On 11/11/2020 8:58 AM, Murray S. Kucherawy wrote:
So the new DMARC document would say: "Determine the Organizational
Domain. The legacy way to do this can be found in [other-RFC], but
other better methods are anticipated." Then when something better
than PSL (maybe a tree walk, maybe something else) comes along, the
IETF publishes that specification and DMARC implementations and
operators switch to that method. The base document doesn't need to
change at all.
On 11/12/2020 8:30 AM, John Levine wrote:
I asked in DNSOP about tree walks and my take on the response is
that
...
_dmarc.sun.com. CNAME _dmarc.oracle.com.
Tree-walking is unacceptable. Tree-walking is now acceptable.
Use CNAMES. Don't use CNAMES.
The issue is not whether a bit of mechanism might work, but that this
realm of activity is both complicated and has a history of controversy,
as well as a history in the PSL case of not working all that well.
All of which should strongly encourage isolating this topic from DMARC
as much as possible. Make sure that changes to this do not require
changing the core DMARC spec.
Murray's proffered text does this nicely.
d/
--
Dave Crocker
Brandenburg InternetWorking
bbiw.net
_______________________________________________
dmarc mailing list
dmarc@ietf.org
https://www.ietf.org/mailman/listinfo/dmarc
