Existential types: HM versus System F

[Eduardo Ochs]

Could someone recommend me a paper that discusses the relation between
Girard's System F and Hindley-Milner type systems? I've been studying
system F lately (and Haskell's semantics is still somewhat mysterious
to me), and in System F existential types can be built from the basic
operations, while in Haskell it seems that one needs language
extensions to be able to use them...

  Thanks & cheers,
    Eduardo Ochs
    http://angg.twu.net/
    edrx@inx.com.br