Re: About the style processing instruction

On Sat, 30 Jan 1999, Didier PH Martin wrote:

> Here is essence of the proposal about a certain usage of the "media"
> property. This is not a big thing but at least an active effort to resolve a
> concrete problem.
> -----------------------------
> XML current specifications do not provide explicit means to convert a XML
> documents into a particular format (Example: a XML document is transformed
> into CGM format). However, The current W3 proposal by being based on CSS
> style properties which allows the capacity to specify the rendering media
> and provide way to add new media parameters.

Surely the media attribute of an HTML link element (or the equivalent
psuedo-attribute of the proposed xml-stylesheet pi) is not so much for
specifying the rendering media, but to allow the
browser/processor/whatever to download or not download a stylesheet and
apply it based on the media it has already decided (by some other means)
that it is rendering to.

A browser will look for a Link/pi with media="print" when the user hits
the print button, and the stylesheet (lets say its an xsl stylesheet
with some <fo:*> like formatting objects for now) will produce lots of
<fo:footnote> elements instead of <fo:inline-link> elements. A stylesheet
for media="aural" would similarly produce lots of <fo:block>s with textual
descriptions rather than <fo:inline-graphic>s.

Note that it was the user clicking a 'Print' button, or a user dialing up
a web-to-telephone service that determined the output format - not the
media attribute. That was merely a useful hint to the browser.

If you really want to specify tex, or rtf, or cgm, then there is nothing
to stop you converting your xml into these formats using whatever
command-line argument, pi or incantation to your favorite processor
uses. You can then put the resulting files on the web. Obviously the less
this happens the better - but it is possible now if you really must use a
particular output format.

Anyway, my apologies if I've entirely missed the point.

Best wishes,

Paul.


Paul Fidler
-- 
Cambridge University Engineering Department
Trumpington Street, Cambridge, CB2 1PZ, UK  



 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list