on Wed, Dec 17, 2003 at 11:39:51PM +0100, Magnus von Koeller ([EMAIL PROTECTED]) wrote: Content-Description: signed data > On Wednesday 17 December 2003 21:36, ScruLoose wrote: > > And, given the popularity of online blacklists that track IPs that > > are _actually__used_ by spammers, how does it make any sense to > > move backwards from something that's more accurate, in favour of > > something that's much, MUCH less accurate? > > Sorry? If the IP is dynamically assigned, how do you block it? That's > the whole point of not allowing DynIPs.
Set a TTL on the block. In practice, most "dynamic" IPs are in use by 24/7 connections which will hold that IP for days, weeks, or months at a time. *Intelligent* use of DynIP lookup lists would involve comparing a DynIP with an RBL, and limiting block times to reasonably appropriate period after first spam activity from the IP. As I've already said: statistically, odds are that *any* IP is going to generate more spam than ham. Why not block 'em all? Peace. -- Karsten M. Self <[EMAIL PROTECTED]> http://kmself.home.netcom.com/ What Part of "Gestalt" don't you understand? Inconceivable! - Princess Bride
pgp00000.pgp
Description: PGP signature
