[lug] Typesetting Programs

[rm at fabula.de]

On Fri, Dec 13, 2002 at 07:06:04AM -0700, Steve Sullivan wrote:
> At one time during the early "XML will cure all" days,
> XML FOP was pushed as a replacement for Latex and other
> formatting programs.

Wishfull thinking & hype, i'd say (sadly).

> In theory, it might have been.
> In reality, it was partially and poorly implemented,
> and then the team started over.  For the last year
> or so the project seems to have stalled.
> Definately not ready for real use!

Yes, the idea is tempting and their page model actually 
pretty good (well, considering that it's based on the 
DSSSL page model that should hardly be a surprise) but
unfortunately there's no (free) renderer that fits the 
specs :-/ The LaTeX approach doesn't really work (LaTeX
page model is not as complex and capable as XSL-FO's).

> I've found Latex powerful and flexible enough for nearly
> all situations.  There are a large number of good
> books on it, and a variety of tools to convert between
> various formats.

Hmm, yes and no. LaTeX is actually a pretty pragmatic mixture
of semantic and typographic markup. This is unclean/non-kosher
from a software/system design point of view but probably the only
way to get convincing typographic output. The problem i see with
automatic conversion from "pure" semantic markup (ala XML) to
LaTeX is that there's no way to provide the often needed typographic
infos (like distinction between a sentence-ending '.' and one ending
an abreviation, italic correction etc.). In a large project this 
info might be added in a separate step (by a typograph/layouter)
and i could imagine "typographic markup" (ala '<typo:nbs/>'). Still,
without a pretty darn smart XML-Diff it will be hard to merge later
changes by the author once this markup was done.

 Ralf Mattes
> 
> Steve Sullivan
> 
> 
> ========================================
> Steve Sullivan    sullivan at mathcom.com
> 
>    Mathcom Solutions Inc.: Custom Software Development.
>     * Mathematical optimization, simulation, and modeling.
>     * Data mining, information retrieval.
>     * Java, XML, C++, Mathematica, Matlab, XSLT, XQuery, SOAP, RMI, ...
> 
> http://www.mathcom.com    303-494-7115
> ========================================
> _______________________________________________
> Web Page:  http://lug.boulder.co.us
> Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug
> Join us on IRC: lug.boulder.co.us port=6667 channel=#colug