On Fri, Apr 13, 2012 at 3:59 PM, Ludovic Courtès <[email protected]> wrote: > Noah Lavine <[email protected]> skribis: > >> I'm somewhat afraid, however, that the real solution is changing how >> we deal with parallelism, and that is a much bigger problem. > > And this is where functional setters come in. ;-)
Yes, functional data structures solve this. But the trouble is that you can't take advantage of this to generate better code unless your compiler knows that you will never mutate certain variables. Maybe you could put something together with lots of caching code and invalidating it if the assumptions change. Noah
