[net.mail.headers] envelopes versus messages

wcwells%ucbopal.CC@Berkeley.ARPA (William C. Wells) (02/25/84)

In reply to:

	Date:  Wed, 15 Feb 84 20:47 EST
	From: "Benson I. Margulies" <Margulies@CISL-SERVICE-MULTICS.ARPA>
	Subject:  envelopes versus messages
	To: header-people@MIT-MC.ARPA
	Message-Id:  <840216014745.993851@CISL-SERVICE-MULTICS.ARPA>

	It seems to me that some of the cross purposes of this
	discussion relate to the role of envelopes.  I always have
	assumed that the content of the envelope is available to the
	mail agent (mail reading program, if you like) of the
	recipient.  Am I out-to-lunch on that?

There is a problem with the definition of envelope. We each seem to have
our own definition. "envelope" information, that is information used or
added by the mail transport agents, should be passed on to the addressees
of a message, but may not be. The mail transport agents job ends when
the message has been spooled to (placed in) the addressees mail box.
Certain information available to the mail transport agent may not be
added to the heading of the message prior to being delivered to the
mail box.  Trace information and other information added by the mail
transport agents should be available to the receiving user agent,
though not necessarily displayed by default.  In the case of SMTP, much
of the SMTP envelope information may also need to be passed on to
a non-SMTP mail transport programs in the heading of the message.

The information also needs to flow in the opposite direction, from the
user agent (the mail formatting/generation program) to the mail
transport agent, and in some cases even down to lower levels.  For
example, the "precedence" and "classification" of a government message
is defined by the user agent, but does not look like the RFC 822 or
SMTP provides for precedence or classification information to be passed
from the user agent down to the IP Header "Type of Service" Precedence
bit and the Security field of the IP Header "Security" Option (see RFC
791 section 3.1).

Bill Wells
wcwells@Berkeley.ARPA
ucbvax!wcwells