For those who have wondered how to implement Neal Gafter's example 'for eachConcurrently' API with BGGA Closures, here's one attempt:
There's a bonus 'for eachEntryConcurrently' in there as well.
I'm sure it can be improved (suggestions welcome) and it certainly deserves a decent explanation, which is coming, but I'll make one comment at this point:
Adding support for nonlocal transfers across different threads required two lines of code.
Two lines is easy. I like easy. Especially when threads are involved.