Re: [Haskell-cafe] Timing in atom -- how consistent?