On Fri, Feb 5, 2016 at 11:13 AM, Ned Freed <ned.freed@xxxxxxxxxxx> wrote: >> Rather than discussing this on the IETF list, wouldn't it be rather >> more productive for the discussants to get together and thrash out a >> draft on how to use STARTTTLS in SMTP? > >> RFC3207 was published in 2002. 14 years and several revisions to TLS >> later, it is probably time for a RFC3207-bis. > > Since the issue at hand is the ramifications of a policy change for > IETF lists, the answer is no, it wouldn't. > > It's unfortunate that such a policy choice requires a deep understanding > of how existing email software implements STARTTLS, but that's the situation > we're in. The point of eating the dogfood is process improvement. Not to get used to the taste. And the point is lost if we then create our own special dogfood. Capturing the process and the special sauce is what I am after. As far as policy goes, the admin of an IETF service should be permitted to change the config any time they like provided that they are compliant with an IETF spec.