Is 0 > None?? (fwd)

[Marcin 'Qrczak' Kowalczyk]

Tue, 4 Sep 2001 09:06:32 +0200, Radovan Garabik <spam at melkor.dnp.fmph.uniba.sk> pisze:

>  : Such program would currently not work *at all* - hardly an improvement.
> 
> At least it forces the author think about the issue - not just
> assume everybody uses latin1 (as is the case in many programs).

The current way lets him assume that everybody uses ASCII (as is the
case in some programs). I said wrong: the program will work as long
as all data supplied is ASCII. This is bad, and by your measure:
worse than a program which doesn't work at all (e.g. doesn't allow
conversion between Unicode and byte strings).

I don't propose to assume latin1! I propose to assume the encoding
which has been set in the system configuration for this purpose:
for assuming the encoding of 8-bit texts appearing in the system
which were not marked otherwise.

Text files in my system are not ASCII. They are mostly ISO-8859-2.
Filenames too.

> well, I would prefer utf-8 encoding - this is what one can expect
> to use when there is already need for unicode strings.

And thus I would set my locale to UTF-8. This is what it is for!
Now all correctly written programs will assume UTF-8. Except
programs written in Python.

-- 
 __("<  Marcin Kowalczyk * qrczak at knm.org.pl http://qrczak.ids.net.pl/
 \__/
  ^^                      SYGNATURA ZASTĘPCZA
QRCZAK