From: Noah Levitt <[EMAIL PROTECTED]> > Should a combination like LANG=fr_FR LC_CTYPE=en_US.UTF-8 > result in something equivalent to LANG=fr_FR.UTF-8?
> If not, is there a good reason why not? > If so, maybe we should separate encoding from language, make > one UTF-8 locale, and encourage people to use > LANG=whatever LC_CTYPE=UTF-8. Yes, we should definitely separate settings of encoding from language. I think it's the same problem I tried to address a few weeks ago here, without much positive feedback, unfortunately. From: "Maiorana, Jason" <[EMAIL PROTECTED]> > what about > LANG=fr_FR.UTF-8 > LC_CTYPE=en_US.UTF-8 > ? In addition to the contradiction between "fr" and "en", the problem with any such setting (also the proposed LANG=ft_FR, LC_CTYPE=UTF-8) is that some operating environments, especially SunOS, don't accept such settings as they think of the combination of language and encoding as being some strictly defined entity they call a "locale". This conception and its rigid implementation is very obstructive to further promotion of UTF-8. To tie these two settings together in the way e.g. SunOS does it means that normal users cannot configure their software to use UTF-8 if they want a locale for which their local system administrator didn't deliberately decide to (compile and) install a UTF-8 version. I hope that there will be an insight at Sun to liberate the use of UTF-8 and that other environments (glibc?) don't follow this bad example. Thomas Wolff -- Linux-UTF8: i18n of Linux on all levels Archive: http://mail.nl.linux.org/linux-utf8/
