[Haskell-cafe] Type inference

[Brian Hulley]

Brian Hulley wrote:
> Brian Hulley wrote:
>>          f :: (forall a m. a -> m a) -> c -> d -> (m c, m d)
>
>
> The above is wrong - there is no way to quantify m properly. This
> must be why intersection types need to be written with "&" after
> all....

What am I saying! It's right after all, and might be better than the & 
syntax because it makes the dependency clearer (assuming you can't write a 
function that is both Int->String  and Float->Int)