Methodology

Published experiments will record the hardware, software versions, settings, prompts, timing and measurement method needed to interpret the result.

Limitations, failed runs and later corrections will be retained where they affect the conclusion.