Steve Atkins wrote:

> A more interesting question is how domain based authentication helps
> domain reputation based systems reduce false positives in spam
> filters, or how domain based feedback loops help ISPs and mailers
> avoid sending unwanted email. DKIM itself doesn't do either of those,
> it's just a platform they're based on.
>
> I don't think we're at a point, yet, where the answer to that will be
> particularly enlightening. It may be time to start tracking the data
> if you're not already, though.

Similarly, I don't think we're at a point where any discussion of 
standardizing domain reputation systems -- any more than we're at a point of 
standardizing IP reputation systems.  There's a lot more experimentation 
that needs to happen (and is happening) before anything could be considered 
standard enough for an IETF document.  I'd expect to see (and may offer) 
competing "considerations for..." drafts before there's anything close to a BCP.

So, where does that leave the DKIM WG?  I think it leaves us exactly where 
we were before: reputation assessment is, and should be, out of scope.

-- 
J.D. Falk
Return Path Inc
http://www.returnpath.net/
_______________________________________________
NOTE WELL: This list operates according to 
http://mipassoc.org/dkim/ietf-list-rules.html

Reply via email to