> The only prudent approach is to assume all results are bad until proven otherwise. As is true with results from people. > But this approach negates much of

> The only prudent approach is to assume all results are bad until proven otherwise. As is true with results from people. > But this approach negates much of the incentive to pay for questionable results. I'm not sure that follows. Even the crack-addled human informant has always been paid for questionable results.