No, not really. In non-EAI mail, the headers are all ASCII, even though some of that ASCII may be MIME encodings of UTF-8. On systems that support 8BITMIME (all of them, these days) the body can contain literal unencoded UTF-8 in the body parts, although the MIME headers between the body parts are still ASCII.
Ok, I was sloppy in my wording.
If the email client do understand MIME and handle that correctly, one can have non-ascii in headers and not only the body already since...hmmm...1991. Thats only 28 years. :-)
In EAI mail, the headers and the body part MIME headers can contain UTF-8 just abuout everywhere other than the field names.
Yes, but for me, if you can encode things and then decode in a non-destructive way, I call that support.
Theyr'e different levels of abstraction. SMTP transport does no encoding
or decoding, so at that level everything has to be ASCII, or with
8BITMIME, strings of bytes under 1000 long with ASCII CR/LF at the end of
each. Go up a level to software that interprets the message and sure, you
can have all sorts of MIME encodings.
That said, yes, I know, have talked with many, and understand people
that want to have non-ascii in email addresses. Specifically for use in
a local context.
Right. There was an interesting talk at the recent ICANN tech day by two
women from Thailand who said they have a lot of mail users who are
literate in Thai but not in any Latin alphabet language.
Regards,
John Levine, johnl@xxxxxxxxx, Taughannock Networks, Trumansburg NY
Please consider the environment before reading this e-mail. https://jl.ly
