W3C home > Mailing lists > Public > whatwg@whatwg.org > July 2004

[whatwg] Re: DOCTYPE shouldn't be optional

From: Ian Hickson <ian@hixie.ch>
Date: Fri, 9 Jul 2004 13:51:44 +0000 (UTC)
Message-ID: <Pine.LNX.4.58.0407091344210.11240@dhalsim.dreamhost.com>
On Wed, 7 Jul 2004, Malcolm Rowe wrote:
>
> I really don't know enough about the issue that the TAG are discussing
> to comment, though I think it essentially boils down to how you can
> determine if a document is 'valid' if it contains elements from multiple
> namespaces.  There's no generic way to validate a compound document, for
> example.

I'm not convinced of the need for validity checking (beyond syntax
checking) for reasons I mentioned the other day, but if one assumes that
one wants validity checking of compound documents, it's not hard: you just
need a schema language that supports describing constraints in terms of
mixed namespaces. For example, "<ul> can only contain <li>" or "<div> can
contain anything" or "<svg:rect> is only allowed inside <svg:svg>" or
"<svg:svg> is allowed anywhere". Then you take all the constraints and see
if all the elements in the document obey all of them.

Not that anyone has actually written such a language, as far as I know.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'
Received on Friday, 9 July 2004 06:51:44 UTC

This archive was generated by hypermail 2.4.0 : Wednesday, 22 January 2020 16:58:35 UTC