On Mon, 29 Jan 2001, Ienup Sung wrote:
> The "much better" in that context was the client programs won't need to
> do any iconv(3) stuff if your input and output will be only in UTF-8.
>
> As I wrote in the previous email though (and other emails related to
> Xutf*() discussion), I also agree that there are cases that users want to
> input and output with their current locale's codeset plus UTF-8/UTF-32 but
> that's not the case that I was mentioning about in the email.
>
> Also, as I pointed out so many times in Xutf*() discussions,
> I also think and believe that each user has freedom of choosing whatever
> codeset and locale that he or she wants to use with his/her applications
> that nobody dare can dictate and we also shouldn't force them to choose
> any one codeset/locale.
But you yourself are advocating forcing people to use UTF-8 locales for
certain features, as indicated below. The point of adding Xutf is to
allow these locales to be continue to used.
> ] > My point was, if UTF-8/UTF-32 is the only multibyte/wide character
> ] > representaion that you need in an application, it will be much better
> ] > just to use xx_YY.UTF-8 locale of your choice instead of using xx_YY.ZZZ
> ] > and then do the iconv(3C) at the client side.
--
Robert Brady
[EMAIL PROTECTED]
-
Linux-UTF8: i18n of Linux on all levels
Archive: http://mail.nl.linux.org/lists/
