On Tue, 04 Jan 2005 15:41:12 -0500, John Denker <[EMAIL PROTECTED]> wrote: > Udhay Shankar N wrote: > > I just got a batch of spam: perfectly justified blocks of random-looking > > characters. Makes me wonder if somebody is trying to train Bayesian > > filters to reject PGP messages.
Or someone is trying to slip messages past bayesian filters trained to allow pgp messages. Most of these spams are awarded insanely high spam scores by spamassassin. > Another hypothesis: Cover traffic, to defeat traffic analysis. > > The procedure: send N copies. N-M of them are spam, sent to uninterested > parties. The other M parties are the intended recipients. Provided N>>M, > and other mild restrictions, they achieve plausible deniability. I've been getting spam with blocks of text strongly resembling pgp signatures appended for years now. Got about 250 of them last year. And, amusingly enough, they seem to keep up on their patches (the versions of pgp seem to keep up with the official releases). Still, the signatures would never verify, as there were invalid base64 characters in the signature block. -- GDB has a 'break' feature; why doesn't it have 'fix' too? --------------------------------------------------------------------- The Cryptography Mailing List Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
