Are NetLogo models runs scientifically reproducible?
Yes. NetLogo’s agent scheduling algorithms are deterministic, and NetLogo always uses Java’s “strict math” library, which gives bit-for-bit identical results regardless of platform. But keep the following cautions in mind: • If your model uses random numbers, then in order to get reproducible behavior, you must use the random-seed command to set the random seed in advance, so that your model will receive the exact same sequence of random numbers every time. Remember that agentsets are always in random order, so anything you do with agentsets uses random numbers. • If your model uses the every or wait commands in such a way that affects the outcome of the model, then you may get different results on different computers, or even on the same computer, since the model may run at a different speed. (Such models are rare. These two commands are common, but using them in a way that affects the outcome is not.) • In order to reproduce model runs exactly, you must be using the exact same versio
