At the risk of sounding hyperbolic, the utility of some means of surveillance in apprehending criminals is not itself a good enough criterion for justifying any

At the risk of sounding hyperbolic, the utility of some means of surveillance in apprehending criminals is not itself a good enough criterion for justifying any means of surveillance. If it were, so the hyperbolic argument goes, we could minimize crime by locking people in cages. Of course, the two extremes are ridiculous and so we use a balancing test: how does the state's compelling interest compare to the cost to personal liberties. If 99.8% of license scans/photos are not useful for apprehending criminals, then I'm not sure the technology can be justified. Moreover, as the parent points out, license metadata is another tool that could be used by third parties (e.g. advertisers, insurance providers, etc). It seems to me like the RFID suggestion could solve (or at least help alleviate) the problem of third party scanning while preserving some of that utility for the police. This, combined with reasonable retention limits (14-21 days) would make the whole matter much more palatable to me.