On Mon, 2 Apr 2007, Bob Fiero wrote: > RBLs such as SpamCop gave me an idea a few years back. We should > build a virtual wall around the country. In theory it is a good idea, but in practice it is not clear if the rate of false positives or false negatives will not ruin the system. Even if you propose to use it only against malware, it can be the case that a server is hosted on an 0wned ADSL host with dynamic IP. You decide to block the whole subnet of the ISP -- now all the ISP customers are in trouble. Aside from technical problems there are also possibility for misuse both from insiders and outsiders. Now you don't need a DDoS to bring competitor's site down: just pay insider to include the site into the block list or put something that can be mistaken for malware in some obscure "user comments" section of the target site. Once the system is in place, it will be used to block other types of "bad" sites as well. Of course, nobody wants to make a public directory of "bad" sites, so there will be no easy way to know what is blocked. See <http://ice.citizenlab.org/?p=260> for discussion of the mission creep in such systems. -- Regards, ASK