Re: [Gnu-arch-users] [OT] Unicode vs. legacy character sets

[Stephen J. Turnbull]

(I'm way behind on my reading; if this is missing important context
from other posts, my apologies---the post I'm replying to accidently
got copied to an urgent folder, so I read it even though I knew it
wasn't dayjob-related.  ;-)

First things first.

    Tom> Mostly, the plan is to wind up with attributed text as one of
    Tom> the fundamental string types.

I doubt you'll like the implementation of
http://www.m17n.org/m17n-lib/, but that's exactly what they do.
Might be worth mimicking their API where it makes sense.  (Or not,
depending on your design goals.)

    Tom> Indeed -- make a hostile fork but get enough people behind
    Tom> and then, boom, that's the new standard.  I don't think it
    Tom> would work but if the dissenters have a compelling point,
    Tom> than it will.

Here there are no "hostile" forks; forking Unicode is more aptly
characterized as "terrorism,"[1] undermining the whole social
framework to achieve tactical goals.  (IMHO YMMV, but I really feel
that strongly about this particular standard, even though in many ways
I think the standard itself sucks in details.)

    Tom> The primary point being that it makes little difference to
    Tom> anything that's in or planned for hackerlab and Pika anytime
    Tom> soon -- swap out some consortium data files for other data
    Tom> files and there you go.

But think about it---you'll need a standard to tell you which files
you're using.  ISO 2022 (aka "modal encodings"), here we come.

    Tom> It was just a thought.  The "fork, don't complain" sentiment.

Normally I'm singing in the choir, Brother Tom.  I think this case is
an exception.



Footnotes: 
[1]  Consider that the only major fork of Unicode currently in
existence is GB 18030, where the Chinese decided to create a version
of Unicode which grandfathers GB2312 in the same way that ISO 8859/1
is grandfathered in standard Unicode, and then mandated support for
that standard if you want to sell in China.

-- 
Institute of Policy and Planning Sciences     http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba                    Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
               Ask not how you can "do" free software business;
              ask what your business can "do for" free software.