Overlapping, undecidable, incoherent -- or worse?

[Simon Peyton-Jones]

| > I agree, I have made it not terminate myself with
| > undecidable-instances,
| > I also think prolog style backtacking would be a good idea...
| 
| For what it's worth, I'll AOL this ("me too").  I know that for the
| area of Haskell I'm exploring (integrating it with OO languages),
| putting backtracking into the type system would instantly enable us
| solve some significant problems.

I'll bear this in mind.  I think the main implementation difficulty concerns functional dependencies: when backtracking out one will have to undo any unifications triggered by functional dependencies. 

Nothing difficult in principle, but the constraint solver is one of the more delicate parts of GHC because GHC's constraint language has become so complex.

Simon