W3C home > Mailing lists > Public > www-html@w3.org > January 2000

Re: Why DOCTYPE Declarations for XHTML?

From: Arjun Ray <aray@q2.net>
Date: Tue, 18 Jan 2000 21:58:22 -0500 (EST)
To: www-html@w3.org
Message-ID: <Pine.LNX.4.10.10001182135210.2094-100000@mail.q2.net>

On Tue, 18 Jan 2000, Murray Altheim wrote:

> I think you're missing the point I've tried to make over and over: we
> can't walk on water. We have to use existing specifications for such
> core functionality. 

Hear you loud and clear, Boss.  The counterclaim is that the existing
specifications, to which you are restricted, don't have anything for this
core functionality.  It's an as yet unresolved *problem*.

> It's not within the scope of the HTML WG activity to be redesigning
> 'standardized mechanisms'.

And, contrarywise, is it in scope to rule mechanisms *out*?  Is it within
the scope to promulgate misconceptions, supposedly because "we got nothing

> You should by now know that an AFDR or PI solution is simply not going
> to fly at the W3C;  they just don't get it or don't like it or don't
> like those who speak a different language than they. And no kludgery
> will do -- it's GOTTA be in a W3C RECOMMENDATION before we can use it.

As a requirement of orderly procedure, this is entirely understandable.
But it's an unwarranted and dreadful imposition to be forbidden to report
that the existing specifications are *incomplete*.
> If I did a James Clark and submitted a NOTE using a PI, it wouldn't
> make it past TimBL, who as I have already repeated said (and even
> included the relevant quote) will simply not use PIs in any further
> W3C spec. It's right there in electronic ink in front of your eyes: NO
> PIs. And I don't see how some namespace kludgery would do here either.

Are you saying that you were *ordered* to mandate doctype declarations for
this core functionality?
> And yes, the DOCTYPE declaration *is* the way to express what we are
> trying to express, 

No.  If that were true, the XML spec would have had no cause to require
(limited) parsing of internal subsets by non-validating processors.  The
fact that the category exists, is defined, and is unquestionably useful,
proves you wrong. 

> just not what you're trying to express, which is impossible.

We can do that at once.  Miracles will take a litte longer;-)

Received on Tuesday, 18 January 2000 21:54:15 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:05:52 UTC