W3C home > Mailing lists > Public > xml-dist-app@w3.org > September 2003

Re: Data model task force recommends adoption of data model formulation

From: Rich Salz <rsalz@datapower.com>
Date: Tue, 16 Sep 2003 21:50:03 -0400 (EDT)
To: "noah_mendelsohn@us.ibm.com" <noah_mendelsohn@us.ibm.com>
Cc: "xml-dist-app@w3.org" <xml-dist-app@w3.org>
Message-ID: <Pine.LNX.4.44L0.0309162126520.17885-100000@smtp.datapower.com>

> > One thing to consider is the amount of "extra reading" being put on the
> > SOAP user and development community.  Less is better. :)
>
> Indeed, but to what conclusion does that lead us?  I had some hope that
> you and others from the DSig community would find the DM approach
> attractive, in part because your own c14n [1] calls on readers to become
> familiar with the earlier version of the XPath data model.  Though it does
> not appear urgent, I would have thought that the natural evolution would
> be to move c14n toward the XPath 2.0 model.  So, what is "extra reading"
> depends a bit on what you were already reading.

I tried to read the XPath 2.0 DM document.  It's a last call draft
whose comment period ended several weeks ago.  I would have thought
the experiences of SOAP 1.1 and XML Schema draft-ness would have
made everyone desperately want to avoid that again. :)  I only
skimmed the document, but there were enough references to the
"Formal Semantics" and "Operators" documents that I think those would
really have to be added to the reading pile.

The DSIG and Encryption WG's are closed, finished, Katy bar-the-door done.
The only way they are going to migrate is if new WG's get created.
I doubt that will ever happen, if only because the security folks are
moving on to higher-level specs like WS-Security et al, and the WS-I's
security profile WG.  I'd be surprised if many crypto folks see a
compelling need to talk about data models, when crypto is all about bytes
and serialization, not abstract data.

> I can see this one either way.  SOAP is Infoset.  Schema is Infoset. Query
> is XPath 2.0/Query 1.0 DM, XML C14N is XPath 1.0 DM.  I think we're
> getting near the point where all of these should come together.

Perhaps.  But I don't need any of that 2.0 stuff to do SOAP or secure SOAP.
I can even pretend it doesn't exist and never be impacted.

> I do take
> the point that, typing issues aside, Infoset is clearly the most natural
> model in which to discuss SOAP, at least for the forseeable future.

Thank you.  So you agree that references to the XPath 2.0 DM should
be excised.

        /r$

--
Rich Salz                  Chief Security Architect
DataPower Technology       http://www.datapower.com
XS40 XML Security Gateway  http://www.datapower.com/products/xs40.html
XML Security Overview      http://www.datapower.com/xmldev/xmlsecurity.html
Received on Tuesday, 16 September 2003 21:50:04 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:59:15 GMT