Bruce Lilly <blilly(_at_)erols(_dot_)com> writes:
I further believe that in the case of a response to a message containing
at least one Resent-Message-ID field, the msg-id placed in the response
In-Reply-To and/or References fields should be that of the most recent
Resent-Message-ID rather than that in the Message-ID field (if present)
or any other Resent-Message-ID field which might be present.
I don't believe those semantics would be nearly as useful as following the
standard References semantics. Consider the case where someone resends a
mailing list message to someone who didn't receive the original, and the
recipient then responds to the message, cc'ing the mailing list. If the
References header is constructed in the traditional fashion, the reply of
the person to whom the message was resent will thread correctly for all of
the readers of the mailing list. If the References header is constructed
from the Resent-Message-ID field, it will not.
I think the best way of looking at the process of resending a message from
the perspective of thread integrity is that it is the after-the-fact
addition of recipients to an equivalent of an MUA Bcc field. In effect,
the distribution of the original message is being expanded retroactively.
It's useful to include those trace headers so that a recipient of a resent
message can figure out what happened, but I believe if the conversation
then continues, much more is gained by unifying the replies to the
original and the replies to the resent message into a single thread than
distinguishing replies to one from the other.
--
Russ Allbery (rra(_at_)stanford(_dot_)edu)
<http://www.eyrie.org/~eagle/>