Re: SDL-News: SDL and safety critical software



Date view Thread view Subject view Author view

Peter B. Ladkin (ladkin(at)rvs.uni-bielefeld.de)
Fri, 18 Jul 1997 17:09:02 +0200


Jean-Loup, > My intention was, and still is, that the example questionned > in my March 3rd message just proves that [SDL use for > safety-critical systems] has been achieved successfully > and that I disagree with the opinion of [Peter Ladkin] I have looked again at my message of March 3 (dated 18:18:27 +0100). You say you disagree with my (at)opinion'. I made a number of assertions in that message. I think all of them are demonstrably correct. Please could you tell us precisely which of those assertions you disagree with, and why? > If the SDL semantic was in any way hazardous, how could any software > generated from SDL ever be qualified? The fact that several software > generated from SDL have been qualified demonstrates that the use of > SDL is not hazardous. You are asking how SDL could be used in safety-critical applications, if the (at)semantics' were (at)hazardous'. First, I feel constrained to point out that you are misusing the engineering term (at)hazard'. According to that use, semantics itself cannot be (at)hazardous'. However, the use of a language with imprecise semantics can certainly lead to system errors, which may create hazards. See Leveson, Safeware (Addison-Wesley 1995) for definitions. Of course, you may have your own meaning for the word, in which case it would help to see your definition. Second, the answer to your question is given by you yourself: > It is for these reasons that I am recommending the use of SDL for > critical on-board data system software [...] A final request. You forwarded your message to the SDL forum, which referenced my views, without copying me. It was forwarded to me by a colleague. I notice that you also did not forward your message to the safety-critical systems mailing list, from which I am sure you will obtain some further commentary on your recommendation. As a courtesy, when you reply to messages of mine, please could you also forward me a copy? Peter Ladkin ---------------------- ------------------------ Peter Ladkin, Professor fuer Rechnernetze und Verteilte Systeme ladkin(at)rvs.uni-bielefeld.de http://www.rvs.uni-bielefeld.de Snailmail: Universitaet Bielefeld, Technische Fakultaet, Postfach 10 01 31, D-33501 Bielefeld, Germany Tel: +49 (0)521 106-5326/5325/2952, Fax: +49 (0)521 106-2962 ---------------------- ------------------------


Date view Thread view Subject view Author view