[SystemSafety] What do we know about software reliability?

Michael Jackson jacksonma at acm.org
Thu Sep 17 16:52:31 CEST 2020


Peter: 

> On 16 Sep 2020, at 10:21, Peter Bishop <pgb at adelard.com> wrote:
> - as a standalone artefact, the software is "faulty" rather than reliable
> As a component in some operating environment, E, "reliability" becomes meaningful.

Yes. 

The software under discussion is a component in a cyber-physical biparite system  
comprising both the executed software and those parts of the physical world whose 
behaviour it is intended to govern. Software developers are programming this bipartite 
system---not the computing equipment alone. 

So regarded as a standalone artefact the software behaviour is meaningless, being 
one side only of a dialogue with an unknown partner. It can have meaning only in the 
context of the governed world with which it interacts. In this context, its "reliability" is 
an aspect of---and conditioned by---the "reliability" of the whole system. 

Michael Jackson



More information about the systemsafety mailing list