OK, here's another tweak, in response to the latest round of comments:
NOTE: In certain transport enclaves, RFC 822 restrictions such as
the one that limits bodies to printable ASCII characters may not be
in force. (That is, the transport domains may resemble standard
Internet mail transport as specified in RFC821 and assumed by
RFC822, but without certain restrictions.) The relaxation of these
restrictions should be construed as locally extending the definition
of bodies, for example to include octets outside of the ASCII range,
as long as these extensions are supported by the transport and
adequately documented in the Content-Transfer-Encoding header field.
However, in no event are headers (either message headers or
body-part headers) allowed to contain anything other than ASCII
characters.
Some specific responses:
Excerpts from TODO: 28-Apr-93 Re: Comment on the draft MI.. Olle
Jarnefors(_at_)othello(_dot_)a (2454)
The proposed NOTE uses very cautious language: "RFC 822
restrictions SUCH AS", "extending the definition of body-part,
FOR EXAMPLE to include octets outside of the ASCII range".
Is this really necessary? _RFC 822_ seems to put no other
restriction on the message body than that octets must not
be > 127. I can think of three other restrictions, none of
which is part of RFC 822:
You listed them all, but I see no harm in the cautious language; it
makes it easier to infer what to do in the event of other
transport-level extensions.
-- In Appendix D, put the definitions of "qp-iffy" and
"discard-text" in proper alphabetical order.
Done.
-- In Appendix D, either insert a reference to the NOTE on
relaxing RFC 822 restrictions on bodies or duplicate it.
Done -- actually, I attached the NOTE to the BNF for "body-part", so it
shows up in both places. -- NB