[SystemSafety] proofs

[Martyn Thomas]

On 22/11/2018 18:02, paul_e.bennett at topmail.co.uk wrote:

> As I know I have stated many times before, like hardware components,
> software components need to be supplied with a descriptive data-sheet that
> explicitly states what the functionality of the component is, the environmental
> expectations and limitations of use. Against such a data-sheet, you can then
> apply the proofs, tests and other quality determination to allow assessment of
> whether such a component is fit for purpose. After all, we already do this for
> the hardware side of things. Again, why is software so different?
>
> Just to be clear, the standard of documentation on most software I have seen
> is thoroughly inadequate to be able to make such assessments and it costs
> time (and hence money) to begin even a cursory evaluation.
>
> Regards
>
> Paul E. Bennett IEng MIET
> Systems Engineer
> Lunar Mission One Ambassador
> -- 

And, of course, the "descriptive data-sheet" should be as formal as
possible so that it provides the basis for proofs and for warranties.

But the software industry won't accept that formal specifications are
necessary or that warranties are desirable, and their customers continue
to let them get away with it.

Martyn


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 488 bytes
Desc: OpenPGP digital signature
URL: <https://lists.techfak.uni-bielefeld.de/mailman/private/systemsafety/attachments/20181123/b70dabf8/attachment-0001.sig>