You had claimed:
The point is in order to accomplish this, the CSV (and others) state
machine must be based on a delay design mechanism.
This is totally false, and is not supported at all by your 15K post,
which I have slogged through.
I am familiar with your claim that one MUST wait 'till RCPT TO: to do
any validation. It's a false claim.
Certainly doing so might be more efficient. That's not a false claim.
Re. Mixed policies discussion:
1)It's CSV, not CVS.
2)What part of
This is out of scope for CSV. IIRC, you've produced a doc that
addresses this scope (or "presumes to", to use your own words) for a
certain set of policies; it could could be revised to encompass CSV.
CSV does not purport to do the coupling you claim it does not do
properly.
do you not understand?
We are in what's called 'violent agreement'. We AGREE: CSV must be
considered in the reality of mixed policies. CSV *ITSELF* does not do
so, however.
We AGREE: doing the checks that are least costly first (which MAY mean
delaying validation) is a good idea!
In terms of the specific mixed policy issues you raised:
One only trusts Accredidation and Reputation services that one feels are
Reputable! If the USA's DMA runs such a service, I doubt many folks
will trust it!
Should ELVEY.COM get blacklisted? or should a report be sent?
IMO, a blacklist that blacklists based on a delivery attempt *without
having seen the content of the email, and therefore having no good way
to know whether it was UBE* should not be trusted; there are legitimate
reasons for CBV failures, IMO. So IMO, no. But again, let me
re-emphasise, this is not part of the CSV protocol, and does not belong
in it. It probably belongs in a BCP around usage. If a Accredidation
and Reputation services maintainer feels that there are no legitimate
reasons for CBV failures, ever, then Yes, the maintainer should ding
elvey.com in the scenario you describe. I believe there is consensus
among abuse desks that email abuse reports that don't include the header
and body of the email are unlikely to be given much, if any weight.