Re: [Haskell-cafe] Dynamic choice of "reverse" implementation