Re: [Texmacs-dev] TeXmacs server

[Joris van der Hoeven]

On Sat, Jun 03, 2006 at 09:48:28PM +0200, David MENTRE wrote:
> That's said, I would prefer seeing currently annoying bugs fixed
> (slowness, weird behavior regarding X focus handling[1], not capable of
> reading its own document produced by earlier versions[2], etc.) rather
> than introducing fancy features. And again, there are other tools
> already doing what Joris wants to implement, like Gobby[3]. But I
> suppose Joris is scratching his own itch.

People are allowed to contribute to bug fixes. I don't have the energy
to do only that. Anyway, many bugs are usually fixed while reorganizing
code in a better way. For instance, the better linking system that
I am currently implementing should make it possible to have a cleaner
concept of views, allowing us to rewrite part of the GUI as TeXmacs
documents. My next aim is to get rid of the current widget engine.
At that point, only the X layer remains and the design becomes more sound.
Another important reorganization concerns the style motor.
But, like many people, I also implement things I directly need,
from time to time (like TeXmacs chatting).

> This approach is clearly not scalable and cannot be used in practice
> beyond a few lines (been there, used that).
> 
> As Joris underlined, a scalable approach:
> 
>  1. allows the user to write a document in his prefered language;
> 
>  2. uses an *automatic* computer system to find "things" to translate
>     and find translation if already known.
> 
> This is the approach used everywhere in the computer industry, like
> gettext and .po format in the Unix world. In this particular case,
> "things" are strings. Fundamentally, Joris approach seems quite good to
> me but, once again, I would have much prefered seen TeXmacs use standard
> tools and format (.po files) than re-inventing its own and incompatible
> world.

Sure. Except that .po files might/should have been used for the menus,
but are less adequate for translations of structured document fragments.

> Those who code decide, but don't be surprised if they are not
> followed. Sorry for the harsh tone of this email, but I was once seduced
> by TeXmacs features description and I was then strongly dissatisfied by
> daily use of TeXmacs.

Once again, if certain bugs really dissatisfy you, then you are free to
solve them if you are a programmer. I am very satisfied with the daily
use of TeXmacs myself and never experience many of the bugs submitted
by several users. From time to time, I take on my time to solve some of them,
but this should not be my full-time job.