I should point out that there are now a couple of large networks playing around with code derived from the paper Curtis Villamizar presented at the NANOG meeting at NCAR. Initial testing with route-dampening seems to hold out some promise. Probably the main question now will be how aggressively to attack the issues of flapping prefixes. With the current default values, and given that it's not exactly peak usage time, there are some 100 prefixes from essentially all over the Internet being dampened. Sean.