spf-discuss
[Top] [All Lists]

RE: the Seth Hypothetical

2004-10-25 22:02:39
--Seth Goodman <sethg(_at_)GoodmanAssociates(_dot_)com> wrote:

From: Michael Hammer
Sent: Monday, October 25, 2004 3:48 PM

<...>

I would ask however, are we really willing to say that there is no
meaningful difference in terms of the order in which scopes are
evaluated?

I think the answer is yes.  Someone please correct me if I am wrong, but
SPF parsing currently works by evaluating left to right until it finds a
pass or hits "all".  Even if it does find a mechanism that passes, the
original parser had to evaluate all mechanisms to make sure no syntax
errors were present, though I don't recall if that is in Mark's current
draft.  The system was designed to give a deterministic result regardless
of the order of evaluation of mechanisms.  If my interpretation is
correct (I believe Wayne is the expert on record evaluation), it
shouldn't matter what order the scope of modifiers was evaluated in, as
long as all listed scopes were evaluated.


Seth and Michael,

I think I'm in complete agreement with you both. I think I misunderstood when someone said ordering that they were using the results of one SPF record to decide what to check next, which seems silly. If SPF is successfully applied to multiple scopes (which I believe it will be, eventually) then I think there is a natural sensible order (such as order appearing in the datastream) or it can be left as a decision for the implementer.

The important thing for SPF to do is to make sure the result is determinable and consistent for each individual header or data parameter, without requiring the parser to go perform checks on other data items (which might never appear or might be in a different domain)


--
Greg Connor <gconnor(_at_)nekodojo(_dot_)org>


<Prev in Thread] Current Thread [Next in Thread>