Repeatability, Reproducibility, Computer Science and High Performance Computing : Stochastic simulations can be reproducible too…

摘要

Many of us know the important work of Karl Popper, as philosopher of Science, specialized in modeling and simulation. Karl Popper is generally regarded as one of the greatest philosophers of Science of the 20th century. Among the criteria of the scientific method, which are intangible principles, we find refutability, non-contradiction and reproducibility. The latter is one of the conditions on which Popper distinguishes between the scientific or pseudo-scientific character of a study. Indeed, Science works by drawing “laws” or “principles” from reproducible observations whose main property is to be true as long as no observation has proved otherwise. Scientific conclusions can only be drawn from a well observed and described “result/event”, which has appeared several times, and has been observed by different people and/or studies. This criterion eliminates what seriously distorts the results as well as errors in judgment or even manipulations by scientists.