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

Re: Doctype detection

From: Daniel Hiester <alatus@mail.earthlink.net>
Date: Thu, 27 Jul 2000 11:13:04 -0700
Message-ID: <001501bff7f6$54f326a0$0100a8c0@distant>
To: <www-html@w3.org>
Mathew Brealy said:
"Microsoft owns the browser market and can do what it
likes. If it released a compliant browser, this would not affect its
market share, on the basis that most people get Internet Explorer with
their computer, and do not even realise that other browsers exist (and
also on the fact that it is popular for its interface more than its
levels of support for CSS and HTML). The people who have created crappy
sites would very soon fix them if they started looking bad in Explorer.
If they chose to follow this root the web would be standards-compliant
within a couple of years."

Microsoft is more concerned about Money. There are people who pay them money
for Intranet "solutions," and thus, that is their priority. Because they've
been paid to provide these solutions, they will not, and cannot provide a
compliant browser, because, unfortunately, their proprietary extensions work
in ways similar, but different from, W3C specs. It's impossible to support
both in the same renderer. (disclaimer: I'm not affiliated w/ Microsoft,
this is merely my loose understanding of the scenario, please take it only
as that, if I'm wrong, you have every right to gently correct me)

Windows IE's CSS support was bar-none the best until this year, but I'm sure
we all know this.

Mathew Brealy also said:
"If they must, the only way this should be implemented is via a HTTP
header (and http-equiv). It is vital that the default be correct, rather
than bug, mode - as I said above, sites would change in a flash if
Explorer was compliant (and this would do Microsoft a lot of good too I
might add)."

The idea is interesting, but I thought that:
a.) XHTML is sent as text/html and I am, for one, unaware of any plans to
change this
b.) not everyone has control over HTTP headers (like the millions of people
who use free webspace services like Geocities et al.)
c.) it's not in the spec

What's more, I don't see how there's a difference in the end-result if a UI
chooses between "quirk mode" (I like to call it "legacy mode") via means of
HTTP headers, or a doctype declaration, except for the fact that doctype
declarations are in the XHTML specs.

I seriously doubt that XHTML 1.1 could be correctly rendered in "legacy
mode." That's my bottom line. Am I missing something?

Received on Thursday, 27 July 2000 14:10:38 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:05:54 UTC