* Re: Language choice and mission-criticality - Ada issues
[not found] ` <58hp0l$1rl@feep.rsn.hp.com>
@ 1996-12-19 0:00 ` Robert S. White
0 siblings, 0 replies; only message in thread
From: Robert S. White @ 1996-12-19 0:00 UTC (permalink / raw)
In article <Ew9Y2RAkLGuyEwCy@jessikat.demon.co.uk>, robin@jessikat.demon.co.uk
says...
>
>In article <5956fp$la8@flood.weeg.uiowa.edu>, "Robert S. White"
><WhiteR@CRPL.Cedar-Rapids.lib.IA.US> writes
...snip...
>> For a really "mission
>>critical" type of language - choose one that really requires you to get your
>>act together. New banking software often use Eiffel or Ada. Avionics
>>use Ada.
>Who's proved the ADA compiler, runtimes, checking software, hardware etc
^^^
>etc etc. Most damning of all who's proved the overall design.
1. Its Ada (not ADA) after Augusta Ada, the Countess of Lovelace. The first
computer programmer. An English lady.
... the rest of my arguments/points are avionics related ...
2. I have not heard of "a formal proof" occurring often for really involved
products. Mathematical routines and functions yes. More complex
multitasking solutions such as Flight Management Systems, GPS's, etc. No.
3. Compilers, runtimes, Real Time Executives etc are subject to U.S. FAA
DO-178B qualification requirements for use in commercial airlines. The
DoD Ada compiler validation suite is also helpful but not as important.
4. Low level unit testing (using the compiler and runtimes) does try to
do a thorough test of all logic paths and input data at and beyond
acceptable limits
5. Formal Qualification Tests are done using representative production
(not engineering built) hardware with the black boxes given subject to
test environments/scenarios designed to verify design requirments are
met.
_______________________________________________________________________
Robert S. White -- an embedded sys software engineer
WhiteR@CRPL.Cedar-Rapids.lib.IA.US --long/cheap alternate I-net address
^ permalink raw reply [flat|nested] only message in thread