tlug: Kinput2 and Wnn6

[Frank Bennett <bennett@example.com>]

I'm using the above with Applixware and Kterm, alongside Emacs
20.2 with Japanese support.  I've gotten Kinput2 to work with
similar keystrokes to the Emacs "egg" interface (thanks, Scott,
for your help on that one).

Now I've noticed another blemish on Kinput2.  When conversion is
done in Emacs, the conversion is recorded in the user dictionary,
so that, say, I only have to tweak わが輩は猫である once to be 我
輩は猫である; on subsequent occasions the second reading will be
presented as the first choice for the string わがはいはねこである.
This behavior is important to me, because a lot of the
expressions used in law are not found in common vocabulary and
need to be teased out of the server the first time around.

Unfortunately, when the jserver is accessed through Kinput2,
these learned entries are ignored.  It appears that Kinput2
entries are recorded (I'm just guessing, but the above string was
tried in Applixware, tweaked, and then tried again --- had to be
tweaked again; when it was then tried for the first time in
Emacs, the hit was correct the first time).  But not even the
strings that are trained up in Emacs turn up in the same form
through Kinput2.

I can't see any good reason why Kinput2 should exhibit this
limitation.  Is there a fix?  A configuration option?

Scott? ... :)
-- 
-x80
Frank G Bennett, Jr         @@
Faculty of Law, Nagoya Univ () email: bennett@example.com
Tel: +81[(0)52]789-2239     () WWW:   http://rumple.soas.ac.uk/~bennett/
---------------------------------------------------------------
Next Meeting: 10 October, 12:30 Tokyo Station Yaesu central gate
Featuring the IMASY Eng. Team on "IPv6 - The Next Generation IP"
Next Nomikai: 20 November, 19:30  Tengu TokyoEkiMae 03-3275-3691
---------------------------------------------------------------
Sponsor: PHT, makers of TurboLinux http://www.pht.co.jp