Re: Commercial/Real-world Semantic Web Services?

The Open Geospatial Consortium (OGC) may be a "good" example for reference. OGC "is an international industry consortium of 332 companies, government agencies and universities participating in a consensus process to develop publicly available interface specifications. OpenGIS® Specifications support interoperable solutions that "geo-enable" the Web, wireless and location-based services, and mainstream IT. The specifications empower technology developers to make complex spatial information and services accessible and useful with all kinds of applications."

Thus, it is a multi-enterprise interaction. OGC standards/specifications enable the industrial wide interoperability issues, both semantically and technically with little logic modelings regretfully :-) But as a shared, common conceptualization, it resovles industrial wide, domain specific semantic issues, in the geospatial industry. Even Microsoft has to pay attention to OGC specs and its TerraServer Web services have to comply with OGC specs. Google is of course another example as we can find many samples that combine google map API with OGC compatible applications. So why some people just don't believe that we may reach an agreement to do something before they ever try to generate an agreement?

More broadly, "The World Wide Web Consortium (W3C) is an international consortium where member organizations, a full-time staff and the public work together to develop standards for the World Wide Web. ". Internet and Web have been based on standards, protocols, agreements for communication among heterogenerous platforms, programs, OS, etc. SWS-IG should not ignore such a fact but have to consider how to realize the pre-requisite specified in W3C documentation "Web Services Architecture" - The important point is that the parties MUST AGREE on the semantics of Web services, regardless of how that is achieved.

Regards,

Xuan

>>> "Sergey V. Mikhanov" <sergey.mikhanov@gmail.com> 10/11/2006 11:37 AM >>>
Greetings!

Thanks for very detailed answer.
Anyway, despite the fact that most of your arguments are true in
general, I still hope that it is possible to point out some successful
attempts to bring SWS to:

* SOA in the scope of one large enterprise (obvoiously, it's easier to
negotiate the use of one ontology on one enterprise). It is promised
to be more effective solution for integration tasks than to use
mapping in BizTalk or whatever else
* Web Services, which are used in the scope of some specific task
(consider Telecom Use Case from DERI as an example:
ieeexplore.ieee.org/iel5/10636/33570/01595724.pdf?arnumber=1595724)

Could anyone extend this list with more examples?

Thanks,
Sergey.

> I agree, semantic Web services (SWS) and this IG have nothing to boast,
> as I indicated in
> http://lists.w3.org/Archives/Public/public-sws-ig/2006Sep/0018.html 
>
> By examining W3C documentation "Web Services Architecture" again @
> http://www.w3.org/TR/ws-arch/ we can see why SWS failed and had
> little accomplishment and progress in the past years.
>
> How to realize and implement (semantic) Web services? W3C said clearly:
> "the requester and provider entities must agree on the semantics and the
> service description that will govern the interaction between the
> requester and provider agents, but it would be more accurate to say that
> they simply need to have a congruent or non-conflicting view of the
> semantics and service description of the interaction."
>
> (Semantic) Web Services Architecture has to be based on "agreement" -
> if anyone in this SWS-IG would like to read this document again, just
> count the number of the repeated word "agree" used in this W3C
> document.
>
> However, the leading roles of this IG believed that NOBODY wants to
> agree with each other, as every developer or service provider has the
> absolute right to do what s/he wants to do. For this reason, they have
> to use varied kind of logical modeling to guess which one might be
> similar to the others, by referencing each individual annotated semantic
> definition to a super-ontology.
>
> Unfortunately, ontology, again by definition, is a shared, common
> conceptualization of a domain knowledge (or again a kind of
> agreement/standard). Then we see, those who CANNOT reach an "agreement"
> have to "share" a super-ontology. This means, after turning around and
> around through modeling, we return to the starting point - we have to
> "agree" something first. But the problem is, referencing to a
> super-ontology promotes the dissemination of individual "semantic"
> definition on varied service and interface, and this means such people
> just do NOT use that "shared" ontology of a domain of the service.
>
> Why people do NOT use that "shared" super-ontology of a domain of the
> service, in case there is such a super-ontology? Because they thought
> standard/agreement-based SWS "takes all the fun out of it", although
> they knew "That's certainly true" - "given enough clear information
> about web services", we can write any desired program for interacting
> with web services, because we reach an agreement first, then those
> artificially designed "agents" know what and how to do with little fun.

Received on Friday, 13 October 2006 18:41:59 UTC