AN> In defense of "people", I reread marid-submitter and marid-core this
AN> morning and could not find the place where this was stated. However,
AN> this statement is significant. I'm not saying it is a good or bad
AN> idea, but it profoundly changes how SUBMITTER is used, the nexus of
AN> SUBMITTER and PRA, and even the semantics of the RR version identifier.
indeed.
submitter has been offered as an efficiency hack on the rfc2822-based
checking of sender-id. there has been no rfc2821.mailfrom in it.
That has been my reading as well. Indeed, the text of the submitter document is
quite clear that the identity involved is the PRA identity. It would also be
rather strange for the SUBMITTER SMTP extension to change the semantics of SMTP
when the extension _isn't_ used.
so meng's statement would seem to change the basic semantics of
sender-id, depending upon whether submitter is present.
The rationale document is also pretty clear that RFC 2821 MAIL FROM isn't
used, and why.
Ned