The non-US non-MIME senders DO assume that something outside the
US-ASCII character set will be understood by MIME capable readers of the
recipient.
That's how non-US implementors have implemented MIME.
This is undoubtedly true.
The only problem is, they don't all make the same assumption as to
what character set will be used.
I don't want MIME to require a certain interpretation for non-MIME
messages. But I don't see why the MIME spec can't point this out,
to encourage implementors to label charsets of outgoing mail.
Keith