LWN.net Logo

SpamBayes 1.0 released

SpamBayes 1.0 released

Posted Sep 30, 2004 12:28 UTC (Thu) by RobSeace (subscriber, #4435)
In reply to: SpamBayes 1.0 released by rmstar
Parent article: SpamBayes 1.0 released

> Secondly, just wading through all that spam essentially means that you are
> looking at it just the same. The fact that it is on its own folder does not
> make any difference.

That's really not true at all... It makes a very large difference, in fact...
It's really rather easy to spot the one good message buried in a pile of
spam... Just as it's pretty easy to spot the one spam in a pile of good
messages... But, it's a MUCH harder task to sort spam from non-spam in a
completely mixed mailbox, with roughly equal amounts of each... Such a
mixed environment requires very close attention be paid to every single
message to make a determination... But, when dealing with a collection of
messages that are overwhelmingly of one particular variety, you can just go
into a much quicker scanning-for-anomolies mode... One good message in a
pile of spam really does stick out like a sore thumb... I've experienced
it on a few occassions... (Though, not that often, because despite your
implications, the bayesian filters I've used really don't make too many
false-positives... Very rarely, someone might send something full of HTML
or some other junk, which trips up the filter, but at least for the mail I
receive, it's been extremely rare...)

> none of these methods has a remote chance of actually groking the
> difference between spam and legit mail

They may not UNDERSTAND the difference, but they sure do seem to be able to
differentiate anyway, despite being simplistic and relatively stupid little
algorithms... Personally, I don't really care if my spam-filter has an
intelligent grasp on what spam is; as long as it successfully weeds out the
spam, and leaves the good messages, it can be as dumb as Dubya, and be
using astrology and numerology to make the determination, for all I care! ;-)
And, from MY experience, bayesian filter really DOES work remarkably well...
So, like I say, that's the important thing to me...


(Log in to post comments)

Copyright © 2009, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds