Message-ID:  <eaa92e19@corona.oche.de>
Reply-To: Mailing list for the LaTeX3 project
              <LATEX-L@URZ.UNI-HEIDELBERG.DE>
Date:         Wed, 3 Nov 1999 21:15:01 CET
From: Achim Blumensath <blume@CORONA.OCHE.DE>
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: on silence
Status: R

Frank Mittelbach wrote:
> but without feedback and even if the feedback is mainly of the type,
> please explain this better i don't understand, it is awfully hard to
> write about this. and if i'm not able to finally get this into a shape
> where most of you on the list do understand it and/or think it is
> being useful this is doomed, or not?

In my case, I think the general design is ok. So there is no need to
comment on that. But in order to comment on the details they should be
published first. You said several times that you couldn't release the
code of template because it depends on some unfinished other things.
Perhaps it would be a good idea to just post a list of those templates
without an actual implementation, just with a description of the
arguments, attributes, and semantics.

Here are two (random) things I've thought of:

a) General structure.  IMHO LaTeX should be much more modular so you can
replace parts you don't like with your own implementation. It could be
divided into the kernel containing just the programming environment
like xparse, etc., and a lot of modules, one for each aspect of actual
typesetting (lists, math, fonts, headings, pagestyle,...).

The structure of a class file would be

  preamble
  including modules
  declaring instances of templates defined in the modules

The advantage would be that you could

 o make cosmetic changes by writing a new class which uses other
   values to instantiate the templates, and

 o make fundamental changes by writing a new module with different
   implementation of templates.

b) Displays.  A basic display template IMHO could have no parameters,
the following attributes

  above-skip, below-skip, left-indent, right-indent
  alignment:  left, right, centered, full width
  mode:       text, math-displaystyle, math-textstyle

and it would define two commands to resp. begin and end the environment.
Other, more convenient templates could be defined on top of this.
Unfortunately, I haven't come up with a satisfactory design of high
level math environments, so far.

Achim
________________________________________________________________________
                                _                             | \_____/ |
    //  Achim Blumensath       | \  _                        \O/ \___/\ |
   //   blume@corona.oche.de   |-< /_\                       =o=  \ /\ \|
 \X/    (p^2 - m^2)\psi = 0    |_/ \_                        /"\   o----|
____________________________________________________________________\___|