Re: [marf] Change request for AS, was Working Group Last Call on draft-ietf-marf-as-05

[Scott Kitterman <[e-mail]>]

On Sunday, February 05, 2012 02:31:14 PM Barry Leiba wrote:
> > Why the AS draft? There are tons of non-spam auth failures. These are
> often the most interesting ones.
> 
> > If this belongs in the AS draft then it's probably misnamed.
> 
> I'm confused by both of the last two sentences... I don't see the
> antecedents to "these", "this", and "it's".

These are auth failure arfs.  This is the text from the DKIM and SPF drafts 
being proposed be consolidated into the AS draft and it's is the AS draft.

> As to "Why the AS [document]?", It's because that document is an
> "applicability statement" for the work being done with MARF.  An AS is a
> standards-track document that describes how to use the underlying protocols
> (as opposed to a TS (technical specification), which defines the
> protocol(s)).  Part of using the protocols involves dealing with feedback
> loops, reporting, failures, ans so on.
> 
> We decided to split certain details off into separate documents, for
> various reasons, but it all fits into the "applicability statement" box,
> and it makes sense to put the common bits into the AS document.

For auth failure reports we (tried to anyway) put the common bits needed to 
extend arf to send messages about authentication failures in draft-ietf-marf-
authfailure-report.  If there is further factoring to be done from the DKIM 
and SPF drafts, that's where it seems to be sensible to put it (I know that's 
procedurally complex at this point).

I don't think what's being discussed fits a draft about anti-spam.  Anti-spam 
and auth failure are different concepts.  In many cases the most interesting 
(to the sender) cases are the ones where legitimate messages failed 
authentication.  There are also interesting authentication failure cases where 
the message is spam, but it's certainly not limited to that.

It would seem odd to me to have to go read an RFC about anti-spam in order to 
implement authentication failure reporting, so if what's in the anti-spam 
draft is really necessary to authentication failure reporting, then I suspect 
it is misnamed.

That's a longer (and hopefully clearer) version of what I was trying to 
communicate yesterday.

Scott K