microcode, compilers, and supercomputer architecture

From: Huw Davies <H.Davies_at_latrobe.edu.au>
Date: Mon Apr 5 21:23:17 1999

At 21:07 05/04/99 +0100, Tony Duell wrote:

>Well, I don't drive yet, but when I do, I sure as hell won't have a car
>with ABS.
>
>The reasons are that (a) I am not going to trust my life, and the lives
>of others to an undocumented system that could possibly fail, (b) a good
>driver can stop a car in a shorter distance than an ABS system can under
>some conditions and (c) if it does fail you have to use the brakes
>differently than you do with a working ABS system.
>
>No thanks. I'd rather trust my skill (and thus have to learn to do things
>properly) than trust a microprocessor.

Have you done an engineering analysis of the comparative failure rates of
ABS and drivers? I know of one person who is alive today because the car I
was driving had ABS (wet tram tracks have little or no grip). I've no idea
if I'd have hit her in my own car (no ABS) I tend to think I'm a reasonable
driver but in a panic situation you can't be sure. One of the really good
things about ABS is that the panic response is just what you need - you
really have to push hard to invoke ABS in the dry. My new car is going to
have ABS, and airbags and probably stability control as well - I just
haven't decided what to buy.

To put this slightly on topic, we all know that computer systems have
become more reliable in the last 20 years but does anyone have any real
figures to back this statement up? I exclude disks which have changed so
much that comparisons are rather unfair....

 Huw Davies | e-mail: Huw.Davies_at_latrobe.edu.au
 Information Technology Services | Phone: +61 3 9479 1550 Fax: +61 3 9479
1999
 La Trobe University | "If God had wanted soccer played in the
 Melbourne Australia 3083 | air, the sky would be painted green"
Received on Mon Apr 05 1999 - 21:23:17 BST