Re: [dmarc-ietf] hints for receivers

[Scott Kitterman <[e-mail]>]

On Saturday, April 20, 2013 10:49:00 AM Murray S. Kucherawy wrote:
> On Sat, Apr 20, 2013 at 10:21 AM, Scott Kitterman 
<[e-mail]>wrote:
> > Most of the current deployment is at large providers that can do
> > significant
> > data analysis to effectively identify lists.  As a small domain, there's
> > no way
> > I can do the same.
> > 
> > If there were a protocol method to share this information, either among
> > trusted receivers or via 'hints', it could be useful to smaller providers
> > that
> > don't have the scale to mine the data themselves.
> 
> Why would you, as a small-scale receiver, believe what an unfamiliar sender
> claims about whether its traffic transits lists, ultimately meaning "don't
> be so strict with this message"?  That seems a big loophole with little
> benefit to me, regardless of your size, given the nature of the problem
> space; malicious senders will always use it, and good senders that use it
> will be spoofed more easily.

I probably wouldn't.  My point wasn't to advance a particular solution, but to 
suggest that the experience base with how large providers with extensive data 
mining capability prefer to do things probably isn't universally applicable.

I solution involving receiver collaboration is more likely to be useful.

Scott K