W3C home > Mailing lists > Public > public-ws-semann@w3.org > September 2006

Re: Review and publishing the Usage Guide

From: Rama Akkiraju <akkiraju@us.ibm.com>
Date: Wed, 27 Sep 2006 20:23:51 -0400
To: Joel Farrell <joelf@us.ibm.com>
Cc: Brahmananda Sapkota <brahmananda.sapkota@deri.org>, SAWSDL WG Private <member-ws-semann@w3.org>, member-ws-semann-request@w3.org, SAWSDL public list <public-ws-semann@w3.org>
Message-ID: <OF4FF444FD.F24C5238-ON852571F7.00014ACC-852571F7.00022F26@us.ibm.com>


Thank you for your helpful comments. I have incorporated most of them in
the latest SAWSDL Usage Guide. Please see my comments below in detail.

Rama Akkiraju

member-ws-semann-request@w3.org wrote on 09/25/2006 01:05:45 PM:

> Hi Rama, Brahmananda,
> Here are some comments
> Section 1 Introduction, I think creation of mediation code should
> also be mentioned in the life-cycle discussion.

Good point. Done.

> Just after Listing 2-4, "The annotation can be associated with the
> interface a Web service as follows" should be "The annotation can be
> associated with the interface of a Web service as follows" (missing word

Thanks for catching this. Done.

> As we found in the f2f in Turin, not everyone immediately relates to
> the concept of a web service representing the requirements you want
> to search for. I think the concept should be explained a bit more in
> the beginning of section 1.2 and section 3.

Done. Section 1 is revised to reflect this.

> Also, the "Web service"
> one matches against is really a "Web Service Interface" or perhaps a
> better term is "WSDL Interface." This terminology change would also
> help to clarify the idea. The WSDL interface is created and
> annotated to codify the requirements for the service you are looking
> for. Then, I think the document should say that these requirements
> could be represented some other way, like semantic data, but the
> references to the concepts in the semantic model would still be the
> same. But, for simplicity, our examples will use this annotated WSDL
> interface technique. I would stay away from the term "Web service
> matching" since that will imply to many readers the matching of
> aspects of the service implementation, which gets into policy, etc.

Terminology also changed per suggestion.

> In the spec, we changed all references to "domain model", to
> "semantic model." The same should apply to this document.


> We had also talked about including best practices. I don't think there is
> experience yet for true "best practices" but some guidance could be
> included, especially where there are conceptual reasons for doing
> something one way versus another. These are items we could start
> collecting from the group.

Good idea. You may recall that at the second face-2-face meeting we have
agreed to take out best practices from the usage guide until such time as
when we actually have a good list of best practices. Not all of us were in
full agreement with the one we had at that time which talked about not use
the ontology deficiencies as an excuse to adding multiple references to an
element. So, we decided to take it out.

> I congratulate the editors. This document has come a really long way
> and will be very helpful to readers of our spec.

Thank you.

> Regards,
> Joel
> member-ws-semann-request@w3.org wrote on 09/24/2006 10:10:37 PM:
> >
> > All,
> >
> > Brahmananda and I made many updates to the SAWSDL Usage Guide in
> > preparation for Last Call publication. Please review and give your
> > before Tuesday's call so that we can incorporate them in the spec in
> > for the .last call. Especially your help in verifying section 4 (SPARQL
> > XSLT transforms) would be appreciated.
> >
> > Regards
> > Rama Akkiraju
> >
> >
> >
Received on Thursday, 28 September 2006 00:24:07 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 19:58:46 UTC