On 22/Dec/11 06:07, John Levine wrote:
> 
> There's two fairly different contexts in which people use ARF.  The
> original one, and still by far the most common, is when two mail
> systems make a private agreement to exchange abuse reports, usually
> reports due to recipients manually reporting messages as spam.

Another original use case is the unilateral FBL from a mail system to
a bulk sender.

> The other one is sending reports between parties that don't know each
> other, with the recipient address typically being abuse@domain, or
> looked up via RIR WHOIS or the like.

One of the like methods should be described in the reporting-discovery
draft.  RIRs' refurbishing of their databases and access methods seems
to be arriving earlier than that, at least for ARIN.

However, whois databases need the cooperation of network providers to
get updated.  PTR records suffer of the same complication, worsened by
the need to maintain consistency between DNS and rDNS.  The other
method --the RFC 2142 "abuse" role-- being what it is, if
report-discovery will make it it seems it will be the one more
universally available to mailbox providers.

> The reports may be manual, or automated due to hitting spam traps,
> or scored high by spam filters, or anything else.

This classification is also worth being made, IMHO.  Not sure where
and why, though.

> I'd suggest reorganizing the draft a little to clarify that prior
> agreement vs. no prior agreement are two different applications, both
> in current use.  If people want, I can suggest specific changes.

I wrote some changes, but didn't post the resulting draft.  For the
time being I copied it in http://www.tana.it/file/
_______________________________________________
marf mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/marf

Reply via email to