Curry's Paradox in Haskell

curry-paradox.hs

data Bottom -- Should be unconstructable.

newtype Curry = Curry { r :: Curry -> Bottom }

f :: Curry -> Bottom
f c = r c c

b :: Bottom
b = f (Curry f) -- But we can construct it!

{-
This is not terribly exciting. Afterall, you can also construct
a Bottom with

b :: Bottom
b = b

to construct a Bottom. And we don't care that Curry's Paradox 
is encodable in Haskell: It is not a total language.

Hence, proof assistants (which are total) ban recursive types
in the negative (contravariant) position.

The only interesting part is that this drives GHC (I'm on 9.4.5)
to err:

  Program error:
  Simplifier ticks exhausted

  When trying UnfoldingDone f
  To increase the limit, use -fsimpl-tick-factor=N (default 100).

-}