Re: Ada is not a failure.

[news.crd.ge.com!e7sa!groleau@uunet.uu.net (Wes Groleau X7574)]

In article <23541k$jja@nic.lth.se> dag@control.lth.se (Dag Bruck) writes:
>............... I was under the impression that software reliability
>in critical applications is primarily governed by the organization of
>the software development process, so applications written in assembly

Close.  The process ATTEMPTS to govern the reliability.  Just like our
political leaders ATTEMPT to govern us.  But we still have drunk drivers.
Now to  r e a l l y  stretch the analogy, do you suppose there might be
fewer of them in jurisdictions that prohibit alcoholic beverages?

Which is more dangerous: a drunk driver, or a bug in the software installed
in every brand X airplane?  (A bug that slipped past testing due to using
language Z; a bug that would never have slipped past an Ada compiler.)

>language need not be less safe than those written in Ada (stretching
>my case, I know); the main motivation for using Ada is the *cost* of
>developing applications with inferior languages.

Cost is part of it, reliability is the other.  Cost is of course higher because
more testing is needed in some languages to get almost the same reliability.