On Oct 6, 2014, at 4:24 PM, Murray S. Kucherawy <superu...@gmail.com> wrote: > On Fri, Oct 3, 2014 at 9:05 AM, Tim Draegen <t...@eudaemon.net> wrote: > On Sep 20, 2014, at 7:41 AM, Alessandro Vesely <ves...@tana.it> wrote: > > IMHO, we should specify a credible MLM model, even if that can be > > slightly off topic for the WG, in order to maximize its probability of > > being adopted. The rest of this message has some notes to this end. > > Can I get some clarification on the intent here? As worded, this paragraph > suggests that we are looking to produce a model for MLMs to follow in a > DMARC-aware world. I was under the impression that (a) this is a > non-starter, and (b) we're not chartered to do so.
Sorry to inject confusion and not immediately follow up. The intention is to have something in place -- the MLM model -- that can be used to quickly identify issues that are related to DMARC interoperability with any given piece of MLM software. I read Alessandro's model as a way to generically describe MLMs, which would make comparing and contrasting of MLMs a lot easier. IOW, fleshing out a matrix of interoperability issues with respect to MLMs is made easier (possible?) if we have a generic way to describe MLM behavior. This is not meant to be a robust exercise in crafting the ideal MLM. If something like this is NOT in place, my concern is that the WG will only look at the big MLM packages. If the WG does not spend time collecting input, the WG will not be able to make informed decisions regarding solutions. =- Tim
_______________________________________________ dmarc mailing list dmarc@ietf.org https://www.ietf.org/mailman/listinfo/dmarc
