Re: glib time functions

[Ali Abdin <aliabdin aucegypt edu>]

* Robert Brady (robert suse co uk) wrote at 18:23 on 30/11/00:
> On Wed, 29 Nov 2000, Federico Mena Quintero wrote:
> 
> > We should do the best we can to support Islamic calendars and any
> > other types of calendars that are in widespread use.  This may be
> > technically hard, just as supporting languages with complex ligatures
> > is a hard thing to do (Thai et al).  But we cannot just say "screw
> > them"; remember that the world is not constrained to your own
> > particular culture.
> 
> Not just hard; impossible.
>
> We shouldn't pretend we can solve this problem; we are no magicians.

Umm - I said that we can not ever be 100% accurate. That doesn't mean we can't
be damn close. At worst the glib functions will be off by one day for a period
of 3 months (although I did propose a possible way to get around this which you
haven't responded to yet).

This beats "no implementation at all". Like I said, programmer and users in
the Middle East will go somewhere where it IS supported (i.e. Windows).

Or even worse, people will implement their own implementations in their
libraries. Imagine every single person implementing their own Gregorian date
handling functions (remember two-digit years, Remember the fact that many apps
didn't recognize 2000 as a leap year, companies then released patches to fix
this (but managed to fuck it up and make it recognize the year 1900 as a leap
year)). 

We are not magicians, no. We are computer hackers that can provide a decent
(but not 100% accurate) technical solution to a particular culture/religion.

Regards,
Ali