Received: by 10.36.80.14 with SMTP id d14mr3254271nzb.1186159627606;
        Fri, 03 Aug 2007 09:47:07 -0700 (PDT)
Return-Path: <ben.peter...@riskcare.com>
Received: from MAILONE.riskcare.com (gateway.riskcare.com [81.3.93.70])
        by mx.google.com with ESMTP id v28si846140nzb.2007.08.03.09.47.06;
        Fri, 03 Aug 2007 09:47:07 -0700 (PDT)
Received-SPF: pass (google.com: best guess record for domain of ben.peter...@riskcare.com designates 81.3.93.70 as permitted sender)
X-MimeOLE: Produced By Microsoft Exchange V6.5
Content-class: urn:content-classes:message
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Subject: RE: Fwd: VIM 7.1 compilation error
Date: Fri, 3 Aug 2007 17:47:11 +0100
Message-ID: <A70226A35219F24CB5AC282C89D2699101D151A4@xtranetserver>
In-Reply-To: <46B359B2.2010509@gmail.com>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: Fwd: VIM 7.1 compilation error
Thread-Index: AcfV7JfLcoDbkvjTS1+qPua7JsN8gwAABIwg
References: <22316227.444471186129992559.JavaMail.nabble@isper.nabble.com> <bcbc1a860708030216t75721c7fvd98954995a06da0f@mail.gmail.com> <46B3149A.1060109@gmail.com> <bcbc1a860708030742y4da084a5hf8a763d9deb107eb@mail.gmail.com> <46B34524.9070701@gmail.com> <bcbc1a860708030823m638667djda0fa306c358eade@mail.gmail.com> <A70226A35219F24CB5AC282C89D2699101D151A3@xtranetserver> <46B359B2.2010509@gmail.com>
From: "Ben Peterson" <ben.peter...@riskcare.com>
To: <vim_multibyte@googlegroups.com>
Cc: <vim_dev@googlegroups.com>,
	"Bram Moolenaar" <b...@moolenaar.net>,
	"vim-dev" <vim-...@vim.org>,
	<vim-multib...@vim.org>,
	<um.ki...@gmail.com>


Hi,

  True, one would have to make two separate attempts, one for old-style
and one for new-style Romanization. =20

  There are other problems too, now that I think about it:

  1)  Some Korean syllables have different final consonants when written
but not when spoken.  It is difficult to express this in romanization.=20
  2) Hundreds of mappings would start the same way as each other (e.g.
Se, Set, Seok, Seo, Seot, Seok etc) which might make entering them via
the mapping system a bit odd (would it pick 'Se' if you hesitated part
of the way through typing 'Seoul'?)

  3) As you know, neither romanization system is unambiguous because it
is sometimes impossible to tell where a syllable ends.  Normally you
wouldn't want to type 'Se-o-ul', but if you did, how would the keymap
tell it apart from 'Seo-ul'?  I think you'd need to actually type the
hyphens or something.

  Hm, this actually sounds very very hard :(

 Still I might generating a keymap (from the composed hangul character
area of the Unicode charts) over the weekend and see what happens.

  Benjamin


-----Original Message-----
From: vim_multibyte@googlegroups.com
[mailto:vim_multibyte@googlegroups.com] On Behalf Of Shawn Kim
Sent: 03 August 2007 17:37
To: vim_multibyte@googlegroups.com
Cc: vim_dev@googlegroups.com; Bram Moolenaar; vim-dev;
vim-multib...@vim.org; um.ki...@gmail.com
Subject: Re: Fwd: VIM 7.1 compilation error


Ben Peterson wrote:
> Hi,
>
>   Wouldn't hangul be a very big keymap?  If the keymap was for
individual jamo it would be manageable but I'm not sure how vim would
then combine the jamo.  If it is for all hangul characters it will be
very large -- unless I am missing something.
>
>   On the other hand, it should also be relatively easy to generate the
keymap automatically...
>
>   Benjamin
>
>  =20
Replying to all,

Hi, Benjamin,

Yes, It is a huge set of letters.
There are more than 10,000 letters that can possibly be made up, even
excluding obsoleted ones.

And there is a criticism I found about that method :

http://en.wikipedia.org/wiki/Revised_Romanization_of_Korean#Criticism

Thanks for your comment

Shawn.