I think the best criminological exposition of Prof. David Kennedy's "pulling levers" approach is in this law review article at Harvard. The rolling stone article makes mention of Kennedy's ideas being the lastest, greatest thing in the war on drugs strategy, and I have seen the way law enforcement embraces it, although in my opinion, an i2 approach would be better than the GIS approach in pulling levers. However, what's probably most interesting is the poor way criminologists conceive of outcomes. Prof. Kennedy actually thinks that in a properly-advertised abstinence regime, offenders will not only get arrested less often, but will start turning themselves in.