Xref: utzoo sci.space:6051 sci.space.shuttle:855
Newsgroups: sci.space,sci.space.shuttle
Path: utzoo!henry
From: henry@utzoo.uucp (Henry Spencer)
Subject: Von Braun quote
Message-ID: <1988Jul10.003611.16575@utzoo.uucp>
Organization: U of Toronto Zoology
References: <1988Jun17.053132.5314@utzoo.uucp>, <3361@phri.UUCP>
Date: Sun, 10 Jul 88 00:36:11 GMT

> > Man is the best computer we can put aboard a spacecraft. --Von Braun
> 
> Last week's successful Ariane-4 launch put AMSAT Phase 3-C and two other
> satellites into [a very precise] orbit...  I'd like to see some fighter
> jock/astronaut do as well by flying a launch manually...

"The Lord has delivered him into my hands."  -Huxley

Yup, the computer did a fine and very precise job.  Now, Phil, cast your
mind back a little ways, to the Ariane launch of AO-10 (I think it was).
Much the same computer, running much the same software, controlling much
the same booster, put AO-10 into a nice precise orbit.  After which the
same computer, mindlessly following orders, proceeded with a venting
procedure that caused the third stage to catch up with and collide with
AO-10, damaging it seriously and causing a lot of headaches for you and
the rest of the Amsat crew.  The greenest student pilot could have
prevented that, if he'd been there.  You, of all people, should not be
lauding the unmanned nature of Ariane as an unmixed blessing.

I think we all agree that machines are generally superior for boring
jobs that have to be done exactly right, especially when there are tight
response-time requirements.  Even Von Braun, after all, built computer-
controlled launchers.  And I think we would all agree that humans
are generally superior for adapting orders to situations and coping with
the unforeseen.  Both the Solar Max repair and the Palapa/Westar retrieval
succeeded (despite one or two false steps along the way) even though the
original carefully-built equipment simply didn't work.  

The debate centers on the extent to which unexpected situations and
unforeseen problems can be removed by advance planning.  NASA, Arianespace,
etc. have been insisting for a long time that nothing is left to chance and
everything is foreseen.  They have consistently been proven wrong.  Sometimes
the equipment can be convinced to cope, and this is trumpeted as further
proof that humans are unnecessary in space.  Sometimes the equipment just
isn't up to handling a new situation, and this is written off as Just One
Of Those Failures One Has To Expect -- even if it wouldn't have been a
failure with a human on hand.  Do I detect a small inconsistency here?

Automatic equipment is the appropriate response to a well-understood,
simple, repetitive job like relaying communications, taking pictures, or
guiding a launcher.  Humans are the appropriate response to complex,
variable, unforeseen, one-of-a-kind situations like equipment failures,
unexpected changes in environment, and exploration of planetary surfaces.
To pick a close-to-home example, the success of satellite repair/retrieval
procedures has been inversely proportional to the reliance placed on
automatic equipment rather than humans -- compare the equipment-intensive
Solar Max repair, a near-disaster, with the human-intensive Leasat repair,
which worked so well that it was boring.

> Perhaps you should limit the scope of this statement somewhat.

If we're being pedantic, note that Von Braun did:  he said "spacecraft",
not "launcher", so Ariane isn't relevant.
-- 
Anyone who buys Wisconsin cheese is  |  Henry Spencer @ U of Toronto Zoology
a traitor to mankind.  --Pournelle   | {ihnp4,decvax,uunet!mnetor}!utzoo!henry