We have three possible protocol modes:

1) Evaluator using RFC 7489 with any domain owner policy
2) Evaluator using DMARCbis with domain owner publishing policies based on
RFC 7489 only
3) Evaluator using DMARCbis with domain owner publishing policies based on
DMARCbis

We know that the first mode is subject to organization consolidation and
fragmentation based on PSL errors

We know that the second mode is also subject to organization consolidation
and fragmentation errors, based on missing policy tags.

We have an opportunity to eliminate organization consolidation and
fragmentation errors completely, for communication using the third mode, by
asking the domain to add a DMARCbis tag to each DMARC policy.   When tags
are uniformly applied and internally consistent, the evaluator has the
right answer and knows that he has the right answer.   This is a big win.

Yet this does not seem to be a goal of our present text.  Indeed, the
discussion has resisted moving in this direction.   Why would we not want
to eliminate risk wherever possible?

Doug
_______________________________________________
dmarc mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dmarc

Reply via email to