On 6/15/2014 4:08 AM, Murray S. Kucherawy wrote: > What about a new canonicalization, which is largely the same as the > existing ones but carries with it the additional semantic that "This can > only pass when accompanied by a Mediator signature"? > > Current verifiers don't know what this is and thus wouldn't know what to > do with it, so unless they do something abysmally stupid like "I don't > know what this canonicalization is, so let's just call it a 'pass' to be > on the safe side", this might be a path forward without a version bump.
The suggestion I made in an offlist discussion was a new header canonicalization, but it hadn't occurred to me to include the contingency of a mediator signature. It's an interesting idea, but does seem like some sort of layer violation. (i wanted some phrase other than 'kludgy' just so i didn't echo john...) Surely there are enough details we could vary in a header canonicalization that are legitimate so that we don't need to toss in odd 'policy' characteristics to the of the algorithm? d/ -- Dave Crocker Brandenburg InternetWorking bbiw.net _______________________________________________ dmarc mailing list dmarc@ietf.org https://www.ietf.org/mailman/listinfo/dmarc
