[Fwd: A URI for the class of SOAP MEPs]

Hi, response to issue 288 from XMLP WG.
Jacek

-------- Forwarded Message --------
From: noah_mendelsohn@us.ibm.com
To: Jacek Kopecky <jacek.kopecky@deri.org>
Cc: xml-dist-app@w3.org, public-ws-desc-comments@w3.org, www-tag@w3.org
Subject: A URI for the class of SOAP MEPs
Date: Tue, 21 Feb 2006 16:41:35 -0500

(resending with useful subject line)

Dear Jacek,

This note is in response to your query at [1], in which you asked:

> For modeling purposes, we want to say that the WSDL SOAP binding
> property soapMEP will point to any SOAP MEP, therefore we need to be
> able to indicate the class that represents SOAP MEPs. Currently, we use 
> http://www.w3.org/2005/10/wsdl-rdf#SOAPMessageExchangePattern
> for that, but this URI is WSDL-specific and that is wrong.
> 
> We intend to use instead the URI that points to the section 3.2 of SOAP
> 1.2 specification, which describes the concept of SOAP MEPs:
> http://www.w3.org/TR/soap12-part1/#soapmep
> 
> Can you please confirm that we may use this URI as the identifier of the
> RDFS class of all SOAP MEPs? Or if you don't like this, can you please
> provide us with a better URI?
> 
> We are asking you because the concept of a SOAP MEP is yours and
> therefore you should be the ones to coin the URI to identify that
> concept.

As best we can tell this query was never logged in XMLP Comments or 
tracked as a formal issue, but we have discussed it as we would such an 
issue, and this note is to communicate our intended resolution.

We note that the URI http://www.w3.org/TR/soap12-part1/#soapmep has to 
date been used not for the concept of MEPs, but for the section in our 
document that introduces the concept.  Though various members of our 
workgroup have differing feelings regarding the importance of this 
distinction, the workgroup has decided to create a separate URI to refer 
to the overall class of MEPs (your note was a bit vague on whether you 
wanted a URI for the class of MEPs, or for the concept, but I'm assuming 
that former).  This will be
http://www.w3.org/2006/02/soap12/abstractions#classOfMEPs. 

As you can see, the workgroup has accepted Tim Berners-Lee's suggestion 
that the URI include a fragment identifier [2]. The workgroup also debated 

Tim's suggestion that the representation of this URI be an RDF document. 
While we agree that such a document would be useful, we do not feel that 
the workgroup has the skills, the staffing or the mandate under our 
charter to do this work.  Accordingly, we plan that the new URI will 
resolve to (an anchor within) an HTML document that will briefly describe 
in prose the class of MEPs, and will include an appropriate reference to 
the SOAP 1.2 Recommendation.  We note that doing this does not seem to 
preclude subsequent development of RDF descriptions, or the use of content 

negotiation in the case that both the RDF and HTML were eventually 
deployed in parallel.  The XMLP workgroup at this point has, for the 
reasons stated, no plans to develop such an RDF description.

This mail will serve to initially document the creation of this URI, and 
we expect reasonably soon to provide an appropriate representation on the 
Web.  The XMLP workgroup may or may not at some future time include a 
description in this URI either in an  erratum to SOAP 1.2, or in a future 
version of SOAP. 

We hope this resolution is an acceptable to all concerned.  Since some 
discussion of this issue was held on www-tag@w3.org, I am copying that 
list as well.

Thank you.

Noah
(for the W3C XML Protocols Workgroup)

[1] http://lists.w3.org/Archives/Public/xml-dist-app/2006Jan/0103.html
[2] http://lists.w3.org/Archives/Public/www-tag/2006Jan/0066.html

--------------------------------------
Noah Mendelsohn 
IBM Corporation
One Rogers Street
Cambridge, MA 02142
1-617-693-4036
--------------------------------------

Received on Tuesday, 28 February 2006 12:53:16 UTC