Clojurians Log v2

I did some more thinking: so a missionary task is required to be cooperative; meaning it can only do operations that don't take a lot of time. A clojure function does not meet this requirement; for this reason m/via was created. So guess that clojure functions run via m/blk or m/cpu go to one of two threadpools. I was wondering why there is no api to send a task to one of m/blk or m/cpu. I get it now: inside a task a time consuming call can ve iffloaded directly. So there is no need .. can be done inside the m/sp expression. But what znow wonder is what is the difference in an ap process that forks off subprocesses. Is it always clear that the fork cpu bound? What if the firk had to (m? ...) something to pass data elsewhere; say to another ap that writes values to disk. Isn't that then a io bound fork? I hope I am not seen as pedantic. It is just that missionary api is so powerful that one really had to think twice about everything.

> I did some more thinking: so a missionary task is required to be cooperative; meaning it can only do operations that don't take a lot of time. A clojure function does not meet this requirement; for this reason m/via was created. So guess that clojure functions run via m/blk or m/cpu go to one of two threadpools. I was wondering why there is no api to send a task to one of m/blk or m/cpu. I get it now: inside a task a time consuming call can ve iffloaded directly. So there is no need .. can be done inside the m/sp expression. This is correct. In general, any effect that is not an m/via is not bound to a single thread, in practice it should be expected to run on multiple threads successively during its lifecycle. The rest of the paragraph I don't quite understand