and the problem with that as an incentive, is that if you're waiting for perfect data before you take any kind of action, and the action is you write a paper, that the disease outbreak is over. that you -- the nature of disease control is sort of taking action with imperfect information in conditions of real ambiguity. it really is like battlefield command. and she just sensed that the institution had lost its nerve. >> you had this amazing moment where you describe how george w. bush reads john berry's book about the spanish influenza, and he decides that, in fact, the u.s. has a bad national health administration and he tries to reorganize it. does it work?