[SystemSafety] Two Preprints

Derek M Jones derek at knosof.co.uk
Tue Mar 15 18:38:35 CET 2022


Bertrand,

> 1 - only outrageously simple software can satisfy the requirements for the methodology. These SW don't need statistical evaluation.

This begs the questions of what counts as outrageously simple, and how do you
know that statistical evaluation is not needed?

> 2 - For those for which statistical evaluation is desired, there is no methodology.

No methodology to do what?

Build the software?  Statistical evaluation can be independent of
how the software is built.

No methodology to build statistical models?  Well not yet,
and perhaps never, but we could at least get out of the
sandpit.

-- 
Derek M. Jones           Evidence-based software engineering
blog:https://shape-of-code.com


More information about the systemsafety mailing list