Just a quick clarification: I'm not saying that these were the *only* issues.
I'm saying that these were the ones that made me queasy enough to abandon
the effort. As an example of another important issue, I also think that a
(possibly limited) form of dependent type seems like a really good idea as
well.

On Sat, Jan 5, 2013 at 2:04 PM, Jonathan S. Shapiro <[email protected]>wrote:

> I seem once again to have a project for which I need a modern, extensible
> language, so I'm considering whether it is possible to resurrect BitC in
> some form. The main issues were:
>
>    1. Type classes vs. overloading vs. genericity
>    2. Subtyping/Subclassing
>    3. We got const-ness wrong.
>
> I'm going to take each of these up in a separate note, and I'd appreciate
> reactions and comments on the appropriate threads.
>
>
> Jonathan
>
_______________________________________________
bitc-dev mailing list
[email protected]
http://www.coyotos.org/mailman/listinfo/bitc-dev

Reply via email to