xml-dev - RE: XHTML DTDs failing in IE5

[Mark Birbeck <Mark.Birbeck@iedigital.net>]

Andrew Layman wrote:
> In this list and in others, there have been strong and conflicting
> opinions expressed on what would be the best behavior of a parser, and
no
> settled standard exists.  I believe that people will continue to post
> opinions on what a parser should do.

Sorry. If you were trying to pre-empt more debate ... too late, I've
sent it off!

> I do not, in fact, know that the IE5 rules are the ultimately best
rules;
> what we can say in their favor is that because they are conservative,
> they do not encourage the creation of documents that are accepted by
> MSXML and yet are later rejected by conformant parsers after standards
> have been worked out.

I'm sure that raised a few smiles!

As it happens, your approach is worse than useless. If in the future the
value of 'xmlns' is actually part of determining whether a document is
valid or not, then so be it. DTDs will need to change. But that also
goes for XLink, fragments, XML-Data and more. The fact is that as far as
I can tell, at the moment there is nothing in XML 1.0 or the namespace
spec that says the value of a namespace MUST be fixed. Yet because
Microsoft have decided they should be - and many may agree - I am now
unable to get perfectly valid DTDs, over which I have no control, past
your parser. This is not conservative ... it is WRONG.

What really amazes me is why you have picked this feature to be
restrictive on. Microsoft have happily issued an implementation of
XML-Data that we all know will have to change, as well as an
implementation of XSL that has changed already and will have to change
again. I don't mind that - I accept that things are still evolving. In
fact I'm very grateful for having the chance to build a test system
around XML-Data, even though I know it will all change. It's even got to
the point where for our own purposes I don't even use DTDs anymore - I
automatically generate schemas from my database and use your excellent
parser.

But your misplaced paternalism is now preventing me from doing real work
with information from people who still DO use DTDs - like the W3C, for
example! I want to apply XSL to XHTML, and I can only think that to do
it I will have to change parsers (unless you tell me there is a secret
switch somewhere, to turn this ridiculous 'feature' off).


Mark Birbeck
Managing Director
IED Ltd.
220 Bon Marché Centre
442-444 Brixton Road
London
SW9 8EJ
w: http://www.iedigital.net/
t: +44 (171) 501 9502
e: Mark.Birbeck@iedigital.net

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo@ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo@ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa@ic.ac.uk)