[PRL] first-class polymorphic values

William D Clinger will at ccs.neu.edu
Sun Feb 25 16:10:30 EST 2007


Dave Herman wrote:
> Sam answered my question pretty well in this thread. I'm curious about
> whether there are opportunities for unintended type variable capture
> when you have a sufficiently sophisticated type language, in particular
> with polymorphic type application and free type variables.

Of course.

Some years ago, one of the quirkN languages I use
for the compiler course featured explicit universal
quantifiers and allowed polymorphic arguments to
functions.  There was no type inference at all, but
the soundness of the type checker wasn't entirely
trivial.

Will



More information about the PRL mailing list