Re: design specifics --- the config label

Hadmut Danisch <[email protected]> wrote:
> When an MTA is receiving e-mail, it could ask the service:
> 
> I received a message from IPADDRESS, with SENDERADDRESS and
> MESSAGEID, is that ok? and the server would answer "yes" or "no".

  By focussing on EHLO/HELO and MAIL FROM, the design and deployment
of a solution becomes simple.  We can then use DNS to do simple domain
checking, which may address a large part of the spam problem.  Using
DNS also allows us to re-use an existing protocol and infrastructure.

  Sender address checking is currently done in some cases via 'ident'.
Message-Id checking may be done via another method.  But both are
outside of the scope of the work on this list.

  I would bet that in 99% of the normal situations, if a domain
consented to an MTA using its name in EHLO/HELO and/or MAIL FROM, then
any Message-Id checks would be redundant.

  Alan DeKok.