Steelman: 1. Perhaps GP is just exceptionally sensitive to bad mental states or addiction. I know the social "sciences" have an entirely well-deserved bad repu

Steelman: 1. Perhaps GP is just exceptionally sensitive to bad mental states or addiction. I know the social "sciences" have an entirely well-deserved bad reputation, but it isn't good science to extrapolate from n=1 either. Pilot programs like this one are useful. 2. The obvious answer to your rhetorical question is the second one, but in this case it is irrelevant; the program is targeted at a random set of applicants, their reliability is unknown.