W3C home > Mailing lists > Public > www-xml-schema-comments@w3.org > January to March 2001

About XML Schema Structures Part

From: Murali Mani <mani@CS.UCLA.EDU>
Date: Mon, 19 Mar 2001 13:33:48 -0800 (PST)
To: <www-xml-schema-comments@w3.org>
Message-ID: <Pine.SOL.4.33.0103191311470.6505-100000@panther.cs.ucla.edu>

Hi,

I would like to mention my strong intuitions about XML Schema - Structures
Part. This is a slightly strongly worded mail, because I find that the
*errors are accumulating*, and no efforts are being made to correct the
errors.

First of all, I am not sure whether research and decision making is done
by the same group with regard to a schema language for XML, if yes, then I
think it will make the decision making body not so broad-minded to see the
good points of other work, sometimes, it might even make them narrow
minded to hide their mistakes, and not accept them. I would ideally like
the decision making body that decides what should be the standard to be
broad enough and technical enough to understand the merits of every
approach.

Regarding XML Schema, I think there already are sufficient proposals and
one of the existing proposals has to be endorsed as a standard, though of
course leaving room for some brilliant new idea. If it is done in pure
broad-mindedness, then I believe the current W3C proposal for XML Schema
has *several irreparable inherent deficiencies* which make it incompatible
to be a standard. In short, it fails short of several features required
for a schema language for DB applications. Also there are several
materials posted such as XML Schema is strictly more expressive than all
other existing schema language proposals, which is wrong. I would like if
such wrong claims are not made, in order that we come to a good schema
language for XML, which I believe and hope is the ultimate goal of the
decision making body.

From my personal research experience, I think the connection between
formal language theory and schema language for XML should be *very*
carefully studied. Also, research is usually co-operative because the
final goal is the same -- come to a good and correct conclusion.

<warning>speaking for himself only</warning>

regards - murali.
Received on Monday, 19 March 2001 16:34:15 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 6 December 2009 18:12:49 GMT