[hidetoshi.tajima@eng.sun.com: Re: IMdkit/onthespot method for xvim]

[Tung-Han Hsieh <thhsieh@twcpro.phys.ntu.edu.tw>]

----- Forwarded message from tajima <hidetoshi.tajima@eng.sun.com> -----

Delivered-To: thhsieh@tlug.sinica.edu.tw
Date: Tue, 09 Jan 2001 13:01:30 -0800 (PST)
From: tajima <hidetoshi.tajima@eng.sun.com>
Subject: Re: IMdkit/onthespot method for xvim
To: hidetoshi.tajima@eng.sun.com, thhsieh@tlug.sinica.edu.tw
Cc: dhson@zino.Physik.Uni-Mainz.DE, Brian.Yuan@eng.sun.com,
	Sriram.Swaminathan@eng.sun.com, katakai@japan.sun.com
Reply-To: tajima <hidetoshi.tajima@eng.sun.com>
X-Mailer: dtmail 1.3.0 @(#)CDE Version 1.4.2 SunOS 5.8 sun4u sparc

Hi,

First of all, I'm sorry for my late reply. I was taking a week longer
holiday last week after regular winter break, and have come to office
since yesterday.

>I am Tung-Han Hsieh, the current maintainer of xcin project
>(http://xcin.linux.org.tw). Thanks very much of your great
>work on IMdkit, which is the XIM server engine xcin using.
>Without this engine the xcin project will not be possible.

Thanks for kindly saying so, but I would have to say sorry by
imagining how many bugs in my old IMdkit you had to fix by yourself..

>The xcin program is designed specially for Chinese character
>input (for Big5, GB2312, Big5HKSCS). Because the Chinese
>input is a very complex issue, so we are interesting in
>writing a portable/stand-along XIM server to handle this
>taske as good as possible. Now xcin only use a small part
>of IMdkit: the Root and OverTheSpot input style, only one
>locale handling at run time. In the next step we will explore
>the multi-locale handling ability of IMdkit and add the
>OnTheSpot input style. Now xcin is successfully ported to
>GNU/Linux, FreeBSD, and HP-UX operating systems.

Perhaps, you may feel it easier with new IMdkit to support multi-locale
handling as well as on-the-spot input style.

In the new IMdkit and all the rests of stuffs in IIIMF/SDK we contributed,
UTF16 is used in text representation form, so multi-lingual scripts
will be easy, as long as the scripts can be mapped in UTF16 and we have
proper code-converter modules between the scripts and UTF16.

>During using IMdkit, we do encouter several problems, especially
>the stability problem. I know that there is a new version called
>IIIMF or something like that (very sorry, I am not familiar with
>your project, I will take time to survey your project contents),
>and I am very glad to try the new version. Now I am re-designing
>the character set sub-system of xcin to enhance its character
>set handling ability. After that I will study the new version
>of IMdkit.

Please feel free to cotact us whenever you get questions during the study.
It is our pleasure to get any interested people in our project and help
them use our stuffs.

>Because our goal is to develop a portable XIM server and hope
>that it could run on most of the UNIX operating system, so currently
>we strictly follow the X11R6 standard and will not go too far
>from this standard. Recently I note that XFree86 project have
>add many UTF-8 handling functions into their newest releases,
>but I think in the near future xcin will not use these calls,
>to ensure the portability. Similarly, when we start to try the
>new version of IMdkit, we will also take this into consideration.

Agreed. And just FYI about portablity, the IIIMF/SDK2 has been already ported
to Linux and Windows. Actually for Linux, "Atok12 for X" product from
JustSystem is based on our code, as you may know, Atok is one of the most
famous and widely-used Japanese language Input system on Windows and Linux.

>I am really very glad to have friends to discuss the XIM protocol
>and its related issues. In the past I have no one to discuss with,
>so the only thing I could do is to try and error and guess. So,
>thanks Son very much to send me the mail and thanks to the reply
>of Toshi. :-))

I'm also excited to get a chance to work with you and Son. Keep in touch!

Regards
Toshi


----- End forwarded message -----

To Unsubscribe: send mail to majordomo@linux.org.tw
with "unsubscribe xcin" in the body of the message