Re: use of locale in "ls" again (Re: Japanese expression of date)

[Tomohiro KUBOTA]

Hi,
At Thu, 20 Dec 2001 16:42:53 -0800 (PST),
Paul Eggert wrote:

> >    Standardization way is possible only when people all over the world
> >    agree.
> 
> Yes, but we don't need complete unanimity.  We need only sufficient consensus.

Sorry for ambiguous words.  I mean, a certain percentage of people from
almost all countries.

> This is true, but only up to a point.  For example, the GNU Emacs
> maintainers could recommend GNU 'ls' to the relatively few users who
> have unusual date formats that GNU Emacs can't parse.  This would
> remove the need for GNU Emacs to support every date format on the
> planet.

If GNU Emacs recommends GNU 'ls' and the GNU 'ls' outputs locale-
independent timestamp, what is the advantage to that GNU Emacs
recommends GNU 'ls', the GNU 'ls' outputs locale-sensible 
timestamp, and GNU Emacs invokes 'ls' with LC_ALL=C ?  Thus,
your plan cannot be a reason to recommend locale-independent
'ls' behavior.

Similarly, Markus' opinion doesn't explain why locale-independent
'ls' is better than locale-sensible 'ls', while it insists locale-
independent 'ls' is not so bad.

---
Tomohiro KUBOTA <address@hidden>
http://www.debian.or.jp/~kubota/
"Introduction to I18N"  http://www.debian.org/doc/manuals/intro-i18n/