Thanks you all for getting on a more constructive line of dialog, a lot more can be accomplished by doing such.
Graham, I'll reiterate as mentioned from a previous post that blacklisting is a good component to an overall perimeter defense, and as I mentioned it has not been left ignored and in fact there has been several discussions over the last 1-2 months about what you've discussed wrt to aggragating attacks from many systems. I personally don't know the deails of how/where the VoIPBL project positions their servers to detect and find the offending IPs to blacklist to have any opinion as to how comprehensive and geographically relevant their sources are.
I have followed some of the fraud trends in the VoIP space ot know that attacks are getting more subtle and clever by attackers in the ways some of them initially do very quiet recon to find their victims which will usually go undetected by any of the various means, followed by their eventual targetted attacks which, if they come from a known blacklisted server would of course be blocked, or if was all concentrated from a single source would also get blocked fairly quickly but neither of these is always the case. As such, all forms of detecting and blocking have merit, and a readiness to track the trends and adapt is also important and what we would like to do going forward.