On 12/9/24 18:37, John R Levine wrote:
> Once again, we are not the Network Police.  If we make the fields 
> mandatory, that means that reports that are missing them but are otherwise 
> OK are ill formed and should presumably be discarded.  Why would that be a 
> good idea?

Implementing DMARCbis and new aggregate reporting should go hand in
hand. The new stuff in DMARCbis should be reflected in the new reports.
E.g. the "np" and "discovery_method" tags should be the reports, and
mandatory.

I don't think anyone, in response to the new RFCs, will update the
namespace declaration, and call it a day.

Presumably, no one will just implement new aggregate reporting on top of
existing RFC 7489 DMARC. And, presumably, no one will update to
DMARCbis, but keep the old aggregate reporting mechanism in place.

DMARCbis require new things; tree-walk, "np", "testing" etc.
We should expect that to be in the report, just as we expect the removed
"pct" not to be there.

Changes are mandatory, old reports are not compatible.

Why should we try as much as possible to be backward compatible in the
aggregate reporting XML format?


Daniel K.

PS. we need to add the "np" element to PolicyPublishedType.

_______________________________________________
dmarc mailing list -- dmarc@ietf.org
To unsubscribe send an email to dmarc-le...@ietf.org

Reply via email to