Louis Wasserman
The point here is that a MonadST instance guarantees that the bottom monad is an ST -- and therefore single-threaded of necessity -- and grants any ST-based monad transformers on top of it access to its single state thread.
This approach sounds like Fluet and Morrisett's monadic regions (ICFP 2004, JFP 2006), for which Oleg and I produced a easier-to-use implementation (Haskell 2008). You may find these papers helpful! http://ttic.uchicago.edu/~fluet/research/rgn-monad/ http://okmij.org/ftp/Haskell/regions.html#light-weight -- Edit this signature at http://www.digitas.harvard.edu/cgi-bin/ken/sig "Attending a mathematics lecture is like walking through a thunderstorm at night. Most of the time you are lost, wet and miserable but at rare intervals there is a flash of lightening and the whole countryside is lit up." - Tom Koerner