--- Arlie wrote:

> As far as I know, the only systems in common use which understand > >
> unicode are recent IBM PCs.

I think it depends on the software, though, rather than the machine.
Right now I am using a Linux box and am doing the mail with Mozilla.
I see now that Mozilla has a menu for different character code pages.

But I must confess that my first attempts were unsucessful.
I think it also depends a bit on how yahoo handles things.
i.e. do they use MIME codes, or do they just flush through
8-bit codes straight. (SMTP standard is still 7 bits I believe).

Under the "View" menu in Mozilla, there is also a submenu
titled "Character Coding". If I follow it, I am led to a sub-menu
with lots of options for different character codes.

Some other day I hope to have more time to explore it.
For now, I'd like to remark that unicode characters seem to
have numeric codes that consist of 4 hex-numbers.
The old iso-codes consist of only two hex-numbers.

As a consequence, I found out that characters whose unicode
begin with two zeros seem to come through fine. It gets much
worse when the first two hex-digits are non-zero.

Examples:

þð
ÞÐ
áeíóú
ÁÉÍÓÚ

(all these should come through fine, I believe
and are quite easy to write with the Linux keyboard.)

But if I, for example, try with a "u" with a bar over it
(unicode 363, I recall = hex 016B), you get all kinds of
unpredictable results.

Best regards
Xigung.