2006 04:08:30 UTC