On Wed 04/Dec/2019 08:13:48 +0100 Scott Kitterman wrote: > I'd prefer to see the new dns ptype separated from the dnswl discussion. I > can see broad utility in the dns ptype (for example, if you want to indicate > that a domain is testing DKIM, I think we need dns because that's where you > find the information - it's not an attribute of the signature).
Ptypes are already shared. Define once, use freely. (Except that each method has to define which properties it uses, and hence which ptypes.) IMHO, a simplification step, whenever someone is going to address rfc8601bis, would be to factor the results as well. For example, every method carefully specifies pass, fail, temperror and permerror to have the same, agreed semantics. The IANA page nicely allows to sort the table by code, a bemusing experience. Best Ale -- _______________________________________________ dmarc mailing list dmarc@ietf.org https://www.ietf.org/mailman/listinfo/dmarc
