Re: [Haskell-cafe] Modeling the performance of a non-trivial software app