On Tue, 13 Apr 2004, Shaun T. Erickson wrote:
But what's the point of publishing my SPF records, if I did what you
suggest? If I tell people they can ignore the result of looking up my
records, then why bother?
The neutral and softfail results are very useful for "fuzzy" spam
logic such as bayesian filters. You can think of a neutral result as
increasing a "spaminess" score - and if other factors push the score
over a threshold, the mail is rejected or quarrantined. Bayesian filters
will automatically measure the proper weight (probability that the
message is spam given a neutral SPF result) if Received-SPF headers
are present.
For instance, the percentage of AOL mail with a 'neutral' result
received at my site that is spam is currently still 100%. However,
such a message would still get through if it was otherwise non-spammy.
--
Stuart D. Gathman <stuart(_at_)bmsi(_dot_)com>
Business Management Systems Inc. Phone: 703 591-0911 Fax: 703 591-6154
"Very few of our customers are going to have a pure Unix
or pure Windows environment." - Dennis Oldroyd, Microsoft Corporation