W3C home > Mailing lists > Public > xml-dist-app@w3.org > September 2000

RE: Removal (Time for XMail?)

From: David Orchard <orchard@pacificspirit.com>
Date: Fri, 29 Sep 2000 08:50:00 -0700
To: <Noah_Mendelsohn@lotus.com>, "'Ed Mooney'" <Ed.Mooney@Sun.COM>
Cc: <bigor@infolio.com>, <cagle@olywa.net>, <xml-dist-app@w3.org>
Message-ID: <006701c02a2c$ed975760$2a19e8d8@bc.hsia.telus.net>
Noah,

I completely agree with your concern about packaging of XML components
within an XML container.  Although I believe the plenary decided to
discontinue plans for a packaging wg, shelving the package so to speak.
Your suspicions are facts.  We have problems using SOAP 1.1 for shipping XML
documents around, in our case valid WML and XML Schema files.  CData
encoding and/or base64 encoding are hacks, IMHO.

The problem of using 1 syntax to contain instances of that same syntax is
occurring in a few places XML.  The issues surround the data model for the
"xmlns" quasi-attribute are one example.  When is it an attribute and when
not?  The same occurs for special elements, say an <xml:include /> element.
I'm reminded of Godel and issues around circular reference.

I'm sure you have encountered these issues before, perhaps in Domino, OLE or
OpenDoc.

I look forward to the debate about the requirement of xml protocols for a
generalized container model.

Cheers,
Dave Orchard

-----Original Message-----
From: xml-dist-app-request@w3.org [mailto:xml-dist-app-request@w3.org]On
Behalf Of Noah_Mendelsohn@lotus.com
Sent: Friday, September 29, 2000 7:11 AM
To: Ed Mooney
Cc: bigor@infolio.com; cagle@olywa.net; xml-dist-app@w3.org
Subject: Re: Removal (Time for XMail?)


I have a suspicion, not carefully considered, that ID= attributes can
cause problems when XML is used as a generalized container for other XML.
For example, let's consider the case where your e-mail has several
attachments, each of them XML, and they make conflicting use of the same
ID names.  As best I can tell, you can work around this as long as you're
careful about what you validate and how, but as I say, I suspect there are
some messy edge conditions here.  Certainly there are likely to be
problems with any tools that take the container document as a whole and
try to blindly interpret ID attributes.

I have had this same concern for SOAP, for example, insofar as it serves
as a generalized packaging framework for assembling XML messages.  Not a
fatal problem, I think, but probably something that deserves a bit of
thought.

------------------------------------------------------------------------
Noah Mendelsohn                                    Voice: 1-617-693-4036
Lotus Development Corp.                            Fax: 1-617-693-8676
One Rogers Street
Cambridge, MA 02142
------------------------------------------------------------------------
Received on Friday, 29 September 2000 11:47:59 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:58:57 GMT