[r6rs-discuss] Why Unicode matters
shiro at lava.net
Wed Feb 18 21:27:42 EST 2009
From: Grant Rettke <grettke at acm.org>
Subject: Re: [r6rs-discuss] Why Unicode matters
Date: Wed, 18 Feb 2009 18:33:32 -0600
> On Wed, Feb 18, 2009 at 5:48 PM, John Cowan <cowan at ccil.org> wrote:
> > As the R6RS process's chief Unicode hound, I'd like to say a word or
> > two about why I think Unicode matters. There are at least three kinds
> > of reasons.
> All good points.
> How much of a cultural shift is this for Scheme?
Having been using a large character set in Scheme for almost
two decades, I'd say there shouldn't be much cultural shift
as far as you won't confuse character sequence and byte sequence.
Distinction of textual ports and binary ports, and strings and
bytevectors, in R6RS, seems a right thing to me.
There are tons of hairy details, but Unicode people have
thought of them carefully.
Whether the language standard should mandate Unicode, or be
character-set independent, is a question of a diffrent layer.
Hypothetically, John's argument applies to any sufficient
I personally prefer character-set independent core, and
seeing Unicode as just an instance of it.
In practice, however, other large character sets (some of
which include more characters than Unicode, or better
compatibility to specific legacy encodings) can't compete
with Unicode in terms of availability and detailed technical
specifications other than codepoint definitions.
In order to *exchange* portable library code, Unicode might
be the only practical choice now.
More information about the r6rs-discuss