[Top] [All Lists]

Re: SPF receivers are weak link

2005-08-26 10:56:45

----- Original Message ----- 
From: "Stuart D. Gathman" <stuart(_at_)bmsi(_dot_)com>

  - The current ugly "kissing cousin protocol" SRS is not part of 
    the package, so those with forwarding issues that they know 
   are real for their system, can only use a neutral/softfail.

Forwarding issues are a problem for receivers.  Senders can't possibly
know about forwarders - because the receiver sets them up!  By the above
logic, senders can never use -all unless they discover all the forwarders
configured by all potential recipients - and track them in real time.

I've tried to stay "neutral" (excused the pun) on the blame issue because I do 
support SPF, but since you forced the issue, I will take a stance.

The problem is the protcol for both the SENDER and the RECEIVER.  SPF was 
introduced with a PARTIAL solution to close one loophole and workaround with to 
address the problem it couldn't address by introduced MORE loopholes.

Forwarding is just one problem.  Authorized MUA Aliasing is also a BIG real 
problem for SPF.

Thats a protocol problem when you attempt to join into the solution spectrum 
and only come out with with a partial solution and a potential for breaking 
currently working operations.

Administrative disciplines will mostly disgree, but we are talking about a 
technical principle only technical disciplines can better understand.  The 
former wishes to dictate POLICY and OPERATIONS over technical mechanical 
considerations. The latter thinks in terms of logic and automation with a 
disregard towards policy.  The two do not mix very well, ergo CONFLICTS.

Hector Santos, Santronics Software, Inc.