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

Re: XHTML, XML, fancy text, and applications

From: Martin Duerst <duerst@w3.org>
Date: Sun, 13 Apr 2003 13:01:29 -0400
Message-Id: <>
To: Linus Walleij <triad@df.lth.se>, ietf-types@alvestrand.no
Cc: www-html@w3.org

Hello Ethan, Linus, others,

At 14:12 03/04/13 +0200, Linus Walleij wrote:
>This discussion is new on this list and sound confusing to me. It would
>be nice to have a briefing on the background facts.

For background/facts, see
1) RFC 3023 itself, in particular Appendix A
2) The archives of the ietf-xml-mime mailing list, at

Please come back to discuss here after you have reviewed
these resources.

Regards,   Martin.

>For example, I want to know where the *existing* and *used* transport
>types application/xml text/xml fits into this, and why it shouldn't be
>used for XHTML like for all other XML derivates. If the browser
>recognize that this XML file belongs to a certain namespace and should
>be rendered in a certain way (xhtml-wise) that is in my humble opinion
>not a *transport* problem.
> > I would see it as "This resource is of such-and-such type AND it is XML in
> > terms of syntax."
>= content type text/xml, and the first row of that XML-compliant file:
><!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
>will tell that it is of such-and-such XML derivative type. This is thus
>not a transport issue. The transport types are designed to be quite
>general. A binary file has no standard for deciphering it's contents
>(well with the exception of say Quicktime and such non-IETF de facto
>standards) why it has to be carried in it's own transport type.
>text/xml and application/xml are reserved for XML derivatives. These
>additionally have clear syntax for describing their context i.e.
><!DOCTYPE ...> so they need no additional transport typing. To add it
>anyway would be unnecessary duplication of information.
> > I have
> > to say that the pipe character ("|", vertical line, U+007C) is too 
> devoid of
> > well-known semantics.
>This is a technical standard, and the pipe sign has been a part of the
>Backuss-Naur form denoting alternatives since the 1960s. It is nowadays
>usually used in the Extended Backus-Naur form for regular expressions.
>The Backus-Naur was developed heavily inspired by Noam Chomskys language
>hierarchy and formalisms for grammar introduced with his book "Syntactic
>Structures" in the 1950s. Chomsky didn't have pipes, but both BNF and
>EBNF has. This is the reason why it was in the DTDs, and why it is part
>of most document description syntaxes. That is: it is not devoid of
>well-known semantics.
>This might be a totally off-topic reply in regards to this discussion,
>but I could not accept that statemant as it stood.
> > Right, and those should be subtypes of "text" if they can reasonably be
> > treated as "text/plain", or subtypes of "application" if they contain
> > non-textual markup.
>application/xml and text/xml both already exist. What is wrong with
>Linus Walleij <triad@df.lth.se>
Received on Sunday, 13 April 2003 14:22:38 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:06:03 UTC