Paul Smith writes:
I think the "guaranteed" 'some response' of DSN is good.
Pay attention. The guarantee, like the analogous 8BITMIME guarantee, is
a myth: it was shredded by sendmail's longstanding disobedience. Even
when your ``some response'' actually appears, it is normally a useless
``I dunno'' response. Furthermore, the attempt to make this guarantee
forced a link-level design that, being rather difficult to implement,
interfered with the goal of producing ``Yes'' responses.
[ Notice-Requested-Upon-Delivery-To ]
What about BCCs?
The mechanism works exactly the same way.
If you're asking about the fact that truly blind deliveries require
separate envelopes: yes, that's true, and obviously the sender has to
split the return receipts accordingly.
---D. J. Bernstein, Associate Professor, Department of Mathematics,
Statistics, and Computer Science, University of Illinois at Chicago