[whatwg] XHTML Considered Harmful

[Brad Neuberg]

Hi Bram. Thanks for your comments; responses inline.

--- Bram van Leur <bvleur at gmail.com> wrote:

> Brad Neuberg wrote:
> > The post is basicly a list of reasons
> > on the mistakes and problems with XHTML and why
> the
> > XHTML religion is impeding progress on the web. 
> In my opinion you fail to make the latter point
> clear. The only thing I
> read in the blogpost is:
> - XHTML doesn't fit in the way you're used to
> develop for the web (1,2,3)

These are really the heart of my argument. I'm not the
only one who uses these techniques in creating full
fledged AJAX and DHTML applications.  If you look over
the code of any sophisticated DHTML framework, such as
the Dojo Toolkit (www.dojotoolkit.org), you will see
these techniques used.  If I were the only one doing
this then of course they wouldn't be a good argument
:)

I have a list of mini-tutorials that uses exactly
these kinds of techniques, documented at
http://codinginparadise.org/projects/tutorials/, to
solve real AJAX problems.


> - You've spotted some problems with the current
> state of the
> implementations (4,5,6)

These are also part of the argument. In an ideal world
the implementations wouldn't matter, because they
would all be following the specs exactly and would
have no bugs. However, in the real world, which is
where I live every day creating real DHTML systems,
the fact that an XHTML system leads to unacceptably
slow performance due to no incremental rendering is a
significant issue.

> - You're irritated by the advocates (7,8)
> 

Yes I am; I'm mostly disappointed.  The main advocates
used to create great standards that helped the web;
now they create stunt double standards that block
better ones from gaining a foothold.  I should have
mentioned that XHTML is one of the failed standards in
that list of W3C black eyes the last few years; I made
a mistake by not mentioning it in point 7.

> Most of these problems affect you, only if you were
> forced to develop
> XHTML pages (which you are most likely not).

A very vociferous XHTML community has developed the
last few years that essentially does force developers
to use XHTML, at least publicly. Using XHTML is
currently considered a best practice; I'm trying to
voice that I believe it is not.

> Visiting XHTML pages can be
> just as cumbersome as visiting HTML4 when the
> developer of the page
> didn't do some proper testing and bugfixing before
> publishing his page.
> Furthermore I tend to agree with the 2nd comment [1]
> on your blog.
> 

Thanks for the dialogue,
  Brad

> - Bram!
> 
> P.S. Messy quotes at the bottom of a mail indeed are
> considered harmful :)
> 
> [1]
>
http://codinginparadise.org/weblog/2005/08/xhtml-considered-harmful.html#c112534835130697531
>