[Mailman-Users] maiman, problem with greek characters

[Stephen J. Turnbull]

Mark Sapiro writes:

 > Yes! base64 is a standard MIME encoding which the recipient's MUAs
 > should recognize. Why do the recipients have difficulty reading a
 > base64 encoded message?

Because they use non-conforming MUAs localized to the environment.
These MUAs are typically popular with users and admins because they
"just work" in the common cases encountered by naive users, and don't
require admins to use MIME-aware tools to examine messages.

We may as well just accept it; even in Japan, which is starting to get
its MUA act together after 30 years of self-imposed confusion (there
are three commonly used native encodings of the national standard
character set, all preferred to Unicode), just yesterday I received a
partially unreadable message generated by a portable telephone, using
"corporate standard" characters with code points in a reserved area of
the national standard set.

Unfortunately, I don't see an "everybody happy" policy for Mailman.

Sure, it's conceptually possible to add a configuration variable that
permits 8-bit transfer encodings, but it seems to me that would need
to be list- and Content-Type-specific (including charset!) which would
be complex.  Or we could assume that the originating MUA's transfer
encoding was well-chosen, but that would require tracking transfer
encodings for each subobject of a message, and deciding on
initialization etc.  Not worth it in a world that increasingly
supports all the MIME standard that Mailman uses, IMO.