spf-discuss
[Top] [All Lists]

Re: IESG evaluation of SPF

2005-04-07 04:39:55
On Thu, Apr 07, 2005 at 03:33:44AM +0000, Mark wrote:

Your suggestion covers that idea. But it may sound a bit too neutral. I
would, myself, probably more go for something which stresses the
(probable) unfavorable result of the action; like:

 "Checking other identities against SPF records is undocumented."

Or (stronger):

 "The result of checking other identities against SPF records is undefined."

And, of course, the existing:

  "Checking other identities against SPF records is NOT
   RECOMMENDED because there are cases that are known to give
   incorrect results."

As long as we can all agree on a wording which makes clear that using
"v=spf1" records for anything else than what they were designed for, is
undesireable because there are cases that are known to give incorrect
results, then I would say we are on the same page.

Part of the problem: The RFC editor seems to think SPF2.0 is
a successor to SPFv1.  As long as this problem is not eliminated,
all discussion about exact wording is useless as the words are
used in a different context.

Alex


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