> So, we cannot trust even formal verification? it's explained in the article, 2 unit tests, 0 integration tests. The formal verification appears to prove corr

> So, we cannot trust even formal verification? it's explained in the article, 2 unit tests, 0 integration tests. The formal verification appears to prove correctness of the 2 pieces independently, but not of the composition. > Haskell, here is your opportunity :-) you're just moving the problem to the correctness of the compiler.