>>>>> "GAW" == Greg A Woods <woodsweird.com> writes:

>> You can do this today with spamassassin -- just forget the scores
>> published by the project and tune it with scores derived from your
>> personal sampling of mail. It will much better match your definition
>> of spam that way.

GAW> You can do that _today_ with a filter based on Bayes' Theorem. Indeed
GAW> that's how Graham's idea is expected to work. Didn't you read his
GAW> paper? You just feed it all your good mail and then feed it all the
GAW> spam you receive. If integrated into something like your IMAP server it
GAW> could classify your e-mail when it's first delivered to your inbox and

I did read his paper. It presented no actual software... A few
others have, but they are just proof of concepts as far as I see.
ESR's bogofilter could be integrated into some mail software, but is
not yet.

One needs such a filter first... You can use SA to tune it to your
mail flow today. What software can you use today that is or can
easily by integrated nicely into your mail system?

In the future, anything can be done (modified IMAP servers, modified
mail clients that identify things as spam for feedback to the filter,
etc.) I'm more interested in what I can accomplish today.

Tomorrow I'll worry about writing code to integrate the baysean
statistics...
-
To unsubscribe, send mail to majordomopostfix.org with content
(not subject): unsubscribe postfix-users

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]