How to read claims on this site

Same line as in the book:

Findings are measured results from named trials. Each should carry the numbers, the source file and the limits. They describe those trials and nothing wider.
Recommendations are my engineering judgement. Where they go beyond what I measured, the text should say so.
Open questions are things not established by the available results.

The first edition’s trials are small and specific. Do not treat them as general laws.