LATEX-L Archives

Mailing list for the LaTeX3 project

LATEX-L@LISTSERV.UNI-HEIDELBERG.DE

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
Subject:
From:
"Denis B. Roegel" <[log in to unmask]>
Reply To:
Mailing list for the LaTeX3 project <[log in to unmask]>
Date:
Wed, 17 Nov 1999 13:38:40 +0100
Content-Type:
text/plain
Parts/Attachments:
text/plain (59 lines)
`Bernard GAULLE' wrote
>

...
>   Since standard TeX engines don't accept dynamic patterns loading
>   (La)TeX formats should contain customized data like language patterns
>   and thus at run time you can't expect that it's the good format
>   for your language (the good fonts, the good patterns) etc. nor force
>   any default depending of theses choices. You only can expect to
>   have CM and US-English patterns, that's all.

However...

When you write \usepackage[frenchb]{babel}, babel tries to load
the french hyphenation patterns, and if they are not available,
tells you so. The fact that such or such patterns are available at
a given installation is a configuration of that installation.
If you get a document with \usepackage[frenchb]{babel} and you don't
have the right hyphenation patterns, you get a wrong output,
but it is your fault, because you have not made sure that the
installation fits the file.

To some extent the same goes with encodings, both input and output.
Given the features of TeX, the choice of the output encoding
is relevant for hyphenation, or if you prefer, the brand of TeX
you use (TeX, MlTeX, ...) is not independent of the encoding of
fonts you use. I think we agree on this. The point is that
common decisions can be made a default, even if they do not suit
everybody. And I think most french users of LaTeX use T1 as an output
encoding (I am confident that such would be the result of a survey).
This is not to deny that there are other ways of doing things,
just that it is sensible to have defaults. In this debate, it is
also useful to keep the distinction clear between "consensual defaults"
and "local configurations." I am a strong opponent to local
configurations that make encodings and other things an implicit *local*
default. This goes against portability. But I am an advocate of global
(consensual) defaults, which do not hinder portability, and make
life easier for a certain number of people.

Returning to babel: it also has some consensual defaults for
things like language specific expressions ("Table of contents,"
"References," etc.) for which, at least in some languages, there
can be several choices. So, what babel currently provides
does not suit everybody, but it suits a majority.

There are various ways of implementing the defaults I advocate,
but, in some way, they should be tied to the language. I mean,
the defaults should be chosen at the time the language is selected
(but the defaults can depend on other factors, as I said in another
message). The defaults can be made part of babel, or they can be made
part of some extra language package. For instance, the french package
could be extended to take options, one of these options being
for instance "latin1-t1." A certain option from the french package would
be the default option.

It is not much work, and it would benefit to all.

Denis Roegel

ATOM RSS1 RSS2