From: Richard Wordingham (richard.wordingham@ntlworld.com)
Date: Mon Sep 18 2006 - 03:01:55 CDT
----- Original Message -----
From: "Don Osborn" <dzo@bisharat.net>
To: <unicode@unicode.org>
Sent: Monday, September 18, 2006 4:30 AM
Subject: RE: Unicode & space in programming & l10n
> Mark, Doug, Steve, Thanks for your replies which are most interesting and
> helpful. I should clarify, having quoted from the article in a way that
> might give a misimpression about the author's intent, that I think he was
> not arguing against Unicode but rather attempting to give an appraisal of
> how well the current state of the technology is facilitating multilingual
> ICT, problems encountered, etc., within a broader discussion of factors
> affecting the potential for linguistic diversity on the net.
>
> My interest in getting some reaction from experts like you on this list
> was
> to better understand his appraisal of Unicode in this particular aspect,
> and
> how much credence to give to the issue of the space requirements of
> Unicode
> in something I'm writing. The context of my work is Africa, where a lot of
> computer systems are older and bandwidth is often very narrow, and any
> discussion of file sizes etc. merits looking at (though the main issues
> for
> Unicode in Africa still probably are (1) systems too old for it and (2)
> computer and software people there not knowing much about it).
One point that was omitted was the difference in size between NFC and NFD.
But, if you'll excuse my ignorance, what are the choices avaiable? As far
as I'm aware it's Latin-1 v. the fully French-supporting Latin-x v. UTF-8 v.
UTF-16. I would be expect UTF-8 to be superior to UTF-16 for Latin and
Arabic based writing systems, as these scripts non-ASCII characters take
only 2 bytes.
Are there even useful ISO-2022 schemes available?
>
> But Paolillo's remarks about programmers and Unicode seemed to go well
> beyond the typical discussion of such limitations, which made the issue
> all
> the more interesting to bring up here. Especially since the document it
> appeared in is intended to inform discussions about the multilingual
> internet.
>
> Thanks again, and I look forward to learning more as the discussion
> evolves.
>
> Don
>
>
>
>
>
This archive was generated by hypermail 2.1.5 : Mon Sep 18 2006 - 03:08:21 CDT