Re: Unicode & Han

From: Martin J Duerst (
Date: Sun Aug 11 1996 - 12:13:00 EDT

Timothy Huang wrote:

>Dear Michael,
>About the Microsoft, I'll give you the source of the news when I can
>find the clip. Right now, at the moment of writing this reply, it's in
>one of the piles somewhere in my corner. When I find it, if you give me
>your fax number, I can even fax you that. I saw the news, I think
>somebody may be interested to know.

This may refer to two things:

- GBK, a method of squeezing all CJK ideographs and lots of the rest
        of Unicode into a 16-bit code that leaves GB codes at exactly the
        same places they are. This can be seen as a step back from
        Unicode, or as a step forwards to Unicode.
- XKP, a collection of software and specifications developped in Japan
        (and maybe also used or promoted in China) to include very
        rare characters or character variants in texts using the OCX
        mechanism. Again, some interprete this as an admission, by MS,
        of a major flaw in Unicode, others just see it as a commitment
        to Unicode.
Anyway, there will always be some gap between the characters in
a standard and the characters somebody needs. If Prof. Huang
discovers a new element, and wants to write a paper about it, he
has to use the private zone and related mechanisms to print it,
he cannot wait for the standard bodies to have it accepted.
(This means that first, it has to be accepted by the relevant
chemical authorities, and then by the character standards.
I don't know for Chinese, but for English/Western names,
there are still heavy disputes about how to name certain
newer elements.)

>Speaking of Ma Jong font, if you like, I think we can make one. I can
>easily scan the images, and you can use some font making software, such
>as fotographer, to make them a MaJong font.

You just seem to confirm what we suggested as the reason there
are no Majong codepoints in Unicode: There is no Majong font
currently, and no other font containing Majong glyphs. This
seems to indicate that there is no real need for these. I could
very well understand this, Majong tiles, as well as Chinese and
Japanese chess pieces, all have their names, and it's these names,
written with ordinary Chinese characters, that appear in texts,
and not the pictures of the pieces themselves, as this is quite
common for Western chess.

>In my understanding on the character coding issue, there is already an
>existing American National Standard, EACC -- ANSI Z39.64, used by the
>libraries worldwide. Why, then, putting so much efforts re-inventing the
>wheel? The originator of the EACC, the CCCII already compiled 75,684
>ideographic characters (in 1989). Why couldn't we start from that and
>support that?

EACC was indeed a very important base for developping Unicode.
This is described in the Unicode book, anyway.

Regards, Martin.

This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:31 EDT