Anthony Atkielski <anthony@atkielski.com> writes: > Russ writes: >> So does trying to find the legitimate mail among a pile of spam. > The difference is that, in the first case, legitimate e-mail is lost, > whereas in the second case, legitimate e-mail is preserved. No, actually, there is no difference. That's the key realization that you've not yet reached. Missing and deleting legitimate e-mail because a human mistakes it for spam has the precise same end result as rejecting a message because a computer program mistakes it for spam, except that in the former case the sender generally does not even get a warning, whereas they may in the latter case. The only actual difference between the filtering methods is that one of them takes more personal time from a human and they have different false positive and false negative profiles. I can think of multiple incidents that I've personally witnessed where the computer program had a *better* false positive and false negative profile than the human. I think you need to think about this a little more. No one is not filtering spam. Everyone filters spam; they just use different techniques which have different impacts on the network, on their personal time, and on the information returned to the sender. If you, like another poster in this thread, are currently only receiving 5-10 spams a day, congratulations, you don't have to care yet. That level of volume is trivial to filter. That was me three years ago; I now get approximately 500 spam messages a day. -- Russ Allbery (rra@stanford.edu) <http://www.eyrie.org/~eagle/>