Re: A draft for a multibyte and multi-codepoint C string interface

[Mouse <mouse%Rodents-Montreal.ORG@localhost>]

> I believe that Arabic text contains the digits in the same order as
> English.

That's ambiguous: the same order meaning "most significant on the left"
or the same order meaning "most significant encountered first in usual
reading order"?

> Even in europe, a French person will write completely different
> looking 1 and 7 to an English person.

Not all taht different.  Not nearly as different as, for example,
Arabic eight and `Western' 8.  See Unicode codepoints 0660-0669 and
06f0-06f9 (the former, according to what I have, for use "with Arabic
proper", the latter "for languages of Iran, Pakistan, and India").

> English has two common glyphs for 3, 4 and 9 (and to a slightly
> lesser extent 6),

Not very different in any of those cases, certainly not as different
as, for example, the two basically different glyphs for `g'.  (One has
a single loop sitting on the baseline, with a descending stem on the
right that curls left under the loop; and the other has two closed
loops, one on the baseline and one below, linked with a short vertical
piece, often curved, at the left.)

At least, I'm assuming your differences are

3 - top half a partial circle versus a flat top and angled line
4 - top open or closed
9 - straight vertical stem or stem curled under the loop
1 - how far the slanting line fromk the top of the vertical extends
7 - crossed or uncrossed

> Maybe that is just part of a font, but most of the differences
> between fonts are much smaller (apart from serifs).

Only if you pick fonts where they're similar.  Compare Magnificat with
Block Up with Audio 2012 with Rollerball 1975 and tell me again how
differences between fonts are smaller than variations such as the
different varieties of 3.

/~\ The ASCII                             Mouse
\ / Ribbon Campaign
 X  Against HTML                mouse%rodents-montreal.org@localhost
/ \ Email!           7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B