Ken,
I agree.
Any one who was an original Unicode evangelist with the loose leaf Unicode
1.0 binder in hand knows that if it were not for UCS-2 that Unicode would
not be used today.
It was a risk for MS to use Unicode in NT
It was a risk for MS to partially implement Unicode in Win95.
It was a risk for Sun to develop Java based on Unicode.
It was a risk for IBM to develop ICU.
These were all UCS-2 implementations that are migrating to UTF-16.
If you are a "Johnny come lately" now that memory and storage are cheap and
communications are fast, implementing UTF-32 or UTF-8 is often the way to go
now. But they must remember that they are standing on the shoulders of
visionary people who were pushing Unicode at a time when people were just
getting used to the marvel of code pages.
Some people were wondering why packages like ICU remain UTF-16. There are
several reasons. I have found that you don't have to add much code to make
them an excellent UTF-8 ore UTF-32 support system. Also if you have
different forms of Unicode to support UCS-2 was simple but now good UTF-16
support is the hardest so I am glad they are doing it. Otherwise they would
probably use UTF-32. However using UTF-32 would double the size of the data
with is most of ICU's size.
There are reasons for choosing different UTF types and the answers are not
straight forward. The only clear cut choice is using UTF-8 for intersystem
data.
Carl
> -----Original Message-----
> From: unicode-bounce@unicode.org [mailto:unicode-bounce@unicode.org]On
> Behalf Of Kenneth Whistler
> Sent: Friday, September 14, 2001 10:43 AM
> To: qrczak@knm.org.pl
> Cc: unicode@unicode.org
> Subject: Anti-UTF-16 Rant (was: Re: PDUTR #26 posted)
>
>
> Marcin,
>
> > IMHO Unicode would have been a better standard if UTF-16
> > hadn't existed.
>
> I won't repeat Asmus' rebuttal, which adequately addresses this
> claim.
>
> However, if people want to go off on the "UTF-16 sucks,
> UTF-8/UTF-32, the way Linux likes it, is far better, and
> besides, Windows sucks" tangent, please use some other subject
> line for the thread, so that any discussion about PDUTR #26 can
> focus specifically on the issues represented by PDUTR #26.
>
> --Ken
>
This archive was generated by hypermail 2.1.2 : Fri Sep 14 2001 - 22:36:59 EDT