W3C home > Mailing lists > Public > public-html@w3.org > July 2008

Re: Doctype on text/html Pages (Was: [author-guide] Character Entity References Chart)

From: Robert J Burns <rob@robburns.com>
Date: Wed, 23 Jul 2008 14:25:39 +0300
Cc: Michael A.Puls II <shadow2531@gmail.com>, public-html WG <public-html@w3.org>
Message-Id: <A856C1C8-107C-4D26-A84D-3BDD6F3839C2@robburns.com>
To: Henri Sivonen <hsivonen@iki.fi>

Hi Henri,

On Jul 23, 2008, at 2:14 PM, Henri Sivonen wrote:

> On Jul 23, 2008, at 13:53, Michael A. Puls II wrote:
>> That makes perfect sense to me. It's making it clear that such XML
>> documents may contain <!DOCTYPE html>.  That sounds a lot better than
>> "a DOCTYPE", which could mean any doctype.
> The spec intentionally allows no doctype or any doctype in the XML  
> serialization, because placing restrictions on the doctype would  
> violate spec layering and micromanage syntax on the XML layer.

We violate spec layering in what the draft says now. The only thing  
HTML5 needs to say about doctype is what doctype declaration authors  
need to use to declare their XML document an HTML5 document. We don't  
have to say they can use another XML document type. They already know  
that. We're micromanaging the syntax to do so. Imagine if XML 3.0  
eliminates doctype declarations entirely. Does an author follow HTML5  
which says they can use any doctype declaration they want? Or does the  
author follow XML that says don't use a doctype. These are fairly  
esoteric discussion, but it is certainly the current state of the  
draft that violates spec layering since we're providing guidance on  
other XML document types which are wholly outside the scope of HTML5.

Take care,
Received on Wednesday, 23 July 2008 11:26:21 UTC

This archive was generated by hypermail 2.4.0 : Saturday, 9 October 2021 18:44:34 UTC