mailing-list for TeXmacs Users

[Joris van der Hoeven <address@hidden>]

> But I really agree with MS that it is just as futile to hope that
> journals are going to accept .tm files anytime soon. I am going to ask
> my editor if they accept .tm files, but it won't do much good. I won't
> press the point, as I am the new author who tries to get published and
> I am not in the position to make demands.

It is not about making demands: it is about letting them know that
they are making you things that you don't want to do (conversions) and
about trying to make them work too.

For instance, I doubt that if you send an article which is correct LaTeX,
but which does not scrupulously follow their guidelines, that they will
not publish your article: you may always object that their instructions
are not clear enough or whatever and, at the end, they will do whatever
they feel is necessary. So the whole thing is about pushing them to do
something instead of putting all burden on our shoulders.

> But even if many journals accepted .tm files, I would still ask for a
> LaTeX export function that generates macro-free LaTeX. Why? Well, no
> matter how much I advertise for TeXmacs among my colleagues, many still
> prefer to use LaTeX. I disagree with their choice, but again it is their
> decision. So, when I want to cooperatively write a document with them,
> we will have to agree on LaTeX as file format, simply because it is "the
> standart" and the least common denominator. Have you ever tried to merge
> LaTeX files written by different people? It is a nightmare, that can
> only be made bearable, when *everyone* agrees *not* to define custom
> macros and *not* to use "non-standart" LaTeX packages.

I think that generating macro-free LaTeX is quite difficult, because at
the other end, you also want LaTeX which has a layout/structure as close
as possible to the TeXmacs document. These two goals cannot be met together.
For instance, should we put

        
T\\kern-.1667em\\lower.5ex\\hbox{E}\\kern-.125emX\\kern-.1em\\lower.5ex\\hbox{\\textsc{m\\kern-.05ema\\kern-.125emc\\kern-.05ems}}

in the text, whenever you write \TeXmacs? And what about reimporting such
a mess (which may also be important if you collaborate with collegues)?

What might be done (as a user preference) is something else:
identify the set of all natively recognized TeXmacs/LaTeX macros and
expand any macros which are not in this set. In parallel, we might
recreate a TeXmacs.cls style file in which all these macros are defined.
We already managed to get an older TeXmacs.cls file included in standard
LaTeX distributions, so this should be possible with an updated one too.

An additional benefit of this approach is that we can be 100% sure
that a generated LaTeX document will compile (unless there is a bug);
this is already the case for our Html exporter. A disadvantage is that
you lose any potential additional structure, including the definitions
of your own macros.