Dear Masayuki TOYOSHIMA
You wrote:
>
>A recent essay titled "Encoding Korean Hangul" by T.J.Kang, appeared
>on the "EnCoding" (Newsletter of the Unicode Consortium) spring 1995
>issue, tells that current code-points of the syllabic (pre-combined)
>forms of Hangul will be removed, and a new (contiguous) mapping scheme
>will be taken.
>
>This is apparently contradicting to the declaration by the Unicode
1.0; i.e.
>
> future versions of the Unicode standard will not reassign, rename
or
> remove Unicode 1.0 characters. (The Unicode Standard 1.0,
p.xix/14)
This continues to be the stated policy of the Consortium. UTC will
resist any efforts to fine-tune the repertoire and allocation by
"correcting" names, "improving" the arrangement of characters, or by
deleting "mistakes". However, exceptions to this policy are made in
what essentially are 'emergency' cases. One such was the strong and
unequivocal desire by customers and industry to avoid duplicate
standards. Therefore, Unicode agreed to merge with ISO 10646, even
though this entailed small changes to the original Unicode design.
(These changes are documented in Unicode Technical Report #4, also
known as Unicode 1.1). The character coding working group ISO/IEC
JTC1/SC2/WG2 has a similar understanding of the need to preserve
ISO 10646. However, because of a historic transition to a larger
character set in Korean computing, which happened in parallel to the
final stages of ISO 10646 approval, editing and publication, the
Korean support in the existing standard was hopelessly out of synch.
After long and agonizing discussions, Unicode, and separately ISO's
WG2 realized that this needed to be fixed. The problem was that a
piecemeal addition to the Korean repertoire would have resulted in
a truly bizarre and quasi-random arrangement of character codes,
impacting implementers for generations to come. When it was determined
that there were no large scale existing implementations, the matter
of removal was put up to a vote. (If there had been large scale im-
plementations, this would not even been considered, and if considered,
any vote would have failed). This vote has recently been concluded and
the results are that the new arrangement has been _approved_ in ISO
(Unicode held a ballot as well, before submitting the matter to ISO,
and the new arrangement was approved there as well). Unicode will
reflect the new arrangement when the upcoming version of its book will
be printed in 1996.
If you are interested in more detail, you might contact the convener of
WG2 or your national standards body (JIS in Japan, X3L2 in the US) to
obtain the copy of working paper N1199 which is the background paper to
the ballot on Hangul.
In the meantime, expect implementations of the new Hangul arrangements
to be available in Korea pretty much immediately.
In summary, this was not an easy decision in any of the groups involved
and few people have any desire to repeat this exercise anytime soon. In
fact, if there is a precedent, it is that the committees are absolutely
unwilling to touch anything that has already seen widespread
implementation. So if you have an implementation of the Unicode
Standard, contact unicode-inc@unicode.org, and have it entered in the
"Catalogue of Implementations of the Unicode Standard". This will help
the committees to protect your interests in a stable standard.
A./
Asmus Freytag
Unicode Liaison to WG2
>The essay admits that "a rationale must is needed", but gives none.
>
>If there is a published rationale for this decision, I would strongly
>like to read it.
>
>Thanks in advance :
>
>Masayuki TOYOSHIMA / Dept. of Japanese linguistics, Faculty of
Letters,
>Hokkaido university, Sapporo 060 Japan / mtoyo@Lit.hokudai.ac.jp
>
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:30 EDT