Johannes Waldmann <waldmann@imn.htwk-leipzig.de> has asked me to clarify my claims here. But I say: why not to ask Chris Okasaki himself? Manuel Chakravarty has recently published this: http://www.cse.unsw.edu.au/~chak/papers/WC07.html so he might take the question too.

From my perspective there are too many language extension proposals like Generic Haskell addressing deriving extensions, Named Instances (http://www.cs.uni-bonn.de/~ralf/hw2001/4.pdf ) and many other I do not care to remember. To my mathematical mind Functor should be made a superclass of Monad before any such extensions were considered. This however costs almost a Ferrari :-) Exactly a Porsche: http://research.microsoft.com/~simonpj/papers/haskell-retrospective/HaskellR... pages 62 - 64.

Regards, -Andrzej Jaworski