[racket] Redex multi-hole contexts

On 2/28/13 1:32 PM, Nicholas Alexander Marquez wrote:
> So in attempting to model an abstract machine for a parallel language (pH), I found it very unpleasant to use Redex.  I didn't care about the non-determinism resulting from which thread might possibly be executing at the moment; they should all conceptually be running in parallel, with the order being unimportant.
>
> Matthias mentioned that at one point Redex had multi-hole contexts, but that they were taken out and he doesn't remember why.  Does anyone know?

This thread doesn't answer the why part, but it gives some alternative 
approaches:

    http://lists.racket-lang.org/users/archive/2012-July/053000.html

David