Re[4]: Comparison of high-level concurrency models?

Hi Tanton,

TG> For distributed workflow, you need (and may have) looked at the
TG> map-reduce model and the open source hadoop implementation.  It is
TG> being used by yahoo!, facebook, veoh, and others and is a nice, easy
TG> to understand distributed model.

I've glanced at hadoop, but our environment is Windows and OS X, and
no Java.

We have a map-reduce model, yes, though I don't place it in the
"workflow" category. We're also looking at a tuplespace model, which I
really like, and could work very well given our toolset. We evaluated
Erlang and decided not to use it, but that doesn't mean we won't
emulate parts of its model.

Has anyone here used E?

I've looked at a *lot* of things, and done my homework to the best of
my ability, but there's nothing like experience. I'm totally with John
on making things mind-bogglingly simple. What I'm interested in are
the "simplifying restrictions" he has found to work well.

--Gregg