[ietf-dkim] Re: Role of Sender header as signing domain
hsantos at santronics.com
Fri Dec 1 18:35:04 PST 2006
Charles Lindsey wrote:
> On Thu, 30 Nov 2006 22:27:04 -0000, Douglas Otis <dotis at mail-abuse.org>
> However, what is become clear is that verifiers will have to be prepared
> to treat things differently according to the ability of their users.
In my mail system engineering experience, thats that last thing you want
in this protocol. All systems must have standard and consistency
protocol behavior. At a minimum, what you are describing is "extra"
local policy based, implementation dependent considerations. Your ideas
for Local policy is not something you can dictate on others.
> Some possible policies might be:
> 1. Drop/quarantine suspicious messages regardless.
> 2. Give the user sufficient information to make the decision himself,
> but by a mechanism that will work with current MUAs.
> 3. For those users with specially adapted MUAs, communicate with them by
> whatever protocol has been standardized for the purpose.
> However, I understood that this group was primarily charged with
> producing a system that would work within #1 or #2.
This is not stuff you put into a standard protocol.
More information about the ietf-dkim