Re: [marf] New Version Notification for draft-li-marf-not-spam-feedback-00.txt

[Alessandro Vesely <[e-mail]>]

On 13/May/11 20:05, Steve Atkins wrote:
> On May 13, 2011, at 10:57 AM, Barry Leiba wrote:
> 
>> Kepeng Li and I have just submitted a draft for consideration by the
>> MARF working group:
>> 
>>> Filename:        draft-li-marf-not-spam-feedback
>>> Revision:        00
>>> Title:           Email Feedback Report Type Value : not-spam
>>> Creation date:   2011-05-13
>>> WG ID:           Individual Submission
>>> Number of pages: 7
>>> 
>>> Abstract:
>>>   This document defines a new Abuse Reporting Format (ARF) feedback
>>>   report type value: "not-spam".  It can be used to indicate that a
>>>   message that was tagged or categorized as spam (such as by an ISP) in
>>>   fact is not spam.
>> 
>> http://tools.ietf.org/html/draft-li-marf-not-spam-feedback
>> 
>> As noted in the draft the requirement for this comes from the OMA
>> SpamRep folks, and also seems like a reasonable need in general.  We'd
>> like the MARF working group to review this, and to consider adopting
>> it.

+1 for adoption.

>> It seems like an easy ride, I think.
> 
> Probably. Seems harmless.

A possible harm is inappropriate usage.  IRC, "non-spam" used to be in
ARF and was removed for this reason.  I suggest to change the label to
something more specific, such as "ham" (to highlight Bayesian training
purposes), "reconsideration-request", or such.

> Is this intended for communication between MUA and message store (doing
> much the same thing as an IMAP move from spamfolder to inbox folder)
> or is it something that's intended for broader use across administrative
> boundaries (like, say, ISP pushback to Postini)?

I hope so, because trusted authors deserve to come to know if their
messages are misclassified, and have a chance to appeal.  In this
respect, the example in section 3 is described deceptively.  Does it
mean "Joe has _authored_ a message..."?

The human readable part of the example does not say /why/ it is not
spam.  (Shouldn't it mention RFC 5965 rather than mipassoc's URL?)