[Haskell-cafe] Equality constraints in type families

[Manuel M T Chakravarty]

Manuel M T Chakravarty:
>> again, i gave a concrete example of how ghc behaves as i would  
>> expect, not as that decomposition rule would suggest.
>
> Maybe you can explain why you think so.  I didn't understand why you  
> think the example is not following the decomposition rule.

Actually, see

   http://hackage.haskell.org/trac/ghc/ticket/2157#comment:10

Manuel