On Tue, Feb 08, 2005 at 02:22:31PM -0500, Jonathan S. Shapiro wrote:
> On Tue, 2005-02-08 at 18:41 +0100, J�rgen Hermanrud Fjeld wrote:
> > Yes, and your current approach seems conservative enough to allow
> > future experimentation and proposals, assuming that you have left
> > overloading entirely as a future exercise.
> 
> I think that for the sake of pragmatics we are going to have to provide
> built-in overloads for arithmetic operators, but I think we can defer
> general overloading.
> 
According to my ideas about extensional polymorphism, this would
require special case handling of arithmetic operators in the
optimization, and that is not entirely shocking, as one would expect
arithmetic operators to be fast.

> Is this good enough for now?
> 
Yes, I think so, but you should certainly get the statements of people
more into type systems than I.

-- 
Sincerely | Homepage:
J�rgen    | http://www.hex.no/jhf
          | Public GPG key:
          | http://www.hex.no/jhf/key.txt

Attachment: signature.asc
Description: Digital signature

_______________________________________________
bitc-dev mailing list
[email protected]
http://www.coyotos.org/mailman/listinfo/bitc-dev

Reply via email to