Re: [dmarc-ietf] what to document about the tree walk

[John R Levine <[e-mail]>]

On Thu, 14 Jul 2022, Scott Kitterman wrote:
> In my view, standardizing two ways to do policy discovery and alignment would be a substantial danger to interoperability and we'd be stuck with it approximately forever.

I agree, it's a self-evidently terrible idea.  "Temporary" transition
periods inevitably turn out to be permanent, or so close to permanent that
we'll all be dead before it's over.

What I expect to happen is that we publish 7489bis with the tree walk as 
the method to find org domains*.

There are a handful of widely used libraries that implmenent DMARC, most 
of which have developers who read this list or are otherwise people we 
know, so we can encourage them to update their software.  Large providers 
like Google and Microsoft have their own implementations but we know them 
too.

So over perhaps a year the places that upgrade their software will get new 
libraries and start to use the tree walk.  There will always be a long 
tail of sites that never update their software, but that's life. 
Fortunately, for the majority of normal mail the old and new methods 
get the same result, so it's unlikely the long tail will run into problems 
any worse than they already have with the obsolete software they use, e.g. 
TLS 1.1 making STARTTLS fail.

R's,
John

* and occasionally PSDs.