Re: [Haskell-cafe] Higher order types via the Curry-Howard correspondence