From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: *
X-Spam-Status: No, score=1.1 required=5.0 tests=BAYES_40,INVALID_DATE,
	MSGID_SHORT autolearn=no autolearn_force=no version=3.4.4
Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site pokey.UUCP
Path: utzoo!linus!philabs!cmcl2!seismo!uwvax!pokey!dan
From: dan@pokey.UUCP
Newsgroups: net.lang.ada
Subject: bug-free programs
Message-ID: <260@pokey.UUCP>
Date: Mon, 7-Oct-85 11:05:08 EDT
Article-I.D.: pokey.260
Posted: Mon Oct  7 11:05:08 1985
Date-Received: Wed, 9-Oct-85 05:22:41 EDT
Distribution: net
Organization: U of Wisconsin CS Dept
List-Id: <comp.lang.ada>


   For anyone who is used to reading the horror stories in Software
Engineering News, the idea of bug-free large programs is a bit hard to
grapple with.  Large programs are so complex, and have so many states,
that exhaustive testing would take hundreds or thousands of years.
Even in the unlikely event that you do have a mechanical proof system,
that system is vulnerable to the input and output predicates for a given
code fragment.

   The important thing about Ada exceptions is that they include the
ability to trap constraint and range violations like the one which
would supposedly appear on the pilot's display.  Proper coding standards
will enforce the creation of programs in which these (unlikely but
possible) errors would be trapped and dealt with in an orderly way.

   Sure beats the heck out a C program that goes off the end of an array
and fires your ejection seat . . .

   -- Dan