Section 18.104.22.168 additionally states that "The reverse-path consists of
the sender mailbox", not a variation thereof. That wording apparently
bans using time-varying tags ...
OK, I have to admit I hadn't thought of or noticed this conflict. As
a practical matter there are already an abundance of schemes that
violate the letter, if not the intent, of this language: SRS and VERP
immediately come to mind. I therefore wonder if this isn't something
we ought to consider "relaxing" in 2821bis.
I don't get how it's a conflict. It doesn't say that mailboxes have
to be unique or have to be valid forever. Section 2.3.10 roughly says
that a mailbox is a thing that can receive mail, but it doesn't say
that it has to be able to receive mail a year from now, and it also
doesn't preclude having a thousand different addresses that happen to
deliver to the same place. I don't see how that contstrains the
address beyond requiring that it can accept a bounce.
In 2821bis the argument to MAIL FROM is a reverse-path which is "the
source mailbox ... which can be used to report errors". The
definition of mailbox is now section 2.3.11 but the wording is the
same. Again, I don't see how that contstrains the address beyond
requiring that it accept bounces.
If the bounce address weren't valid at all, or it deliberately stopped
working in 20 minutes, that could reasonably be said not to comply,
but the intent with the BATV timestamps is that the addresses will
remain valid at least for the week during which the message might be
retried and bounces sent.