LATEX-L Archives

Mailing list for the LaTeX3 project


Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Torsten Bronger <[log in to unmask]>
Reply To:
Mailing list for the LaTeX3 project <[log in to unmask]>
Sat, 19 Jul 2003 15:54:41 +0200
text/plain (71 lines)

Joachim Schrod <[log in to unmask]> writes:

>>>>>> "TB" == Torsten Bronger <[log in to unmask]> writes:
> TB> It's very very difficult to parse arbitrary TeX.
> I differ here, about the "very very". For one, there exist parsers.
> Second, a professional versed in formal languages and compiler writing
> can create a new TeX parser from scratch in 2-4 weeks. It is difficult
> not because of technical reasons but because existing programs are not
> put into real end-user products -- the demand doesn't seem to be
> there.

But the authors need a certain amount of discipline, and there I see
the problem.  My own thesis included 10--20 packages and used many
of their commands, which must be parsed too.

> [...]
> TB> I think eventually we won't actually see XML anymore. We will use
> TB> e.g. systems like LyX that use XML as the underlying format, and
> TB> that call TeX for decent typesetting. XML is not for the human eye
> TB> in my opinion.
> TB> Until then, you can make XML rather bearable. Joachim's problems
> TB> with micro-typography, space handling, and TeX markup in special
> TB> situations (in particular in formulae) can be dealt with in XML
> TB> applications.
> I keep reading that argument for almost 15 years now, since I started
> to work with SGML. This statement was taken over unaltered by the XML
> fork. Yes, "XML is not for the human eye" -- and why are we forced to
> work with it that way?

Because systems like LyX haven't got so far yet, and systems like
OpenOffice still must prove that they are suitable for this task.

> The problem is not "can it be dealt with". The problem is "it is not
> dealt with". Lots of promises about tools that are elementary to the
> approach, since the markup is not to be meant to be written by humans.
> But no strong progress. Where is the Apache XML project to provide a
> good interactive document input and manipulation facility?

General XML editors are not very interesting in my opinion.  To be
honest, Emacs is sufficient for me.

The intersting thing are editors for *special* XML applications.
Let me dream a bit: All scientific publishers agree on an XML format
and order a simple-to-use GUI program that can create these
documents.  It runs on Linux, Windows, Mac, etc (because it's simple
itself).  Authors can download it and write their articles with it.

Then there are no authors anymore that use exotic file formats,
format their text in a very strange way, no employees of the
publishers have to re-type the articles, authors don't lose time
with superfluous typographical fine tuning, guideline can be made
much simpler, archiving and retrieving is much simpler etc.

The reason why such things are not realised are probably rather
complex.  I've seen only one DTD so far that tried something like
that, and my impression was that the creators were not competent
enough.  Moreover one publisher alone would never dare such a thing.


Torsten Bronger, aquisgrana, europa vetus