Next message: Terje Bless: "Re: Any C++ programmers around? (was: Unix --> NT (source code stuff))"
Date: Fri, 9 Mar 2001 12:13:55 +0100
From: Terje Bless <link@tss.no>
To: W3C Validator <www-validator@w3.org>
cc: Liam Quinn <liam@htmlhelp.com>
Message-ID: <20010309122715-r01010600-edca2bfc@10.0.0.2>
Subject: Re: Table Validation
On 06.03.01 at 16:31, Liam Quinn <liam@htmlhelp.com> wrote:
>On Tue, 6 Mar 2001, Terje Bless wrote:
>
>>On 27.02.01 at 13:33, Liam Quinn <liam@htmlhelp.com> wrote:
>>
>>>On Tue, 27 Feb 2001, Nick Kew wrote:
>>>
>>>>Last time I checked, the correct behaviour here is to validate against
>>>>HTML 2.0, but none of them do that.
>>>
>>>I don't think that is correct. The HTML 2.0 standard says [1] "To
>>>identify information as an HTML document conforming to this
>>>specification, each document must start with one of the following
>>>document type declarations."
>>
>>IIRC, we've rehashed this a couple of times and the conclusion was that
>>HTML 2.0 is the _only_ version of HTML that makes the DOCTYPE declaration
>>optional
>
>This assertion is frequently made, but the quotation that I cited proves
>the assertion wrong.
*sigh* Could those that had some sort of conclusive proof that HTML 2.0
made the DOCTYPE optional ("should" or "may"?) please holler? AFAICT,
Liam's citation [1] specifically states, in prose, that a DOCTYPE is
required for a conforming HTMl 2.0 document. Even if it later says that
User Agents are allowed to guess in the absense of a DOCTYPE and still be a
conforming _UA_, this does not negate the requirement that a DOCTYPE be
present to actually be a conforming _document_.
[1] - <URL:http://www.w3.org/MarkUp/html-spec/html-spec_3.html#SEC3.3>