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

RE: Requesting a revision of RFC3023

From: Francois Yergeau <FYergeau@alis.com>
Date: Thu, 18 Sep 2003 10:16:48 -0400
Message-ID: <F7D4BDA0E5A1D14B99D32C022AEB73660EB380@alis-2k.alis.domain>
To: "'Tim Bray'" <tbray@textuality.com>, "C. M. Sperberg-McQueen" <cmsmcq@acm.org>
Cc: MURATA Makoto <murata@hokkaido.email.ne.jp>, ietf-xml-mime@imc.org, WWW-Tag <www-tag@w3.org>

Tim Bray wrote:
> I agree entirely with Michael and feel that the, er, 
> textuality of XML is at the centre of everything.

I happen to agree with that.  But please consider that saying 'XML is text'
is not at all the same thing as saying 'XML is "text/"', as the latter is a
MIME concept that carries a lot of additional (and unfortunate) baggage.

> Thus, I disagree with Francois and 
> Makoto in their contention that XML is not usefully 
> considered as text for humans to look at.

I'm afraid this is the case for most humans in most situations.  Sure, there
are people who like to view and edit XML as plain text (pleading guilty
here;-), but deprecating "text/*xml" won't stop me: save to disk and fire up
the editor.  I don't fell I'm going to lose anything of great pragmatic
value, although there is indeed a loss of conceptual value in not calling
XML what it is: text.

> Having said that, I feel that the use of media 
> types beginning with "text/" remains inappropriate, but primarily 
> because of the charset defaulting baggage that comes with those five 
> characters.

Yes.  Oh, and "text/" also implies the MIME canonical form with short lines
delimited by CR-LF, making UTF-16 and UTF-32 impossible, etc.  Broken.  To
be deprecated.

Received on Thursday, 18 September 2003 10:19:39 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:32:39 UTC