[SCIM] Re: [m17n-lib:00079] Proposal: add minput_reset_ic (MinputContext *ic) method.

Choe Hwanjin choe.hwanjin at gmail.com
Tue Dec 21 18:14:55 PST 2004


On Wed, 22 Dec 2004 10:13:14 +0900 (JST), Kenichi Handa <handa at m17n.org> wrote:
> In article <604f48110412202355306b358c at mail.gmail.com>, Choe Hwanjin <choe.hwanjin at gmail.com> writes:
> > I mean general korean input method. Generally, korean does not use
> > romaja input method. We are using korean input method which has
> > keyboard layout 2beol, or 3beol etc, ie nabi or imhangul. With this
> > input method, we don't need explicit commit key.
> 
> Thank you for the explanation.  I checked nabi.  It seems
> that they use different key for CHOSEONG (initial consonant)
> and JONGSEONG (final consonant) and thus the input methods
> are deterministic.  Correct?

No, and partially yes.
There are 2 types of korean input method. 2 set system and 3 set
system. 2 set system (2-beol-sik) use same key for choseong(L,leading
consonant) and jongseong(T, trailing consonant), so input method
should determine the syllable boundary automatically. However, the 3
set system (3-beol-sik) use different keys for choseong and jongseong,
so with this method, the syllable boundary is deterministic.
Nowadays, 2 set system is more popular than 3 set system.

> Long ago, when I implemented a Korean input method in Emacs,
> I heard from a Korean person that they prefer using the same
> key for them, for instance as this:
> 
>   한글:   ㄱㄲㄴㄷㄸㄹㅁㅂㅃㅅㅆㅇㅈㅉㅊㅋㅌㅍㅎㅏㅑㅓㅔㅖㅕㅗㅛㅜㅠㅡㅣㅐㅒ
>   KEY:  r R s e E f a q Q t T d w W c z x v g k i j p P u h y n b m l o O
> 
> So, to distinguish "가가" and "각ㅏ", one have to type
> "rkrk" and "rkr<Ctrl-SPC>k" respectively (Ctrl-SPC is the
> explicit commit key).  But, as the latter case is not that
> frequent, this system is better than assigning different key
> for CHOSEONG ㄱ and JONGSEONG ㄱ.
> 
> Do you think that this argument is not correct nowadays?

Yes, I think so.
Almost every korean use 2 set input method and some people use 3 set
method. And neither method use commit keys. 'rk' for '가' mapping is 2
set layout. And 2 set method always convert 'rkrk' to '가가' never '각ㅏ'.
2 set method determine the syllable boudary automatically.

You can see how this input method is work in this animated gif:
http://paero3.myzip.co.kr/sebeol_keyboard/garam.gif

To input '가람':

ㄱ + ㅏ + ㄹ + ㅏ + ㅁ

2 set method cannot determine the syllable boundary untill 'ㅏ' is
inputed. But 3 set method don't need to wait next jamo to determine
the boundary, because it is obvious on 3 set method.


More information about the scim mailing list