> Only the latter (that the order where pure functions are applied doesn't matter) has been touted forever as a benefit of functional programming, and what supp

> Only the latter (that the order where pure functions are applied doesn't matter) has been touted forever as a benefit of functional programming, and what supposedly makes it "trivially parallelizable". Functions that return the same values when given the same arguments are substantially easier to reason about than functions that return values that depend on both the arguments passed in and some state somewhere else in the system. Additionally, functions that don't modify their arguments are relatively easy to parallelize. (Due in part to the fact that it's really easy to figure out locking in a share-nothing system.) Anyway. I'm no scholar, but I'm pretty sure that no knowledgable FP proponent has ever seriously said "FP lets you call your pure functions in any order whatsoever and get the same end result, every time.". Perhaps I'm misunderstanding your statement?