17 Oct
2010
17 Oct
'10
6:15 a.m.
At any rate, if you intend to prove this for any arbitrary f, I can't tell you how to prove it formally because it's not true.
How do you know that? Can you give an example where it fails?
The problem with the code you originally posted was that it looked like this: f r = do r' <- something f r' something else -- this is dead code That is, the computation is non-terminating, because f simply calls itself recursively, with no base case. Regards, Malcolm