Matthias Buelow <mkb@mukappabeta.de> wrote:

>  * files with filenames containing non-utf8 8-bit characters couldn't
>    be accessed anymore, neither in the shell nor in the filemanager
>    the user in question was used to (kde konqueror)

This is a generic problem for any pair of encodings.


>  * documents containing 8-bit non-utf8 characters, such as tex files
>    with iso-8859-1 characters, couldn't be used anymore, because either
>    the text editor (emacs in that question) recognized the utf8-locale
>    (can't remember actually if it did) and then misinterpreted the
>    8-bit characters in the file

Sticking -*- coding: iso-8859-1 -*- where emacs can recognize it
(first line of the file, usually) should help in this case.  Again,
this is in no way unicode specific, I use this for the mix of koi8-r,
cp1251, cp866, latin1, latin2 and utf-8 that I need to deal with.


SY, Uwe
-- 
uwe@ptc.spbu.ru                         |       Zu Grunde kommen
http://www.ptc.spbu.ru/~uwe/            |       Ist zu Grunde gehen