And that's not even getting into the issue of folks who are blurring the lines between genuinely empirical #science and computer-simulated "science", who pretend as if setting up something in a #software package and reporting on the results is like conducting an experiment, an _in silico_ experiment.
Is it? Computers do make slight subtle *errors* when doing heavy-duty numerical calculations. The models which are programmed into the software are themselves subject to doubt: at any time, a scientific model once thought to be rock-solid may be proved false by an awkward experimental observation.
(cont'd)