Re: OWL-S and WSMO

Don't be confused - there is a big differentiation in this SWS-IG as
semantic-Web services (SWs) vs. semantic Web-services (sWS).

SWs, including OWL-S and WSMO, focus on the services on the semantic
Web, thus in nature, SWs are (semantic) Web-based services. Here are the
problems:

1. first of all, Carine should tell all people that SWs are not about
Web services specified by W3C, as I quoted many many times:

W3C said @ http://www.w3.org/TR/2004/NOTE-ws-gloss-20040211/ 

"Web service
There are many things that might be called "Web services" in the world
at large. However, for the purpose of this Working Group and this
architecture, and without prejudice toward other definitions, we will
use the following definition:

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). Other
systems interact with the Web service in a manner prescribed by its
description using SOAP-messages, typically conveyed using HTTP with an
XML serialization in conjunction with other Web-related standards."

Unfortunately, OWL-S declared clearly, from 2001 to 2006 that:

"Among the most important web resources are those that provide
services. 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. The Semantic Web should enable users to locate, select, employ,
compose, and monitor Web-based services automatically."

By service, it means a Web site or Web-based service. Thus the
so-called services in OWL-S is included in that "many things" rather
than Web services. WSMO claimed that it's almost the same as OWL-S with
some differences. Then, regretfully, WSMO is almost the same problematic
as OWL-S since it talks about semantic-Web (based) services. Such
conclusion has been demonstrated by SWs tutorials like:

http://www.wsmo.org/TR/d17/resources/200507-ICWS/SWStutorial-iswc05.ppt


In this typical tutorial,  the focus of OWL-S and WSMO is VTA (slide
#63), a Web-based services, rather than FlightBooking or HotelBooking -
the two real WSDL Web Services.

For semantic-Web (based) services, or services on the semantic Web, the
interest is in VTA for service aggregation or mediation.

For semantic Web-services, however, the interest is in FlightBooking
(e.g. US Airway) or HotelBooking (e.g. Hilton) - the two real WSDL Web
Services. Then Hilton has no interest, responsibility and right to
handle service issues of US Airway. That's a big difference.

For this SWS-IG, it's clearly stated that "The Semantic Web Services
Interest Group is part of the Web Services Activity", not part of
semantic Web activity. Regretfully in that tutorial, there is little
about those two Web-services, but full of service aggregation and
mediation - with little relation with the semantics of those two real
Web-services.

2. OWL-S is based on assumption - although Parry denied such a fact in
the past month.  Please see this article - "Bringing Semantics to Web
Services: The OWL-S Approach" @
http://www.cs.cmu.edu/~softagents/papers/OWL-S-SWSWPC2004-final.pdf in
which it clearly stated that "OWL-S is based on the assumption that the
definitions of these semantic concepts are available at referenced URIs
on the Semantic Web, ...". 

Where is the semantic Web? It's not in existence yet, while McCool said
in IEEE Internet Computing that Semantic Web failed and will fail as OWL
is incapable to do certain kinds of jobs, although such a conclusion may
result in controversy. 

As for the automatic service discovery, matchmaking, composition and
excecution, it is more problematic for OWL-S as you see:

- there is no real Web service registry, then where can service
discovery be implemented? SWs may have a solution as they are Web-based
services and then they may search on the Web?
- there is no real Web service, let alone service semantics (McCool,
2005/2006), then how can service matchmaking be implemented? 
- OWL-S has been focusing on service composition based on its
assumption, without the support of a service registry (and matchmaking
what?). OWL-S ends with its composition document because:
- automatic service invocation/excecution is almost impossible at this
moment, according to Dr. Burstein's envision on the dynamic invocation
of Web services - "without any reprogramming, a software system could
have the flexibility to use various services that do the same kind of
job but have different APIs" - don't tell me you can invoke such
funtions like the meaningless "stockquote".

Semantic-Web (based) services (SWs) has been misleading this SWS-IG for
years, because such approaches started from a wrong concept, then it
cannot lead to a correct result. SWs had little interest in the real Web
services but almost all their enthusiasm is on SW, not WS. However,
SWS-IG is part of the Web Services Activity. To handle service semantic
issues, SWs have to remove those components of service
aggregation/mediation, etc. which has little relationship with "a" Web
service.

For all of us who are pursuing or obtained a higher degree, we know
such a basic fact: whenever anyone uses/used a wrong concept in the
research, the first response from the committee is to ask this guy to
make correction first before moving step further, although such
correction may take months or more than a year. It's the same to those
SWs in this sWS community - make corrections first to comply with W3C
specification on WS, not any Web-based service, or Web site. Otherwise,
people will be confused and misled all the time.

Regards,

Xuan


>>> Carine Bournez <carine@w3.org> 9/13/2006 7:17 AM >>>

On Wed, Sep 13, 2006 at 04:52:02PM +0800, merrychang79 wrote:
>  
> Hello Everybody,
>  
> Both OWL-S and WSMO could be used to annotate the service to realize
the automatic service discovery, composition and excecution.
> But which one will be better or promissing as a standard upper
ontology for service?


Merry,

You may find useful information in the report from the workshop held in
2005.

http://www.w3.org/2005/04/FSWS/workshop-report.html 

The SAWSDL Working Group has been created to work on a semantic
annotation
mechanism in WSDL 2.0, see [1].
A charter for a characterization Group [2] had been written and no
clear 
support for further work has been expressed so far.


[1] http://www.w3.org/2005/04/FSWS/workshop-report.html 
[2] http://www.w3.org/2005/10/sws-charac-charter.html 

-- 
Carine Bournez -+- W3C Europe

Received on Wednesday, 13 September 2006 15:09:43 UTC