Re: Agenda for XML Core WG telcon of 2004 November 17

Paul Grosso scripsit:

> Everyone on the call felt that use of the xml-stylesheet
> here was wrong.  We had some discussion.

On the contrary; I thought it was correct.  I probably won't make the
call this week, as I'm at XML 2004 and already have one teleconference
that day.

> Most of us are certainly opposed to them using 
> type="application/xml".

I think it is the Right Thing, since 1) it tells the truth and 2) XML
stylesheets are self-identifying (you can look at the root namespaces to
see what kind of stylesheet you have).

OTOH, you can't reliably tell just by inspection whether a document
is a CSS stylesheet or not, so the type pseudo-attribute text/css is
necessary to mark it.

> Arnaud would like to know more about the motivation and
> why they are using the xml-stylesheet PI instead of
> something else.

Because it's functioning as a stylesheet.  If document A points to
document B, then when A is to be displayed, B is displayed instead,
with parts of A plugged into it as specified by XForms rules.

> We noted that the XSLT spec says (last para before section 2):
> 
>   The MIME media types text/xml and application/xml [RFC2376]
>   should be used for XSLT stylesheets.

That certainly does not mean that text/xml and application/xml should not
be used for other kinds of XML documents.  A fortiori, it does not exclude
their use for other kinds of stylesheets, either.

In other news:

> Norm has suggested a possible update to XLink at
> http://lists.w3.org/Archives/Public/public-xml-core-wg/2004Nov/0033
> where we say that an element that has an xlink:href but 
> does not have an xlink:type should be treated as if it
> had a "simple" link type.

+1

> Norm continues to recommend that we make a profile that is the 
> same as XML 1.1 except to change the bnf so that you can't have 
> any sort of doctype decl.

+1

-- 
If you understand,                      John Cowan
   things are just as they are;         http://www.ccil.org/~cowan
if you do not understand,               http://www.reutershealth.com
   things are just as they are.         jcowan@reutershealth.com

Received on Monday, 15 November 2004 16:31:59 UTC