On Mon, 24 Jul 2006, Andy Dingley wrote:
Who gives a stuff what Hixie thinks?
Quite a lot of what he says is food for thought, even though you might
disagree with him.
XHTML as text/html is kosher,
It *relies* on at least one widespread bug in browsers. That alone
would put it beyond the pale.
But one point that I *do* take from Hixie's presentation is that the
vast majority of soi-disant XHTML on the web today is no such thing.
It's in reality defective HTML, fixed-up by browsers. If that stuff
were ever to be offered as real XHTML, to a real XHTML client agent,
it could not work. Authors have taken XHTML on board because it was
sexy - but without taking on board its principles.
Fact is, browsers are mostly attuned to HTML (and to fuxing-up the
faults in defective HTML). There's IMHO no need to challenge them
with XHTML/1.0 Appendix C, and almost[1] nothing to be gained by it.
The fact that XML-based tools might be used inside your authoring
process is irrelevant - they can just as well emit HTML as their end
product.
We're still waiting for widespread deployment of anything worthwhile
in XML-based markups, such as SVG, Math, etc.
he just has a grudge against it.
That's as may be, but some of his points are on target anyway.
And as for "XHTML/1.0 Transitional", surely that's a sick joke (or as
some deem it, a bogosity indicator)?
ttfn
[1] in fact, if the W3C hadn't gone so head-over-heels for XML-based
markups, it would have been feasible to devise a new DTD for HTML,
which more closely matched the recommended usage of HTML, such as
decoupling the different features of SHORTTAG and setting them
separately to appropriate values; amending some of the rules allowing
opening and/or closing tags to be optional, and so on. Could even
have called it HTML/4.2 ;-)
Yes, I know Hixie wants to throw that out of the window for HTML5,
along with the whole business of DTDs. Don't have to agree with
everything.