W3C home > Mailing lists > Public > xml-dist-app@w3.org > January 2002

Re: Media types

From: Paul Prescod <paul@prescod.net>
Date: Tue, 15 Jan 2002 09:33:41 -0800
Message-ID: <3C4467F5.51A78CD3@prescod.net>
To: www-tag@w3.org, xml-dist-app@w3.org
Mark Baker wrote:
> 
>....
> > It's a perfect example. This document is logically XSLT, not HTML.
> 
> In this example, I'd say it's both HTML and XSLT.  However, HTML has
> the advantage in determining how that XSLT should be interpreted, since
> it's the container.

Let's put it this way. The XSLT fully defines the meaning of the
document. In fact, this document was cut and pasted OUT of the XSLT
specifications. That specification claims that it is XSLT. The XHTML
specification, on the other hand, specifically says that such a document
is not "a strictly conforming XHTML 1.0 document". So one specification
claims the spec. is correct and describes how to process it. The other
says it is "not strictly conforming" and says nothing about how to
process it.

> For example, if HTML had an element called "do-not-process" that meant
> that any content whtin should not be dispatched to alternate processors,
> and that your XSLT was within this element, would you still say it was a
> stylesheet?

Absolutely. And it isn't a matter of opinion. XSLT says nothing about
the do-not-process element, therefore it would be invisible until after
the XSLT parts had been processed. The HTML processor doesn't even see
the content until after the XSLT processor is done with it.

> I agree with TimBL when he says;
> 
> "The significance of any nesting of one withing the other is to be defined by
> the nesting (outermost) specification [...]"
> 
> (from http://lists.w3.org/Archives/Public/www-tag/2002Jan/0081.html )

That's a good theory but there is at least one W3C spec that does not
work that way. Perhaps it needs to be fixed.

 Paul Prescod
Received on Tuesday, 15 January 2002 12:38:02 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:59:06 GMT