On Apr 26, 2014, at 4:02 PM, Dave Crocker <[email protected]> wrote:
> >> Forwarding alone shouldn't be blowing up messages. If you're doing >> something to the message content that invalidates the DKIM signature >> or causes the the from address and return path to not be aligned, >> that's where you're going to run into problems. If you don't modify >> the message in any way while forwarding it on, you shouldn't run >> into issues due to DMARC. > > > The above first sentence is a widely held view, using typical language. > > However the sentence is wrong in so many different technical and operations > ways and levels, that it mostly services as an example of the problem in > talking about email. At base, it relies on a simplistic model that produces > Procrustean assertions. At base, it's just wrong. > > Mailing lists take delivery of a message and posts a new one. Most mailing > lists get their utility by facilitating communication between original > authors and final recipients. Mailing lists vary in the value add they > provide in this process; it often includes modifying the original message in > a variety of entirely legal -- and often useful -- ways. > > Again, what they do is legal and useful, and has been for 30 years. > > Any assertion or implication that a mailing list, which re-posts a > legitimately and usefully modified message, is somehow doing something wrong, > is itself the problem. > > The mailing list is not the problem. The problem is over-application of > mechanisms or policies that render legitimate email non-functional. > > There are serious email abuses motivating the over-application. They > shouldn't be ignored. However, the fact of those abuses are serving to > create new ones. Calling the new problems are abuses sounds impolitic, but > how else should breaking long-standing, independent, legitimate email service > be characterized? > > It's entirely possible that the work-arounds being pursued, with new > modifications to rfc5322.From and rfc5322.Reply-To will suffice. But they > carry their own downsides. > > On the average, a hack to remedy a hack damages the system and often requires > more hacks. > > Architectures usually layer nicely. Hacks rarely do. > Dave, I’m not sure the original email from Paul Scott, was about him running a mailing lists, or something like this… So I think, it was best to put aside the mailing list issue and help him to solve his problem. Let’s focus on problem solving. It seems he just forward emails from the internet to their customers to their yahoo/gmail address via their personal domain he hosts … He should have noticed these errors earlier (DKIM failing), but it may not have had the level of visibility the yahoo/aol policy change brought. There are a few well known large forwarders/hosting providers that breaks DKIM when doing just a forwarding. DMARC is only highlighting them and encouraging them to fix their infrastructure. DKIM is an IETF proposed standard since 2007. Time the infrastructure be friendly with it.
signature.asc
Description: Message signed with OpenPGP using GPGMail
_______________________________________________ dmarc-discuss mailing list [email protected] http://www.dmarc.org/mailman/listinfo/dmarc-discuss NOTE: Participating in this list means you agree to the DMARC Note Well terms (http://www.dmarc.org/note_well.html)
