Note how the provided code enables a bunch of language extensions, most notably UndecidableInstances. (Yep, that's exactly what it sounds like!)

It's quite far removed from standard Haskell or full type inference.

On Sep 12, 2014 4:30 PM, "Brandon Allbery" <allbery.b@gmail.com> wrote:
On Fri, Sep 12, 2014 at 7:25 PM, Jay Sulzberger <jays@panix.com> wrote:
But this is impossible!  The type system of Haskell can be
handled by a Hindley-Milner-Damas style type checker.  Such a
type checker always comes to a halt with a Yea or Nay answer.  So
one cannot get a simulation of the combinators S, K, I inside the
Haskell type system.

What have I misunderstood?  And, in case GHC really does now
handle stuff beyond the HMD horizon, what does the New Core
language look like?

Standard Haskell is (or was) H-M extended with typeclasses. GHC moved beyond that years ago; internally it's System Fw (http://en.wikipedia.org/wiki/System_F#System_F.CF.89), and its Core language reflects this.

--
brandon s allbery kf8nh                               sine nomine associates
allbery.b@gmail.com                                  ballbery@sinenomine.net
unix, openafs, kerberos, infrastructure, xmonad        http://sinenomine.net

_______________________________________________
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe