W3C home > Mailing lists > Public > www-tag@w3.org > April 2003

Re: internet media types and encoding

From: Tim Bray <tbray@textuality.com>
Date: Mon, 14 Apr 2003 09:52:00 -0700
Message-ID: <3E9AE730.2070401@textuality.com>
To: Paul Grosso <pgrosso@arbortext.com>
Cc: www-tag@w3.org

Paul Grosso wrote:

> So now it doesn't sound like, in fact, you were "arguing
> convincingly that the C1 control characters should be
> excluded from XML 1.1."
> 
> Concentrating on this issue for a moment--which is the one
> I had raised and the one asked in the CR request--then how 
> do you come down on the question of whether we should:

hmm... and sidestepping the C0 issue, which is the one I really care 
about ... but anyhow:

> 
> a.  say that the presence of these characters (other than
>     via an NCR) means the result is not well formed XML 1.1; or
> 
> b.  say that, since these characters are allowed (unescaped)
>     in XML 1.0, the benefit of backward compatibility suggests
>     we should continue to allow them in XML 1.1 (unescaped).

I'd go with (c.), not allowed either native or NCR'ed.  I don't think 
(a) makes any sense, I'm worried about them being in the infoset not 
being in the syntax.  However, I can see the other side of the argument, 
and option (b) is clearly defensible.   What I still can't see is why 
introducing C0 controls into 1.1 is a good idea.
-- 
Cheers, Tim Bray
         (ongoing fragmented essay: http://www.tbray.org/ongoing/)
Received on Monday, 14 April 2003 12:51:58 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:17 GMT