On Wed March 9 2005 18:27, David MacQuigg wrote:
I was trying to move the discussion away from "what is spam" ( Irrelevant
question, in my opinion ) back to what to do with "Spam Bounces", assuming
the subjective judgement has already been made by the user.
If a user has already seen the message, there is no "bounce" (a bounce
would be a transport-related notification). A user *might* elect to
compose an entirely new message (e.g. to an abuse contact), but that
situation (new messages) is already covered in the architecture draft.
Abuse handlers *might* get together and define a structured format for
such messages based on the RFC 3462 multipart/report MIME media type
(as used by DSNs, MDNs, and MTSNs), but they haven't. And that would
be outside of the scope of the architecture document.