We are seeing the usual criticism of the Ferguson model: a mound of poorly documented C. But Britain is two months into the epidemic, and after much initial bun

We are seeing the usual criticism of the Ferguson model: a mound of poorly documented C. But Britain is two months into the epidemic, and after much initial bungling, new cases are still very high but dropping, and R_0 is < 1. If there was something seriously wrong you could by now point at the discrepancy between model and data. That doesn't seem to be the case, because the worst they can come up with is the shape of the code. It's a bit like the computerized proof of the four-colour theorem from UIUC that at that time mathematicians were deeply uncomfortable with. If there was a counterexample you could show it, and all the discomfort would suddenly vanish! But there isn't one, and we have to live with the uncomfortable fact that four colors suffice.