Re: XHTML 2.0 considered harmful

Tantek Çelik:
> On 1/14/03 3:56 AM, "Christoph Päper" <christoph.paeper@tu-clausthal.de>
>
> I too prefer <L> over <BR>, but why not just keep <BR> and deprecate it?

Maybe the solution was to introduce XHTML 1.x and probably HTML 4.x (x>1) or
5 with the plans for XHTML2 in mind, i.e. deprecate elements and attributes
that will be removed in 2.0 and introduce some new like nl, but don't
introduce new concepts like non-empty meta.

OTOH you could of course just introduce XHTML 2.0 Transitional...

I think the WG should examine common websites to find out what web
developers (not web designers) want/need. The most obvious sign, that
there's a need for a new element, is when on different sites authors have
given styles to certain "words", e.g. names or navigation.
The results of such an study should be published.

>> There's no need for deprecation in a not backwards compatible language.
>
> And perhaps that is one of the cruxes of the problem.

Not perhaps, it is.

> "not backwards compatible" has become the most
> abused cop-out for bad design in XHTML2.0.

Well, it should be used to drop bad design that else had to be inherited
from previous HTML versions. This currently includes most empty elements,
abbr/acronym ambiguity, anchor synonymous hyperlink etc., but lacks others.

A very bad thing excused with this is the quote element, IMHO.

>> He doesn't seem to have realized that the major problem of XHTML
>> in practice is, that it isn't supported by the major browser.
>
> Why is that "the major problem"?

Because you can't really use it to see advantages, thus people don't.

>> [XHTML2] should clearly be considered and advertised as a
>> long term project.
>
> And given lower priority than...

I don't know how high the priority for XHTML 2.0 is, but I see other W3C
projects that are IMHO more important.

>> there should be HTML 4.1 with SHORTTAG NO etc.,
>> resembling current browser behaviour like CSS 2.1.

Just to make that clear: this shan't bless bad practices of browsers, like
allowing inline elements to contain blocklevel ones.

['style' attribute]
> Plenty of other authors do see a use for it.

Yeah: sloppy presentational markup, but with the style attribute rather than
elements like font.

Christoph Päper

Received on Tuesday, 14 January 2003 09:19:26 UTC