Type tree traversals [Re: Modeling multiple inheritance]

Ken Shan ken at digitas.harvard.edu
Wed Nov 5 01:47:23 EST 2003


Brandon Michael Moore <brandon at its.caltech.edu> wrote in article <Pine.GSO.4.58.0311041405360.25739 at sue> in gmane.comp.lang.haskell.cafe:
> There are two extensions here:
> 
> More overlapping: [...]
> Backtracking search: [...]
> 
> Overloading resolution: [...]

I'm sorry if I am getting ahead of Simon or behind of you, but have you
looked at

Simon L. Peyton Jones, Mark Jones, and Erik Meijer. 1997.  Type classes:
An exploration of the design space.  In Proceedings of the Haskell
workshop, ed. John Launchbury.
http://research.microsoft.com/Users/simonpj/papers/type-class-design-space/

?  There is quite a bit of design discussion there, and I am not sure
how much has been obsoleted by more recent advances.  A primary
consideration seems to be that the compiler should be guaranteed to
terminate (so type checking must be decidable).

-- 
Edit this signature at http://www.digitas.harvard.edu/cgi-bin/ken/sig
hqrtzdfg
aooieoia
pnkplptr
ywwywyyw



More information about the Haskell-Cafe mailing list