home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
kermit.columbia.edu
/
kermit.columbia.edu.tar
/
kermit.columbia.edu
/
newsgroups
/
misc.19990725-20000114
/
000445_news@columbia.edu _Wed Jan 12 10:24:38 2000.msg
< prev
next >
Wrap
Internet Message Format
|
2000-01-13
|
4KB
Return-Path: <news@columbia.edu>
Received: from newsmaster.cc.columbia.edu (newsmaster.cc.columbia.edu [128.59.59.30])
by watsun.cc.columbia.edu (8.8.5/8.8.5) with ESMTP id KAA07891
for <kermit.misc@watsun.cc.columbia.edu>; Wed, 12 Jan 2000 10:24:37 -0500 (EST)
Received: (from news@localhost)
by newsmaster.cc.columbia.edu (8.8.5/8.8.5) id KAA23520
for kermit.misc@watsun.cc.columbia.edu; Wed, 12 Jan 2000 10:21:25 -0500 (EST)
X-Authentication-Warning: newsmaster.cc.columbia.edu: news set sender to <news> using -f
From: fdc@watsun.cc.columbia.edu (Frank da Cruz)
Subject: Re: Arabic support in Kermit
Date: 12 Jan 2000 15:21:23 GMT
Organization: Columbia University
Message-ID: <85i65j$muu$1@newsmaster.cc.columbia.edu>
To: kermit.misc@columbia.edu
In article <85i4c3$les$1@newsmaster.cc.columbia.edu>,
Jeffrey Altman <jaltman@watsun.cc.columbia.edu> wrote:
: In article <85hoqo$ouc$1@techftp.technion.ac.il>,
: <JODY@lib.haifa.ac.il> wrote:
: :
: : I am trying to find out about support for Arabic in Kermit, in
: : particular the version for Windows 95/98.
:
: There is currently no support for Arabic in Kermit 95.
:
Except, of course, to the extent that Kermit 95 supports Unicode.
: : Currently, we display Arabic records from our bibliographic catalog on
: : terminals capable of displaying a special font. But we are gradually
: : phasing out our terminals, replacing them with PCs, and there the font
: : won't work. There is a terminal-emulation program, Reflection, that
: : supports Arabic well enough, but the cost for the number of stations
: : we'd need to install it on is prohibitive. So, the idea came up of
: : checking out Kermit.
: :
: : The website at Columbia implies that Arabic is not yet supported (it's
: : on the wish-list), but I wondered whether there have has been any work
: : done, by the Kermit staff or outside collaborators.
:
: In order to support Arabic two things must happen:
:
: . we must complete the GUI version (which has taken us much longer
: than we could ever have expected.)
:
: . the internal virtual display model must be modified to support
: arbitrary double width characters.
:
: . someone is going to have to pay for the development of the
: Arabic portion of the Unicode font to be used
:
Arabic support can come in different levels. Proper Arabic support that
would be acceptable to people who read and write Arabic does indeed require
all sorts of complex context-sensitive shaping and rendering (and therefore
a GUI screen), and possibly also implementation of bidirectional writing
algorithms by the terminal. As Jeff says, this is no trivial undertaking,
and one for which we, on our own, lack the needed expertise. And as Jeff
says, there is the matter of the font. There is still, as far as we know,
no monospaced Unicode font suitable for terminal emulation that includes the
glyphs needed for Arabic -- let alone Armenian, Hebrew, Georgian, or for
that matter full VT100 graphics. (Let alone Chinese, Japanese, and Korean!)
At the opposite extreme is "character-cell Arabic", similar to how Hebrew,
or for that matter, Greek, Russian, or English are handled by a terminal.
In this case the host character set would be something like ISO 8859-6 and
the local one would be something like PC Code Page 864 or 1008, and writing
direction is handled by the host application. This we could do in short
order, but of course the font would still be an issue. The only reason we
haven't done this so far is that nobody has ever indicated they would be
satisfied with it (and many have indicated they wouldn't).
Somewhere in the middle is a duospaced character-cell approach, in which
each Arabic character occupies either one or two cells on the terminal
screen, and there is some agreement between terminal and host about how
how this is done. Something like this will probably also be needed for
Chinese, Japanese, and Korean.
As you can see, it's a complicated question -- or at least a complicated
answer!
- Frank