MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
References: <199812092035.VAA16014@na6.mathematik.uni-tuebingen.de>
            <199812140945.JAA02920@nag.co.uk>
            <13941.21350.836780.848801@fell.open.ac.uk>
Message-ID:  <13942.15655.945558.913902@srahtz>
Reply-To: Mailing list for the LaTeX3 project
              <LATEX-L@URZ.UNI-HEIDELBERG.DE>
In-Reply-To:  <13941.21350.836780.848801@fell.open.ac.uk>
Date:         Tue, 15 Dec 1998 10:42:47 +0000
From: Sebastian Rahtz <s.rahtz@ELSEVIER.CO.UK>
Sender: Mailing list for the LaTeX3 project
              <LATEX-L@URZ.UNI-HEIDELBERG.DE>
To: Multiple recipients of list LATEX-L
              <LATEX-L@URZ.UNI-HEIDELBERG.DE>
Subject:      Re: portable LaTeX
Status: R

Chris Rowley writes:
 > > or whatever. You can (or rather will be able to) then specify in a style
 > > sheet how this is supposed to get transformed to MathML.
 >
 > So you cannot put these specs into the document itself
you can embed a reference to the style sheet.

 > Is there anything corresponding to macro parameters?
give an example of what you want

 > > If you compare the first draft submission of XSL to the first working
 > > draft of XSL 1.0, which came out only a few months later, you will
 > > see that they are essentially completely different languages.
 >
 > I am not sure what I studied, but it was in mid-1997.
in that case, start all over again. thats eons ago... did XSL
_exist_ in mid 1997?

 > On the contrary, worry is all we can do, given the fate of other
 > attempts at formatting specification languages and their imposed model
 > of what a formatter is, what it can do and what interfaces it provides.
which are these other attempts? we had FOSI, but who ever used that?
and we had DSSSL, which has not been fully implemented by anyone, so i
don't see how you can dismiss it out of hand.

why do you think people are imposing on you? "they" are "we". join the
W3C and get on the XSL committee if you care deeply and have time to
contribute.

 > > alright on the day, you will be able to do the transformation in a lower
 > > level language, such as java or ecmascript interfacing through the DOM.
 >
 > What is the DOM?
Document Object Model, a kind of abstract API for XML documents. A W3C
recommendation, widely implemented

 > But what will "these" be?  Will they be extensible (not so much the
 > maths ones, I am more worried about the pargaraph ones)?
what is an extensible formatting object? you want to add
characteristics? new ones?

 > That sounds very sensible but we (and I really do mean almost
 > exclusively the readers of this list) therefore need to ensure that we
 > can easily plug-in something based on TeX as the formatter for XML;
sorry, but the Web world cannot afford to use TeX as a dynamic
formatter, its just too heavyweight. for print, maybe, but in that
case you'd need to make a reliable batch page formatter macro package. which of
course LaTeX is not. face it, conventional TeX/LaTeX as of today
cannot meet the demands of automated large scale formatting of XML
documents. when you can do multiple columns and single column floats,
give us a call...

 > Agreed, although the world of those who understand the idea of a
 > platform-indpendent, programmable system for document processing (as
 > provided uniquely at present by TeX-related systems) and formatting
do who cares about platform independence? if you drop that, you can choose
from a raft of other formatters which equal or surpass TeX in some
ways

 > seems far wider to me than the world of those shackled to what giant
 > corporations, and sebastian, think is making people happy.

this is getting like an episode of the X Files. shall we get Mulder
and Scully to investigate the huge conspiracy to fetter international
document formatters?

sebastian