From: Philippe Verdy (verdy_p@wanadoo.fr)
Date: Mon Mar 01 2004 - 17:24:01 EST
From: "Peter Kirk" <peterkirk@qaya.org>
> Aha, here's my way to get the characters I want into Unicode although
> they have been rejected! I find some near-bankrupt island state and
> persuade (with a little financial lubrication) its government to set up
> an official standards committee with me as the chair. I then issue an
> official national standard including the characters I want to get into
> Unicode.
Possible way to get it... Provided that this country has a chair for its
national standardization body in some international standard body. That country,
once it promotes its national standard would have to explain to other countries
why it needs it and why others should accept to manage with it.
Your rarely used characters you seem to want are probably not a high priority
for this government, to costly discuss these issues in international standard
bodies (think about the cost of air trips, hotels, personnels...)
> And, from what you say, Unicode will be obliged to accept my
> characters.
> :-)
Tajik is an important language for Tajikistan as it's an official language. It
can't be ignored if Tajikistan chooses to represent it the way it wants for its
legal publications...
> As for Tajikistan, don't go off into wild speculations about what they
> might want, but let them say.
I'm not speaking about them. It's just that it is a legitimate wish to have a
usable 8-bit encoding standard IN ADDITION TO Unicode, simply because there are
so many softwares today (and still for long time) that will depend on, the
capability of representing a language efficiently with a 8-bit coded character
set.
If you think that they MUST use Unicode for Tajik, when you are given the choice
for English and French so that you can use the cheapest and most widely
available software and hardware solutions you want, you put this language and
the country that uses it in a very unfair situation.
Yes they can start using Unicode each time it's possible. But lack of money for
newer developments will still require them to use the tools they have today,
including lots of legacy software made for 8-bit character processing systems.
It's really a question of cost, and an important and justified economic
decision, notably for a country which does not have the same financial power as
US which can benefit of the largest collection of software and hardware
solutions in the world.
I wouldn't object against another 8-bit charset provided that it is well-thought
and correctly matches the language need, and it can really be used as a standard
without too many variants (think about the nightmare of the various 8-bit
charsets for Russian used today, and now look at what a Tajikistan may think
about this legacy past where they may have today to manage various softwares
that work more or less well with ISO-8859 Cyrillic, KOI-8U, KOI8-R, ...)
I wish they could use for now the most widely accepted standard for Cyrillic
which is the Cyrillic variant of ISO-8859, and forget about other KOI-8 like
charsets, but still this charset does not fit all their needs, and going to
UTF-8 would also cause them processing problems or incompaitiblities (need to
patch the used softwares, if their sources are available), which could be easily
avoided by using a nationally approved variant of the ISO-8859 Cyrillic charset
(with the sam level of variation as between ISO-8859-1 and ISO-8859-15, i.e.
just a few characters they need and that would replace characters they don't
need).
This archive was generated by hypermail 2.1.5 : Mon Mar 01 2004 - 18:00:02 EST