[ietf-dkim] protecting domains that don't exist

[Arvel Hathcock]

> It's called feature creep. 

Hooey.

Really.

> It is also entirely outside the scope of this working group.

Substantiate this incredible claim or please stop making it.  It's just 
plain false.

> Really. It has nothing to do with DKIM and nothing to do with Author 
> Domain *signing*.

We're talking about an algorithm for what to do in the absence of DKIM 
and signing.  The NXDOMAIN check is a piece of that algorithm that makes 
it work.  Without it, our purpose is trivially defeated.  Why can't you 
see that?

> Really.  A working group's protocol specification effort is supposed to 
> be guided by more than a current feeling that it should tell folks what 
> is good for them.

That's not fair.  Really.  Really, really.

I and others have offered concrete specifics on why we know this is a 
critical piece of the algorithm.  That's not guidance by "current 
feeling" or a desire to "tell folks what is good for them."

Arvel