procmail
[Top] [All Lists]

Re: Spammish? (Final for me)

2003-02-18 11:08:48
* fleet(_at_)teachout(_dot_)org <fleet(_at_)teachout(_dot_)org> [02-18-03 
11:59]:
On Tue, 18 Feb 2003, Professional Software Engineering wrote:

I don't understand what you're arguing about, or who with.

Sean, I'm not sure I do either.

I think I'm going to bow out of this thread, even though I started it.  I
understand what you are saying about degrees of spammishness; but to me it
just doesn't make a lot of (practical) sense.

My arguments are weak; primarily because I have insufficient numbers/data
to back them up.  Sooooo - I'm about to embark on a project to collect
those numbers.  Each of my recipes will be incrementing a counter.  I'll
let the counters collect data for six months, then review publicly.

False positives should be the only thing I have to account for manually.

All messages will go through all recipes and spam hits will be commented
in the message.

This is a false premise.  If I set up a spam trap that traps 500
positive spams in a period of time and there are *no* false positives,
I have not necessarily accomplished a reasonable goal.  During that
same period of time I may have received 2000 spam posts and I have only
trapped a small percentage.

You must also consider the number of spam posts *not* trapped by your
algorithm, ie: False Negatives.  And, you will have to also account for
these manually.

just my thoughts.  I usually just sit and watch, although I become
quite frustrated by TOFU and excessive quoting.
-- 
Patrick Shanahan
http://wahoo.no-ip.org                  Registered Linux User #207535
icq#173753138                                 @ http://counter.li.org

_______________________________________________
procmail mailing list
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail

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