[feedback-report] Usage Scenarios

[Yakov Shafranovich]

Prompted by a suggestion from Dave Crocker, I wanted to see if we can
come up with a list of possible usage scenarios for this format. Here
are some of my thoughts:

1. Feedback loops - ISPs providing reports back to email senders about
spam reports from their users (like SCOMP at AOL). This would imply that
the ISPs and email senders establish a relationship in some way (which
is out of scope for this spec).

2. ISP-to-ISP reporting - ISPs exchanging reports among themselves -
i.e. spam originating from ISP A's network going to ISP B, and ISP B
reporting it back to ISP A. This also implies some form of an
established relationship among ISPs.

3. Users reporting spam to their ISP - this would be like the "Report
Spam" function in AOL's and Yahoo's interfaces but on the MUA. In this
case, an ISP user (lets say Earthlink) would be able to report a spam
message arriving in his MUA (Outlook, pine, Thunderbird, etc.) to his
own ISP which would then aggregate and process them further before
taking some sort of action on them. This implies that the ISP has a way
of checking whether this user is actually theirs and would help in cases
where ISPs provide a "report spam" button in the webmail interface but
nothing for POP or IMAP users.

4. Users reporting spam to aggregate services - this would be intended
for services like SpamCop where users can report their spam via MUAs and
the service will process them further before sending them off. This
implies that the service has a way of checking if the user is actually
part of that service.

5. Users to ISP reporting - this would allow regular users to report
spam directly to ISPs. However, given that most ISPs and users will
probably not be able to do this properly, this scenario might not work
very well (but scenario #3 instead might).

Yakov