This thread http://www.haskell.org/pipermail/glasgow-haskell-users/2010-October/019360.html convinced me that GHC's rules for implicit quantification are unnecessarily complicated.

---

== The current spec == The current spec seems to be this:

• Define an "implicit quantification point" to be (a) the type in a type signature f :: type if type does not start with forall; or (b) a type of form (context => type), that is not immediately enclosed by an explicit forall
• At each implicit quantification point 'ty', working outside in, GHC finds all the type variables a,b,c in 'ty' that are not already in scope, and transforms 'ty' to (forall a,b,c. ty).

Note that

• The argument of a constructor is not an implicit quantification point, so that

data Foo = MkFoo (a -> a)

is an error, and does not mean

data Foo = MkFoo (forall a. a->a)

• Implicit quantification points may be nested but the inner ones are effectively no-ops. Example

f :: Int -> (Eq a => a -> a) -> Int

There are two quantification points: the whole type, and the (Eq a => ...). But when the outer quantification point wraps forall a around it, the inner quantification point has no free variables to quantify. So we get

f :: forall a. Int -> (Eq a => a -> a) -> Int

---

== The new proposal ==

The proposed new rule is this:

• Implicit quantification applies only to an entire user type signature that does not start with forall.
• For such signatures, find all the type variables a,b,c in the signature that are not already in scope, and prefix the signature with forall a,b,c.

I believe that the only observable changes in behaviour would be

• In a data type declaration

data Foo = MkFoo (Eq a => a -> a)

you'd get an error "a is not in scope", just as you would with

data Foo = MkFoo (a->a)

To me this seems more consistent behavour.

• Inside an explicit forall you would get no implicit foralls:

f :: forall a. (Eq b => b->b) -> a -> a

would yield "b is not in scope", whereas at present it behaves like

f :: forall a. (forall b. Eq b => b->b) -> a -> a

Trac metadata

Trac field	Value
Version	6.12.3
Type	FeatureRequest
TypeOfFailure	OtherFailure
Priority	normal
Resolution	Unresolved
Component	Compiler
Test case
Differential revisions
BlockedBy
Related
Blocking
CC
Operating system
Architecture