W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > January to March 2005

Re: Serving XHTML as XML

From: Jim Ley <jim.ley@gmail.com>
Date: Thu, 24 Feb 2005 09:36:24 +0000
Message-ID: <851c8d3105022401367d66ea7e@mail.gmail.com>
To: w3c-wai-ig@w3.org

On Thu, 24 Feb 2005 10:17:05 +0100, Jesper Tverskov
<jesper.tverskov@mail.tele.dk> wrote:
> Tina mentions the Q parameter used in some accept-headers. I'm not using
> it in my testing at the moment. As far as I know, and I know almost
> nothing about that subject,

Perhaps writing a document where you need to explicitly know about it
was a big mistake then.  I provided a direct link to the relevant
parts of RFC 2616, please read it and understand why q values can not
be ignored.

> I log the accept-header of all browsers, other user agents, search
> engines, web crawlers, etc. visiting my web pages. If problems appear, I
> must fine tune my testing,

Your techniques are obviously failing, as I've reported a failure to
you, I could not view your site in my copy of IE, because you sent me

> and it could be necessary to start using the
> Q parameter. If someone can convince me it is necessary

It's required by RFC 2616.

> In order for the background-color to fill out the whole wiewport in
> Mozilla Firefox we also need to use the background-color property in the
> html element (the top element).

Or you can change the margin/padding on the relevant elements, it's
not a problem, it's a specification.
> JavaScript. You can not use document.write(), (yes it was bad anyway),
> or other old JavaScript methods to add something to a webpage, hide or
> change something. You must use the methods specified in W3C DOM.

No you mustn't - there's nothing inherently incompatible with
document.write and XHTML that means you mustn't use it.  The only
reason is that Mozilla ddevelopers have chosen not to provide it.
(which is their prerogative of course)

> "Use W3C technologies when they are available and appropriate for a task
> and use the latest versions when supported." [Priority 2]

HTML 4.01 is the latest HTML, XHTML 1.1 is the latest XHTML -
therefore people using HTML 4.01 are meeting that requirement
considerably more reliably than you who are still using XHTML 1.0

> "When serving XHTML with mime-type "application/xhtml+xml" the web page
> must be well-formed. Just one violation of the markup rules of
> well-formedness and the browsers will only show an error message. 

Tina has already pointed out how the error message (or the XML
mark-up) is inherently inaccessible, could you please respond to that
point and explain why error messages and raw XML mark-up are

Ensuring your content is well formed and valid (something your
mime-type doesn't give) is important - but it's not something to
impose on your authors, ensure it earlier in your process - either at
author time if it's static, or using something like mod_publisher or
mod_tidy or whatever other means, but don't throw errors at users,
it's inaccessible.

>That is the recipe of quality web pages based on modules 
>of xml applications.

Could you explain more please?  There's very few client-side XML
technologies that work successfully today, certainly many less
deployed than are successfully deployed with non-XML technologies.

> Why not be ready to start
> experimenting as soon as the browsers give us support for SVG, etc?

Authoring XHTML doesn't make us ready to experiment with those, it's a
completely different technology.  tidy can change all your valid HTML
4.01 to XHTML in a few cpu cycles.  There's no need for it.

Received on Thursday, 24 February 2005 09:36:25 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 13 October 2015 16:21:31 UTC