My general approach to programming is that every language at the moment is bad in serious, fatal ways, and that we are all just experimenting and practicing so

My general approach to programming is that every language at the moment is bad in serious, fatal ways, and that we are all just experimenting and practicing so that 50 years from now we might have a decent approach. Haskell is horrible in some ways, excellent in others, and this paper seems to offer a promising technique.