[SystemSafety] CbyC and unit testing

[Olwen Morgan]

On 04/07/2020 22:12, andy at the-ashworths.org wrote:
>
> In fairness a client report would usually include caveats about the 
> use of the statements contained therein and would usually be part of a 
> contract which from personal experience will also define how comments 
> from a client can, and should, be used. Taking an email comment out of 
> context and extrapolating it to be equivalent to a conclusion 
> contained in a client report is a little unrealistic.
>
I agree but it remains the case that PBL's statement as given is so 
lacking in due qualification that it *invites* misunderstanding and 
glosses over things that might foreseeably go wrong. Moreover, I believe 
that there are quite reasonable grounds for believing that the 
unqualified statement may actually be incorrect.

It's one hell of a big claim to say that there are circumstances in 
which you can omit UT. I can think of many reasons why it may be wrong 
and that, in consequence, failure to retain UT as a check on the tool 
chain and process is unacceptably risky.

Does anyone here honestly believe that you could successfully defend 
omitting UT in an action for negligence if a system developed using CbyC 
failed and killed someone as a result of a defect that could have been 
detected by UT?

I'm beginning to wonder whether some in the formal methods community are 
becoming as brain-dead as the lumpenengineeriat.


So far my take on this thread is:

David Crocker: Got it right.

Martyn Thomas: Terse (not a fault) but position not yet fully clear.

PBL: Has made a statement that is wide open to misinterpretation and 
should not be allowed to pass without vigorous challenge.

Me: Still not quite seeing what MT is getting at and utterly staggered 
at what PBL said - indeed very trenchantly so.


I have asked: What could go wrong that might bring into question the 
proposition that CbyC makes UT unnecessary?

In response, I have received counter-question but so far no answers.


Olwen


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.techfak.uni-bielefeld.de/pipermail/systemsafety/attachments/20200705/f6d9692e/attachment.html>