I've added a new module to Data-Aviary that specializes operations from Applicative, Monad etc. to the function type. This might be useful for spotting patterns in the type signatures of the operations that might not be obvious in their more general (type-class) type.
For instance to work with records I have a family of 'substs' - extended arity versions of the S (Starling) combinator:
subst :: (r -> a -> b) -> (r -> a) -> r -> b
subst f g x = f x (g x)
subst2 :: (r -> a -> b -> c) -> (r -> a) -> (r -> b) -> r -> c
subst2 f g h x = f x (g x) (h x)
subst3 :: (r -> a -> b -> c -> d) -> (r -> a) -> (r -> b) -> (r -> c) -> r -> d
subst3 f g h i x = f x (g x) (h x) (i x)
Now because I don't use them too often, when I go back to them I'll often think "Oh, they're just liftA2, liftA3" and swap them in the code introducing new type errors everywhere. I tend to forget that liftA2, liftA3 are a family of S' combinators rather than S combinators:
liftA :: (a -> b) -> (r -> a) -> r -> b
liftA2 :: (a -> b -> c) -> (r -> a) -> (r -> b) -> r -> c
liftA3 :: (a -> b -> c -> d) -> (r -> a) -> (r -> b) -> (r -> c) -> r -> d
