L2/00-148

From: Kenneth Whistler [kenw@sybase.com]
Sent: Friday, March 31, 2000 3:16 PM
Subject: It's Springtime -- Let a Thousand Grasses Bloom!

Unicadettes,

Well, it's that time again. And I thought I should bring
everybody up to date regarding my favorite Unicode character:
'grass'.

Among the 61 compatibility characters for JIS X 213 now accepted
for encoding in the BMP, it turns out there are *another* two
instances of the grass radical. This brings our total to:

U+8278  CJK UNIFIED IDEOGRAPH-8278  <== *This* is the real character.
U+8279  CJK UNIFIED IDEOGRAPH-8279  <== The 3-stroke radical form
U+4491  CJK UNIFIED IDEOGRAPH-4491  <== The non-crossing 3-stroke radical form
U+2F8B	KANGXI RADICAL GRASS        <== Radical symbol that looks like U+8278
U+2EBE	CJK RADICAL GRASS ONE       <== Radical symbol that looks like U+8279
U+2EBF  CJK RADICAL GRASS TWO       <== The 4-stroke radical form
U+2EC0  CJK RADICAL GRASS THREE     <== A variant 4-stroke radical form
U+FA5E? CJK COMPATIBILITY IDEOGRAPH-FA5E <== Looks like U+2EC0
U+FA5F? CJK COMPATIBILITY IDEOGRAPH-FA5F <== Looks like U+2EBF

What is that, *nine* grasses now?

The last two, by the way, are the Shibano grass radicals, 
since they were apparently included in JIS X 213 at his
insistence -- and that is appropriate, I guess, since we all know
that "shibano" means 'grassy field'.

Apparently the Japan NB was under some pressure to not accept a
mapping of these two additional grass radicals to the already
encoded, identical-appearing CJK radical symbols. They had to
be "compatibility ideographs", and not "radical symbols". Urk?

I would suggest that UTC take another hard look at these two --
and the other two "compatibility ideographs" proposed in this
set that are actually radicals: FA4A = 2EA4, and FA66 = 2ECC.
If JIS X 213 actually contains two copies of all four of these --
one in a set of radical symbol forms, and the other buried in the
general set of ideographs, then maybe Unicode is also going to have
to live with these duplications. But something is rotten in
Den^H^H^HTokyo if we are being forced, after the fact, to keep applying
the source separation rule to obvious and ridiculous duplicates in
newly developed standards.

--Ken