On Feb 1, 2012, at 12:14 PM, Scott Kitterman wrote: > On Tuesday, January 31, 2012 04:23:33 PM Barry Leiba wrote: >> I am re-posting this without the extra recipients; please reply to >> THIS message, and NOT to that other one. We can discuss later who, >> exactly, should get the truncheon treatment here....... >> --- >> >> Here begins a last call for the MARF working group on the subject >> document, detailed below. Please make any comments you have on this >> version no later than 10 Feb 2012. That's a week and a half, which >> should be enough for this active and vocal group, wot? Please do not >> wait until the last minute, and especially do not wait until the >> document goes to the IESG. You will be beaten with a rubber >> truncheon. > > It looks good to me. Just a couple of comments: > > 3. The Mailbox Provider SHOULD send reports to relevant parties who > have requested to receive such reports. The reports MUST be > formatted per [RFC5965], and transmitted as an email message > ([RFC5322]), typically using SMTP ([RFC5321]). The process > whereby such parties may request the reports is discussed in > Section 3.5 of [RFC6449]. > > Although I understad the MUST here in context, it could be misread out of > context by people trying to insist on ARF. Could we have some kind of "To > implement the recommendations of this draft, the reports MUST ..." or similar? > > 12. Although [RFC6449] suggests that replying to feedback is not > useful, in the case of receipt of ARF reports where no feedback > arrangement has been established, a reply might be desirable to > indicate that the complaint will result in action, heading off > more severe filtering from the report generator. Thus, a report > generator sending unsolicited reports SHOULD ensure that a reply > to such a report can be received. Where an unsolicited report > results in the establishment of contact with a responsible and > responsive party, this can be saved for future complaint > handling and possible establishment of a formal (solicited) > feedback arrangement. See Section 3.5 of [RFC6449] for a > discussion of establishment of feedback arrangements. > > The SHOULD seems strong here. While I agree it's a nice idea, the odds of > this actually happening are vanishingly small in my opinion. Something > without a RFC 2119 keyword would be better here.
Unsolicited reports sent from an undeliverable address aren't terribly useful, as you can't ask the sender for additional context (data that's already there for FBLs). They're also more likely to be discarded or blocked. I think the SHOULD is a reasonable level of strength (though I wouldn't object to MUST). Cheers, Steve _______________________________________________ marf mailing list [email protected] https://www.ietf.org/mailman/listinfo/marf
