Let me start by saying that I like the CBV idea, but also see (at least one)
problem with it.
But reflecting on the problem, Lets call it Reverse CBV (RCBV) it is solvable.
without using the the MailFrom <> option, an option wich I dislike. (This is a
personal opinion)
RCBV could be better solved as the two of the two (or tree!) parties involved
knew that they where using CBV, wich means there
A: Should be an SMTP keyword to indicate that the server might use CBV
B: the Client that does the CBVtest needs to be CBV-aware.
C: The CBV-aware client MUST respond to this keyword. and so prevent looping
D: A B and C don't validate CBV, so there need to be more...
This solution requires an RFC and lets work (together??) on this basis.
Let persons who are pro-CBV make a Internet Draft, let the opponents shoot
holes in it, let the pro-CBV close the holes, ect, ect,
And at the end.
We (We are the internet) have good reasons why CBV is bad or we have an CBV-RFC
that can used within SMTP.
Ps this solution only solves the RCBV problem, does anybody see other problems
with CBV?
And the big question off course:
Does it stop spam?
I recognise the problem that this can double or even quadripple the number of
SMTP-connections if everybody starts using it, but even then I am wondering, if
CBV stops 80% UBE will it not decrease the total number of SMTP connections.