Re: byteoffset() in lutf8lib.c from 5.3, work2

[Dirk Laurie <dirk.laurie@...>]

2014-05-14 2:13 GMT+02:00 Coroutines <coroutines@gmail.com>:
> On Tue, May 13, 2014 at 4:56 PM, Coda Highland <chighland@gmail.com> wrote:
>
>> I mildly disagree. While I agree that Unicode isn't perfect, I think
>> it HAS successfully addressed the goals it set out to accomplish. In
>> my opinion and experience, Unicode is better than any extant
>> alternative.
>
> It's nice that in ASCII you can do simple range checks on the
> codepoint to find out if something is a control character or numeric
> or alphabetic.  It's very easy to go from lowercase to uppercase and
> back.  We don't have that convenience in UTF8/unicode.  I'm not even
> sure if it could be done like ASCII, but I feel like things were
> poorly organized for programmers -- while it made sense for everyone
> else to group math symbols together or diacritic marks, etc.

It's not Unicode's fault. It's merely that real life is complicated.

When there are only 26 letters, and they are physically implemented
on your keyboard by hitting a button that originally was a lever that lifted
the entire platen, of course it is easy to know what is uppercase and what
is lowercase.

The moment you leave the comfort of the (maybe accented) Latin
alphabet, it becomes less obvious. You don't need to go far. As recently
as 250 years ago, English itself still imitated Greek by using different
symbols for lower-case `s`. Lower-to-upper is simple. Upper-to-lower
is ambiguous.

Unicode is nothing less than the 21st-century rebuilding of the tower
of Babel. In the long run, it might not be any more successful than
that tower in symbolically unifying our species.

But while it stands, it is a magnificent structure.