The issue is not whether, necessarily, there's any problem with a positive coefficient. It was a mismatch between the predicted in their safety case in the computer codes that said it should be negative and the actual measurements in the reactor that turned out to be positive, which means there was something in those computer codes that was not modelling what was really happening in the reactor.
It is very important, certainly in a production reactor like the MAPLEs were, that you fully understand the neutron physics of what's going on all the time through the process. So it wasn't the fact that it was positive or negative. It was the fact that there was a difference between the prediction and the actual.