Stuart & Steve,
>As for Unicode... To get Unicode support using MS products, (at least in
>Japan) you have to use Win-NT. 95 and 98 are not unicode-compliant, so if
>you want to use Unicode as your character set, you'd have to convert to
>Shift-JIS (which Win98 speaks), JIS (Mac? Maybe. Unix - E-Mail is
>basically JIS), or EUC (Unix) at every HotSync... Not a pretty sight.
Win95 & 98 both support Unicode, but to varying degrees. Win98 is pretty
complete. MacOS uses Shift-JIS.
>A bit off-topic, but the Japanese front-end for the Psion makes use of
>UTF-8 (I think) unicode, and their synchronization software (when it
>comes out) is supposed to convert back and forth every time. Why they
>can't just use the same charset is beyond me, so don't ask <g>
Supporting multiple simultaneous character sets (ala MacOS) is hard work.
So if you're going to support a single character set, then Unicode is a
good choice. Unfortunately you still have to interface with the desktop
world, where Unicode is the exception rather than the rule.
Using UTF-8 means you can support Unicode using a narrow (byte-oriented) OS
API vs. creating a new, parallel "wide" API.
>On Thu, 26 Aug 1999 00:54:17 -0700
>Steve Sabram <[EMAIL PROTECTED]> wrote:
>
>> What you are looking for is called Unicode. It is a 16 bit character set
>that
>> tries to pull in all of the major world alphabets into one group. Unicode
>> characters are known as "wide characters" in the Windows world and all of
>the high
>> level coding (i.e. CString, COM interfaces, VB) stores in Unicode.
Actually I think what he was referring to was a multi-byte character set -
for example Shift-JIS, where every character is either one or two bytes.
With Unicode and the UCS-2 encoding scheme, every character is always 2
bytes, while with the UTF-8 encoding scheme, every character can be 1, 2,
3, or even four bytes.
-- Ken
Ken Krugler
TransPac Software, Inc.
<http://www.transpac.com>
+1 530-470-9200 (direct) +1 408-261-7550 (main)
