Re: SPF isn't going to change, was Deprecating SPF

On Fri, Aug 23, 2013 at 06:40:05PM -0700, David Conrad wrote:

> As far as I can tell, the issue isn't whether the SPF RR is being
> used (it appears to be, albeit at a tiny percentage relative to TXT
> currently)

Right, despite the fact that 4408 actually suggested that it'd be good
to use SPF.

> , rather it's whether or not a handful of major mail
> service providers will use SPF RRs in the future if the IETF were to
> document a fix to the broken transition strategy in 4408. 

How would the IETF do that?  The strategy in 4408 was roughly, "Try
both," though admittedly in a flawed way.  But this is exactly the nut
of the problem.  There is no way to create an incentive, given the
actual world we live in, for anyone either to publish or to check 99
as the primary mechanism.  As long as 16 is what everyone is already
doing, network effects say that 16 wins.  There is no advantage to
anyone in moving to 99: neither on the check side nor on the publish
side do you win.  So nobody will.

We either need something new that will make SPF worth doing -- call it
SPF v3 that expresses new policy that's even better (and I can think
of something!  Think of EAI-using domains that need better
granularity!) -- or else we need to embrace the lowest common
denominator.  Nobody in any of this has offered the slightest
incentive to anyone, AFAICT, except, "DNS is better this way."  Who
cares?  I do, you do, but we don't organize the universe.

The above is my claim: nobody cares about this hygiene.  If I'm wrong,
then I want a rigourous study to show it, because none of the
empirical evidence suggests so far that accepting and documenting the
facts as they are today is different than the effects of saying, "You
oughta like TYPE99 more."  We don't have legislative power.

A

-- 
Andrew Sullivan
[email protected]
_______________________________________________
dnsext mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dnsext