As far as I'm aware the strongest objection to a well-designed sender
rewriting scheme is that it imposes costs on operators of forwarding
systems (while providing some small benefits to their users and to
senders).
The strongest objection of which I'm aware is that nobody does it, and
few people are likely to do it, so any design that depends on sender
rewriting isn't going to work.
John Levine:
When it's a bounce of an actual no kidding message and I can look at
what's quoted in the bounce to see what I wrote, extremely useful.
I find that my messages usually bounce for reasons not depending on body
content. Often, the last thing I'd want is another copy of the message - I
have one in 'Sent' already. The address I sent it to and the Subject would
normally be enough for me. Do DSNs always include the body? Must they?
Read the RFCs and figure it out for yourself. But they do have enough
to recognize the bouncing message.
And anyway - who's saying that bounces be done away with?
Practically speaking, you are, saying the bounce should go back to the
forwarder rather than the original sender. In my experience, there's
rarely anyone paying attention at the forwarding host. On my system,
for example, I let people set up their own forwards, and I chronically
have problems where they set up a forward, but out of caution or
intattention still have mail delivered locally, and then forget about
the local mailbox which fills up and overflows. If you don't send the
bounces back to the original sender, nobody's going to see them.
If people want to do sender rewriting, essentially treating forwards
as a one-member mailing list with a robot to catch the bounces and
turn off the forward (and perhaps the orignal address) if there are a
lot of bounces, that's not evil or anything, but it's never been the
standard way to forward mail and I don't see any likelihood that it
ever will be.
R's,
John
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg