Re: Problem with iso8859 on xterm/with man(1)

[joostje@komputilo.org (joost witteveen)]

> > Also, why isn't latin1 or UTF-8 default after installation?
> A.) Debian is a multilingual distribution. Why not latin2 or Japanese
> encodings? 

OK, but then still: why not UTF-8? That's not Western-European biased,
is it?

> B.) In the new glibc you have to create the locales you need:
>     B.1) Uncomment the lines you need in /etc/locale.gen
>     B.2) Execute (as root) /usr/sbin/locale-gen

Done so.

> First try the new locales on a console, for instance
>     LANG=de_DE date
> or your language till you see the messages in German.

Well, neighter nl nor bg (the ones I tried) work, but that's
OK, as I don't really want the messages to be translated:
I just want to be able to see letters from other languages.

> Then add them to /etc/environment or /etc/profile
>     export LANG=de_DE.ISO-8859-1
>     export LC_ALL=de_DE
> 
> The C locales and the Xlib locales are not the same and sometimes they
> seem to conflict. Any suggestion?

I can do so for both bg and nl (uncommented in /etc/locale.gen, and
yes, I did run locale-gen).
But if I run `xterm', it says: `Warning: locale not supported by C library, locale unchanged'.

Ah, now I see what I want: I have to _add_ the line 
  en_US UTF-8
to /etc/locale.gen, and then (after locale-gen) I finally get
UTF-8.

However, that's still not quite OK, as mutt now still doesn't
show mime messages that with ISO-8859-1 coding correctly:(.

I'm giving up.

> btw, in debian-l10n-spanish we are discussing a move to ISO-8859-15
> (how if not are we going to input and display the Euro?). Is it being
> considered by other national groups?

What's wrong with UTF-8? (I just don't know anything about this all,
but hear that UTF-8 is `the best').

Thanks,
-- 
joostje