> On Sep 14, 2014, at 10:56 AM, Scott Kitterman <scott@xxxxxxxxxxxxx> wrote: > > > >> On September 14, 2014 10:40:51 AM EDT, Hector Santos <hsantos@xxxxxxxx> wrote: >> >> On Sep 13, 2014, at 9:49 AM, "John Levine" <johnl@xxxxxxxxx> wrote: >> >>>> Agreed, but just wanted to add one thing- doesn't the details of the >> whether the sender >>>> has to align or not depends on whether SPF or DKIM is used as the >> authentication method? >>> >>> No. Neither DKIM nor SPF have any connection to either the From: or >>> Sender: header other than what DMARC is trying to do. >> >> DKIM has a required hash bind to the 5322.From field data -- the only >> 5322 header signing requirement in DKIM. It's burned into the now DKIM >> now STD level specification. That's not a DMARC requirement, but one >> DMARC relies on having with DKIM. >> >> Maybe an errata is in order? > > The field is required to be signed. It's not required to have any particular value. It's most certainly not required to be related to the signing domain in any way (which is what DMARC does). > Major disagreement and why this problem is not solved. There is tremendous inherent value that could only be understand by working on mail systems offline, online for 30 years. The from field is a treasured field Scott and it's not just for network mail but local and all forms of telecommunications. There is a reason why it's the only one required to be signed and that's strongly because it was the original proof of concept since day 0. Domainkeys started the policy concept. DKIM improved it. One group tried to separate it and that failed. -- Hector Santos http://www.santronics.com