Re: [Haskell-cafe] QuickCheck's co-arbitrary