Graham's spam filter

Paul Rubin phr-n2002b at
Thu Aug 22 23:04:55 CEST 2002

Erik Max Francis <max at> writes:
> This doesn't sound like the right approach to me.  Instead, you should
> perhaps start with a "global" database that is a sample of fairly
> typical mail from your clients and typical spam.  These should be used
> as an initial "seed" to the system only; once a user starts actually
> actively using the system to filter his mail, it can tailor itself to
> his specific needs.  

I hope there's not a proposal up there to seed a global database with
words from users' non-spam private mail.  That would be a bad privacy
breach.  Who needs Carnivore/Echelon scanning your private email for
keywords if a spam filter is doing it instead?

More information about the Python-list mailing list