[PRL] POPL NOTES

Sam TH samth0 at gmail.com
Tue Jan 18 19:19:40 EST 2005


Some notes on mitch's notes.  I'll post my trip report later.


On Tue, 18 Jan 2005 18:21:19 -0500, Mitchell Wand <wand at ccs.neu.edu> wrote:

> 
> Sat Jan 15 09:11:44 2005 FOOL

> Note: KB used the PPT fadeout/fadein effect, which was very gentle
> and pleasing.

He was actually using Keynote, the Apple presentation application.
 
> Sam TH* & Eric Allen
> A Core Calculus of Metaclasses

> Andrew Black: what about upward and downward compatibility (well-known
> in smalltalk?)?.  Sam gave an answer that AB seemed to find ok, though
> I didn't understand it.

In smalltalk, metaclasses have a hook to get their instance class (and
vice versa).  This means that if you specialize the class, you have to
be careful not to break things.  In the model I presented, there are
no such hooks, so the problem doesn't arise.



> Ernst*, Togerson, Hansen
> Wild FJ
> 
> Wildcards are partially specified type argments to types.  Claim is
> that these improve interaction among parametric & subtype polymorphism

This is now in the new Java 5.0, but their formalization of it is not yet sound.


-- 
sam th
samth at ccs.neu.edu



More information about the PRL mailing list