안녕 하세요 신정식씨?

On Tue, Feb 24, 2004 at 08:43:09PM +0900, Jungshik Shin wrote:
>   Please, read what I wrote more carefully. I did write that deleting
> the last letter is more useful when you're in the middle of typing a
> sequence of letter to form a syllable. 

I think we're talking past eachother here...  I noted that and I agree
with it.  It's specifically the fact that once I type the third
character of a hangeul glyph, I can't backspace and change ONLY that
last character, that annoys me.  You say that most Koreans prefer that
behavior, and I believe you.  But I can't for the life of me
understand why...  ;-)  To me, it seems unnatural and inefficient.

Almost invariably once I've committed an erroneous syllable, it's not
the whole syllable I need to replace, but only the last character
which I flubbed.  Otherwise, if I made a mistake before the syllable
was committed, I already fixed it.  [Assuming, of course, that it's a
typing mistake, and not a language mistake... ;-)] The default
behavior of Ami seems strange to me, in that regard.   But it's
probably just because I've been typing western alphabets all my life,
I guess...


> Once a syllble is committed into the backing store, however, most
> Korean people want the cursor movement, the selection and editing
> operations like deletion/insertion to be done syllable by syllable.

Sure... I can see that it's useful, but it would be nice to be able to
do both.

> However, incremental search needs to be done with individual letters
> as unit instead of syllables. I think Indian people have similar
> needs.

알겠어요.  

>   These behaviors are default with XIM servers for Korean like 'Ami'
> (http://kldp.net/projects/ami ) or 'Nabi' (http://kldp.net/projects/nabi).
> 
> > blunt, I find that really annoying, and if there's a way to change
> > that behavior, I certainly would like to know how...
> 
>   What input method server do you use? The msg strings for Ami are available
> in English, too.  That is, setting LC_MESSAGES to en_US.UTF-8 gives you
> English menus in Ami.

Indeed.  I'm using Ami, and what you say is true.  The trouble is that
if I try to set LC_MESSAGES to a non-korean value when I start some
other program, then the application in question seems to not work
properly with ami.  I'll get to that in a moment...

> > can't see how...  Perhaps my biggest problem is that I can't find any
> > documentation about using Korean with Linux which isn't written in
> > Korean.  Which is all well and good, if you already happen to speak
> > Korean fluently...  ;-)
> 
>   I used to post 'Hangul and Internet in Korea FAQ' to
> soc.culture.korean regularly, but that's a way too outdated by now.
> Pls, feel free to ask me off-line if you have any problem.

Well, I will take you up on that.  I have a few...

> > It seems like a perfectly viable solution.  But I can't help but think
> > that it would be better if the kernel allowed for language-specific
> > IME modules in the console/tty drivers.  Then you could deal with it
> > uniformly at all levels of input management...  One API to enter
> > characters, whether you're typing in a terminal emulator or at the
> > console.  What I'm essentially envisioning is that all input
> 
>   It's not for kernel, but you may find it interesting to know more
> about IIIMF and SCIM. http://www.openi18n.org/subgroups/im/IIIMF/

I'll have a look.  Since I moved to Korea, Internationalization and
multi-language support has (understandably) become more interesting
and important to me...  :)

> > about the right way to be able to enter hangeul, while still
> > maintaining English menus and messages and such.  So far, my research
> > has turned up precious little, and I have only been able to type in
> 
>   Well, it's easy. I always do that because I don't like the quality of
> Korean translation in most software, commercial or open-source.  Add
> this to your ~/.i18n (or equivalent. ~/.profile )
> 
> ----------
> LANG=en_US.UTF-8  (or en_GB.UTF-8, en_CA.UTF-8)
> LC_CTYPE=ko_KR.UTF-8
> LC_MESSAGES=en_US.UTF-8 # not necessary unless LC_ALL is set, but
> LC_TIME=en_US.UTF-8     # just to be sure.
> -----------

Hmmm...  I've been managing Unix systems for years in a multi-language
environment for part of that, so I'm no stranger to locales.  I more
or less do this, but I so far have found that it doesn't work.  Well,
honestly I use an all-english locale generally, but then I start
specific applications with a Korean locale.  Specifically, I do this:

  # .profile (or whatever)
  LANG=en_US.UTF-8
  LC_COLLATE=C  # I like ASCII sorting for most applications...
  ...
  export LANG LC_COLLATE ...

Then, when I start up an application where I want to type Korean, I
originally tried startiing it like this:

  $ LANG=ko_KR.UTF-8 LC_COLLATE=ko_KR.UTF-8 LC_MESSAGES=en_US.UTF-8 gedit&

I also tried other combinations of both UTF and non-UTF locales.  One
of two undesireable things always happened:

1. Menus were in Korean
2. Hangeul input via ami simply didn't work.

Now, I admit I haven't reade the locale man pages in a long time, so
it may be that this particular combination of variables isn't
quite right to accomplish what I want to do.  But it seems like it
should be sufficient.

As it happens, until recently the most common case I want to do this
was with mozilla.  It wasn't a major problem then, because my
installation of Mozilla had no Korean.  But as my Korean improves, I
have more and more cases where I want to do this.  Of course, I'm also
better able to navigate the menus, but that's beside the point...  :)

I'll play around with my locale settings again, and I'll get back to
you with the results.

Thanks,
Derek

Attachment: pgp00000.pgp
Description: PGP signature

Reply via email to