Re: let authors choose text/html or application/xhtml+xml (detailed review of section 1. Introduction)

Robert Burns wrote:

>> That's right, because of this millions of people out there are 
>> thinking that there web documents are valid XHTML, but they're not, 
>> they are in fact invalid HTML.
> 
> They are often valid XHTML. Many who make use of XHTML have a stronger 
> tendency to use validators too. You may be right that these files may 
> not validate as HTML too. But then again any document that uses EMBED 
> will not validate as HTML. Clearly browsers use a different notion of 
> "valid HTML ' than the validators do. That's a problem that could be 
> easily fixed by making an HTML 4.02 DTD that includes EMBED and disabled 
> nulll end tags. Voila! Now all of the invalid XHTML-like HTML files 
> suddenly becomes valid HTML.

No, they don't, because they still commence with a pre-HTML 4.02 DOCTYPE :-(

Although you and I agree on many things, this is one area over which
we differ.  For years, people have sent bug reports to the validator
team, saying (in effect) "my document displays perfectly in IE / Mozilla /
Opera  / w-h-y, yet your validator keeps claiming it is faulty.  Why
can't you fix your validator ?", to which the validator team, politely
but firmly, reply "Your document is /invalid/, that is why the
validator tells you it is faulty.  You need to fix your document
rather than us "fixing" the validator".  Now this whole HTML 5 exercise
is seeking to turn that situation on its head, and to re-define "valid"
as "works inter-operably across the major browsers", tweaking and
adjusting the HTML 5 specification until that end is accomplished.
I am very much afraid that, if this succeeds, the W3C will lose any
last vestige of credibility, and all those who have claimed that
their documents were valid (when they were, in fact, bug-ridden)
will trumpet loudly "We were right, you were wrong : na na nee na na".

Received on Friday, 31 August 2007 16:14:14 UTC