🚀 go-pugleaf

mohta@necom830.cc.titech.ac.jp (Masataka Ohta) writes:

> In article <DwqPwB3w165w@blues.kk.sub.org>
>       kosta@blues.kk.sub.org (Kosta Kostis) writes:
> 
> >I agree that UniCode is not very good for Asian languages, but
> >for European languages (and some more) it's really OK.
> 
> Then, could you get out of here, comp.std.internat?

Masataka-san, if you are unable or unwilling to understand my posting,
please refrain from pointing your publicly known flame thrower at me.
Thank you very much, indeed.

> >10 bits? You want to keep Asian languages out of the game, too?  :-)
> 
> Or, you want to keep Asian languages out of the game, too?  :-|

Iie, soo dewa arimasen.

Is it possible that you have become a little "paranoid" during
your "fight" against 16-bit UniCode.  =;^)

What makes you believe that I want to keep Asian languages "out, too"?
It's the other round. 

My statement was that UniCode supports Latin-like language well while
Asian languages are *not* covered well. Does that imply I *like* UniCode?
No, so please calm down. Arigatoo gozaimasu.  :-)

> >Should
> >we ever decide to use the full 32-bits ISO 10646 intended to
> >allocate for character codes, we should be able to cover almost
> >all languages (to some extend).
> 
> Full 32 bit? 18 or 20 bits will be suffice for the real universal
> character code set.

Well, at least one that is "universal" enough from a Japanese point
of view  ;-), but maybe you're right. So, just to satisfy my curiousity:
how do you represent 18 or 20 bits with 8-bit octets?
Wouldn't be there be some "wasted" bits?
Please don't let us make the same mistake that was done when SMTP was
designed to cut off the MSB when discussing a "universal" character set.

> >Who defines "required"? This is a classical "don't care about the
> >users culture/needs" approach that was so harmfull in the past.
> 
> ISO 8859-1 is not very good for all European languages, but
> for central European languages (and some more) it's really OK.

Right, but there are more character sets needed like ISO 8859-7 for
Greek (I am a Greek), ISO 8859-5 for Cyrillic (allthough Vadim doesn't
like it), ISO 8859-8 for Hebrew, ISO 8859-6 for Arabic and many more.

I want to be able to compose a document containing all the languages,
many more *including* the Asian languages. That's what leads us to a
"universal" character set. Note, that I have never claimed, that 16-bit
UniCode *is* the "universal" character set I'm looking for.

As I stated before, I think 16-bit UniCode is one step "aside" and one
into the "right" direction. Does that sound like "fandom"?

> Then, you wrote:
> 
>       I agree that UniCode is not very good for Asian languages, but
>       for European languages (and some more) it's really OK.
> 
> What was so harmful in the past? What you are repeating?

Well, the same thing that all non-english speaking people suffered
and still suffer from because of the fact that the U. S. were and
still are leading and ruling the computer industry.
Don't get me wrong, I have no objections against the U. S., they
just ignored the needs of their international customers for much
too long.

We suffer(ed) from Systems, that don't support anything other than
ASCII or at least nothing more than *one* 8-bit character set (like
ISO 8859-1, which is better (for me) than US ASCII but not enough).
In Japan you have your own local solution which may be the better
for you than 16-bit UniCode, but that isn't "universal" either.

I think we can agree we all should want to have a "universal" character
set. At least as a basis for *international* data/document interchange.

> >The other languages you stated, like Russian, Greek and English
> >seem to be served well by UniCode, I think.
> 
> But, as for reg-exp pattern, Vadim's suggestion is quite right.

What do you use reg-exp patterns for? I think it's irrelevant or at
least no real problem. Local versions for that will never go, I believe.
Do you know any broadly used computer language that supports more than
US ASCII in qualifiers, names, etc.?   :-)

If you want to stay with two "alphabets" fitting in 8-bit, you can
always convert your ISO 10646 data to KOI-8 (or whatever) and use
your old programs. Right, Vadim?   ;-)

	Kosta

PS: I am *not* a "UniCode"r, but I strongly feel the need for a universal
    character set and I want it *implemented* in *this* century. :-)

-- 
  Kosta Kostis, Talstrasse 25, D-6074 Roedermark 3, Germany
  kosta@blues.kk.sub.org                                        (home)
  sw authors: please support ISO 8859-x!      äöüÄÖÜß = aeoeueAEOEUEss