'

[Irl-dean] A nice summary of why not to use client-side XHTML ...

Barry McMullin barry.mcmullin at dcu.ie
Tue Jul 31 12:40:14 IST 2007


Hi Josh -

On Tue, 31 Jul 2007, Joshue O Connor wrote:

> Because of _one browsers_ lack of support. If it did - this would be a
> non issue.

The IE lack of support is the single biggest objection to serving
XHTML (properly - i.e., as XML); but there *are* problems with
almost all other "mainstream" browsers also.  Check out the full
article for details.

> I agree that there will be cause for validity errors as the markup is
> not being parsed correctly, (as it is being parsed by an SGML parser)
> but again if there are any _real_ accessibility issues that arise, or if
> indeed you find any issues (like show stoppers and not objections such
> as 'Its not an ideal way to deal with code', or minor CSS mishandling)
> again please send them my way.

Sorry, I probably presented this poorly by quoting only the
conclusion, without more of the context.

The key point is *not* whether serving XHTML presents any
"showstoppers", accessibility related or otherwise.  The point is
whether serving XHTML offers any *benefit*, accessibility or
otherwise, that would justify serving it in preference to HTML,
given the fairly long list of more or less serious deficiencies
in the deployed support for it - ranging from IE's complete
rejection at one end, to the relatively more minor but still
problematic issues even with opera or browsers using the
mozilla/gecko engine.

That is (and now this is purely me speaking, not David Hammond),
in my opinion, the burden of the argument should be properly on
those wanting to serve XHTML, to explain *why* this is better
than serving HTML (without going into red herrings about serving
valid rather than invalid code, or strict rather than
transitional code, or using XML internally on the server side, as
all of these are completely independent of serving HTML versus
XHTML on the client side).

But finally, I didn't actually mean to re-open the whole
debate here (honestly!); I posted the link simply because I
thought Hammond had lined up and articulated the arguments on
this much more comprehensively and clearly than I have seen
anyone else do, to date; and certainly much more clearly that I
have ever managed myself.

Best - Barry.

--
Barry McMullin, Dublin City University
  phone: +353-1-700-5432
  web: http://www.eeng.dcu.ie/~mcmullin/




More information about the CEUD-ICT mailing list