[SystemSafety] Logic

[Derek M Jones]

On 18/02/2014 22:10, Heath Raftery wrote:
> I hope I'm not misrepresenting Derek's argument, but it seems to me this
> is the core issue. Derek claims that using formal methods on a small
> slice of a larger project does not constitute an example of formal
> methods scaling to a large project. I'm inclined to agree with Derek's
> contention that formal methods scale very poorly to large projects, and
> this paragraph from the paper backs that up.

Exactly.

> I however, still remain hopeful, and look forward to reading more
> examples of large scale use of formal methods.

Formal methods do have their uses and software developers should be
aware of them as an option.

-- Aimed at the general audience

The current culture of intellectual dishonesty in the formal methods
community is switching off interest from potential users.

Justifying excessive claims by using weasel words to show how
wording can be interpreted to reflect reality is not the way to
convince people to try something new.

Researchers should ask themselves whether their paper would stand
scrutiny from the advertising standards authority.
Soap power manufacturers have their claims scrutinized and I think
we all agree that claims made about the use of formal methods should
also meet this minimum requirement.

-- 
Derek M. Jones                  tel: +44 (0) 1252 520 667
Knowledge Software Ltd          blog:shape-of-code.coding-guidelines.com
Software analysis               http://www.knosof.co.uk