-Z- (Z@Gundam.Com)
Thu, 10 Feb 2000 19:02:16 -0800

At 17:34 2/10/2000 -0500, you wrote:
>-Z- wrote:
>> Anyone encoding the kanji for "Kidoh Senshi" in Unicode should be advised
>> that I made a mistake on my Gundam High Frontier pages.
>Err, actually I did mention that mistake twice back on Jan 5 or so. I
>think I even sent you the correct phrase of Kido Sensi in both Big-5 and
>S-JIS encodings. But since we don't know about any freeware/shareware for
>converting between Unicode and existing encodings, I couldn't give you the
>correct Ki in Unicode.

I do not recall receiving these messages. I received at least one message
that was garbled to such a degree that it may as well have been encrypted
-- even the Subject and From lines were unreadable. Could that have been

The message that I received from Lersak today was sent as HTML and I was
unable to read it until I switched to the Microsoft viewer, something I'm
loathe to do because it's a vector for the recent batch of e-mail macro
viruses. With the plain-text viewer that I normally use, JIS and Big-5
encodings are garbage.

>Is there a more usable methods for entering Japanese and Chinese in
>Unicode now? Right now I use a pen to "draw" Chinese and kanas into Big5
>and then if necessary, run convertor programs to convert to GB (for
>Chinese) and SJIS (I wrote that one myself, works only for kanas). This
>method is reasonably fast for entering short Chinese and Japanese text, I
>can do about 15-20 words a minute. The big advantage is that there is a
>ZERO learning curve.

CJK (Chinese Japanese Korean) character entry has been the Holy Grail of
data processing since the days of the dedicated word processor. So far,
the best system that I've seen is UniPad, a Notepad type program with a
pop-up virtual keyboard for each character set. The CJK version, still in
development, will require nearly 100 such keyboards, each of which can
accommodate 256 characters.

Unicode is, of course, still in development; the CJK code set was just
recently introduced in Unicode 3.0 and being developed by a separate body
as the Unihan specification.

>There are good freeware convertors within each language, but I haven't
>found one for Unicode yet... Wait... NJStar claims to have one:
>I will try it some time.

The much-maligned Microsoft has been working to "internationalize" Windows
for some time -- they don't like having multiple code bases any more than
their customers do -- and decided to standardize on Unicode with Windows NT
4.0 and Internet Explorer. Office 2000 already supports Unicode, albeit
only up to 2.1 and this not yet inclusive of CJK. The Arial Unicode MS
(ARIALUNI.TTF) TrueType font, released on 7 April 1998, contains 51,180
glyphs encoded per ISO 10646-2 and supports Code Page 932 (JIS/Japan), 936
(Chinese Simplified/PRC & Singapore), 950 (Chinese Traditional/Taiwan &
Hong Kong), 949 (Korean Wansung), 1361 (Korean Johab) and all Unicode 3.0
ranges. They cobbled up the Input Method Editors (IME) for Internet
Explorer, but their goal is the Multilingual User Interface (MUI) that will
debut with Windows 2000 a week from now.

Expect the next iteration of Microsoft Office to incorporate MUI at the
application level and use Unicode tagged Dynamic HTML in place of Rich Text
Format (RTF) as the native document interchange format. Expect the Office
Millennium Edition to also run within a Web browser, with the applications
appearing only as context-sensitive toolbars -- when you're in a text
block, you get a Word toolbar; when you're in a table, you get an Excel
toolbar; when you're in a data field, you get an Access toolbar; and so
on. Similarly, you can specify a language and it'll shift to the
appropriate code page -- type French and the diacritical marks will appear
as needed, type Japanese and you'll get a table of characters from which to
pick the correct kanji or kana. It'll be very neat when they finally get
it working.


Gundam Mailing List Archives are available at http://gundam.aeug.org/

This archive was generated by hypermail 2.0b3 on Fri Feb 11 2000 - 12:09:09 JST