A comparison of work-rest models using a "breakpoint" analysis raises questions. Issue 4 (1st October 2020)