> > I thought it had been assigned code point U+20A0 EURO-CURRENCY SIGN a
> > long time ago? (It's listed in Unicode 2.0; I am pretty sure it is
> > older than that.)
> You are joking, right? Do you really expect people and
> programmers to start using and supporting Unicode in the near
> future?
Partially agreed -- Now the 8-bit character sets are simply defined as
subsets of Unicode, the definition must be done first in Unicode (at
least, the ISO-8859-x and its ECMA (43 and following) counterparts were
looking this way last I checked, about a year ago. From what I understood,
the ECMA has the effective responsibility for the ISO-8859 serie)
My brother and I tried to have emails from the Commission's DG X about
what were they doing about normative changes to support the EUR symbol
into existing charsets -- got no responses besides "we're following up to
the right person"... then, a few month later (around september ?) MS came
up with something (which does look like an unilateral move) :
http://www.microsoft.com/typography/faq/faq12.htm
(where one can see something interesting : they have chosen positions for
the Euro symbols in Windows 125x charsets (which happen to be close to the
ISO 8859-x but not identical), but they'll implement them only in Win98,
NT5, perhaps the SP4 for NT4. Nothing for Win95 unless people use Word97.
Now talk about dominant position abuse <grin />)
Having checked back this afternoon, the CEN/TC304 seems to have some
activity on the topic : http://www.stri.is/TC304/Euro/eurostatus.html
>From there, one can read the JTC1 has decided last July to add the Euro
symbol to the Latin1 table... Curiously, the US rep at this commitee
voted against the inclusion (the reasons why aren't stated, though).
There's a joint proposal proposal from the AFNOR (Association Francaise de
NORmalisation) and the Conseil canadien des normes (sorry, I don't know
the official English name) to amend the 8859-1 in order to replace
character 10/04 : CURRENCY SIGN into 10/04 : EURO SIGN (that's 0xA0)
(among others), as in http://www.stri.is/TC304/Euro/ISOIEC8859 .
Curiously, this document mentions the initiative to put the EUR symbol
into 8859-1 as a Canadian one (!).
On Feb 3rd, the TC304 seems to have seen pretty hot battle over whether to
support at all the EUR symbol in 8-bit charsets (some of the members
wanted to promote 10646-only), while some even talked about 7-bit or
EBCDIC... No definitive solutions were approved or even drafted that day.
I haven't checked much more, but although I have the feeling they're late
and have really poor PR staff, something's being (or has been) done.
Perhaps sending a mail to the ECMA may be a good idea (they were very
responsive last year).
> list.) If the transition from seven to eight bits causes this
> much pain, I don't see Unicode coming very soon. At least not
> if the only driving force is one single symbol.
Well, good standard frameworks are a real help -- if for instance the
bleeding edge GNU tools and coding practices (such as the ones from GNOME)
require 1) localisability 2) usage of a specific format for
localisable strings (the N() or L() macros) and 3) unicode or at
the bare minimum UTF-8 awareness (with probably an updated catgets() et
al), things are probably to move a little bit faster. Also, the now
widespread use of MIME encodings have practically solved what used to be a
quite common mail gateway disaster (anyone using any continental European
language has seen at least once mail turned into garbage by 8-bit-vs-7-bit
MTA wars). Finally, there's not only GNU development tools ; Java is said
to talk only Unicode, MS is regularly warning that it will push Unicode
real hard once Win 98 begins to phase out (yes, that's pure vapourware.
Hello, Consumer NT 6.5 !), but we'll never see that happen : Linux and
Hurd will be the World Dominators(tm) by then <grin />. Etc.
> Well, maybe the year Sweden joins the EMU...
We may have to support the change in Central European charsets, perhaps
even in KOI8-Unified before <grin />
-- Cyrille
------------------------------
Zog Zog
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu