In countries needing Latin-1 characters there is usually no problem.
If they don't show up correctly on the other side of the world who
cares -- they wouldn't understand it anyway. Mosaic supports it ok,
I guess Lynx works just fine if the terminal understands them, so
my advise would be to use Latin-1. I mean those ampersand things
make the source text look like hell, and currently it is still the
poor human-beings writing it with Emacs, not some HiTech HTML editors.
> - Using the SGML entities ensures that the file can be e-mailed (see
> what became of your öaut; above...)
The guy designing these SGML entities was clearly someone whose mother
tongue didn't contain any of them. Imagine what a pain it would be to
always type '&blaa;' instead of 'a' and '&bloo;' instead of 'o'. This
text would look mighty horrible, never mind if I can email it. (BTW,
ö doesn't map to | if you clear the high-order bit, it's just
something he typed in because scandinavian terminals can be configured
to show the (7bit) pipe char as ö).
> - Browsers that cannot display the characters, can -- in principle --
> approximate them.
Well, none of them do. And this is the 90's, machines should have
those characters and not just approximate them.
-- Cheers, Ari "why-does-this-gimme-an-open-brace-when-I-wanna-have-an-ä-my-emacs-setup-sucks-or-is-it-this-damn-window-system" Luotonen --