Posted: May 22, 2010 1:50 pm
I'm asking this on behalf of a friend...
- David Hume cautioned that we cannot assume that a theory is forever good just because it worked in the past. So how can we know when to call a scientific theory "robust"? OK, we can do the observation/experiments and dismiss the null hypothesis. But how many repeats are required to call this theory robust? One? ten? A hundred? A thousand? A million? A billion? What?