[Haskell-cafe] Re: (a -> [b]) vs. [a -> b]

Chad Scherrer wrote:

Unfortunately, I was trying to give a simplification of the real problem, where the monad is STM instead of []. Based on apfelmus's observation of why they can't be isomorphic, I'm guessing I'm out of luck.

http://www.haskell.org/pipermail/haskell-cafe/2006-December/020041.html

So in reality, I'm trying to construct something like f :: (a -> STM b) -> STM (a -> b)

Indeed, such an f most likely does not exist. What is the task you tried to solve with the help of f? I guess that either there is a way without or it just cannot be solved for mathematical reasons. Regards, apfelmus