On Tue, Feb 17, 2015 at 12:21 AM, Matt Oliveri <[email protected]> wrote:

> On Tue, Feb 17, 2015 at 3:00 AM, Jonathan S. Shapiro <[email protected]>
> wrote:
> > But I now think that is the wrong approach, and it further turns out that
> > all my noise about static definitions was wrong. We *do* need to be able
> to
> > determine arity given type, but we do *not* need to worry about the cases
> > whose arity we cannot statically determine.
>
> Interpreted literally, that sounds like a contradiction...


Yeah. I wrote that really badly. What I was saying was: we can
syntactically infer arity at all statically visible definitions, and if we
treat it as a form of specialization, it will turn out that all of the
abstracted arities get solved by specialization.
_______________________________________________
bitc-dev mailing list
[email protected]
http://www.coyotos.org/mailman/listinfo/bitc-dev

Reply via email to