W3C home > Mailing lists > Public > www-tag@w3.org > April 2005

Re: XML validity and namespaces

From: <noah_mendelsohn@us.ibm.com>
Date: Mon, 4 Apr 2005 22:42:33 -0400
To: Robin Berjon <robin.berjon@expway.fr>
Cc: www-tag@w3.org
Message-ID: <OF0E4BF008.36082F62-ON85256FDA.000BB9B8-85256FDA.000EE2C3@lotus.com>

Robin Berjon writes:

> And is it not one of the greatest ironies of XML
> Schema that it failed to learn from that lesson
> and therefore didn't provide simple and
> straightforward means to describe extensibility in
> schemata?

It is indeed, though to be fair, there was and to some degree still is a 
great deal of disagreement as to what evolution strategies people wanted 
to use for their instances.   The job of schema is to make it easy to 
describe evolving constraints on those instances, and the community was 
nowhere near consensus on what idioms were to be described.  For example, 
there was strong belief on the part of some that even a small change (bug 
fix?) to a language would result in complete republication putting either 
(a) the entire language into a new namespace or (b) at minimum the new 
constructs into a new namespace.   We have since seen important XML 
languages that do neither (a) nor (b).

In 1999, I suggested to the XML plenary that the XML community as a whole 
should consider just these questions  [1], in part so that XML schema 
would have a context in which to explore the requirements.  The response 
from the CG was basically:  versioning is a known hard problem and if we 
try to tackle it, we'll spend more time and energy than we can afford. 
Let's not try now.  I don't think it's fair to blame only the Schema WG 
for that decision.  Note that the current schema wg charter specifically 
makes versioning  a priority [2], and a very serious effort is being made 
to explore use cases and requirements, and then to develop suitable 
support in Schema 1.1.

So, ironic as it may appear now, there was a conscious decision by the CG 
not to make a frontal assault on versioning for the community as a whole. 
There were scattered attempts be Schema WG members to add features that 
might be helpful.  Wildcards, substitution groups and the <redefine> 
mechanisms are all examples of mechanisms that provide at least limited 
support for certain versioning idioms.   The WG is now, belatedly, 
involved in a much more serious attempt to understand the requirements and 
to define more robust mechanisms to support versioning.


[1] http://lists.w3.org/Archives/Member/w3c-xml-plenary/1999Oct/0019.html

Noah Mendelsohn 
IBM Corporation
One Rogers Street
Cambridge, MA 02142
Received on Tuesday, 5 April 2005 02:42:45 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:32:45 UTC