Re: HTML 3.2 PR

jon@oaktree.co.uk
Sun, 17 Nov 1996 18:06:21 +0000 (GMT)


From: jon@oaktree.co.uk
Message-Id: <199611171806.SAA09252@white.oaktree.co.uk>
Subject: Re: HTML 3.2 PR
To: sja20@hermes.cam.ac.uk (James Aylett)
Date: Sun, 17 Nov 1996 18:06:21 +0000 (GMT)
Cc: www-html@w3.org
In-Reply-To: <Pine.LNX.3.93.961117173741.3390C-100000@crystal.clare.cam.ac.uk> from "James Aylett" at Nov 17, 96 05:45:01 pm

James Aylett wrote:
> Therefore the text '<title*' on its own is meaningless; it opens a tag,
> but doesn't close it. I imagine most browsers will then skip everything up
> to the next '>', assuming that title* is some tag it doesn't understand.

The significance of the '*' is that it means that '<title*' is *not*
an unknown start tag. The character '*' is not allowed in a tag name.
(AIUI) So, my question is whether we are supposed to treat '<title*'
as illegal markup (and hence we can do whatever we like with it,
i.e. probably skip until '>'), or whether we should treat it as
perfectly legal data characters which should be displayed on
the screen.

> The reason the 'text' (by which I assume you mean the RFC for 2.0, or
> perhaps the 3.2 docs, or whatever) doesn't define what to do with
> <FLIBBLE> or whatever is that it's already covered by the fact that HTML
> is an SGML application.

I was talking about the specific example in RFC 1866, <!- stuff ->,
not unknown tags. You don't need to know about SGML to do HTML,
you can go just by RFC 1866 - except for a couple of questions
like the one I'm asking here. AIUI '<!-' would mean an SGML
declaration with name '-', which is presumably an illegal name
and hence we are back to the same position as with '<title*'.

Cheers


Jon
____
\  //    Jon Ribbens    // 10MB virtual-hosted // www.oaktree.co.uk
 \// jon@oaktree.co.uk // web space for 49UKP //