home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.lang.c++
- Path: sparky!uunet!gatech!darwin.sura.net!udel!rochester!rocksanne!news
- From: kirby@xerox.com (Mike Kirby)
- Subject: Re: Give me safe C++
- Message-ID: <1992Dec22.182256.3797@spectrum.xerox.com>
- Sender: news@spectrum.xerox.com
- Reply-To: kirby@xerox.com
- Organization: Xerox Corporation, Webster NY
- References: <1992Dec21.234459.20895@ucc.su.OZ.AU>
- Date: Tue, 22 Dec 1992 18:22:56 GMT
- Lines: 40
-
- In article 20895@ucc.su.OZ.AU, maxtal@extro.ucc.su.OZ.AU (John MAX Skaller) writes:
- >In article <1992Dec18.134937.14313@bony1.bony.com> richieb@bony1.bony.com (Richard Bielak) writes:
- >>
- >>If you are writing safety critical software - let's say software that
- >>controls the brakes in *my* car - I'd rather have you code in PASCAL
- >>and run with all the runtime checks on. I don't want my brakes to stop
- >>working just because you forgot an ampersand.
- >>
- > I'd rather have the code
- >implemented in a language in which all the run-time checks
- >had been optimised away by PROVING they were not required.
- >
- > What would you have your brakes do if the run-time
- >checks detected a program error? I think if you lived
- >you would be entitled to sue the manufacturer.
- >--
- >;----------------------------------------------------------------------
- > JOHN (MAX) SKALLER, maxtal@extro.ucc.su.oz.au
- > Maxtal Pty Ltd, 6 MacKay St ASHFIELD, NSW 2131, AUSTRALIA
- >;--------------- SCIENTIFIC AND ENGINEERING SOFTWARE ------------------
-
-
- In most saftey critical systems, you ASSUME that there will be errors in the
- software. With this assumption in mind, you create physical fail-safe's or
- programatic failsafes to ensure that specific problem states can never be
- entered.
-
- The problem with formal mathematical methods for proving software implementation,
- is that the proof only ensures that the implementation is a valid implementation
- of the requirements. It does not ensure that the requirements are correct.
- Faulty requirements will lead to failure, and formal proofs will NOT catch that.
-
- This discussion is starting to drift out of the realm of C++ discussions thouugh..
-
- Mike Kirby
- Xerox Corp
- E-mail: kirby.roch803@xerox.com
-
-
-
-