Russell McMahon wrote:

> Time after time after time this sort of thing makes the case for manned 
> presence in space on leading edge missions to the maximum extent 
> possible.

You're kidding, right?

> Much more expensive, certainly. Much more demanding of 
> resources, certainly. Much more liable to produce tragedy rather than 
> just disaster. certainly. Much less likely to have really really really 
> silly things happend that blow the mission.

Lets see now...

Launch space shuttle Challenger when we KNOW that the boosters
have leaky 'O' rings and that the problem is exascerbated by cold.
Overrule the engineers opinions against launch when really cold
because of political expedency. "Must get teacher to orbit".

Launch space shuttle Columbia when we KNOW that falling tank debris
is an ongoing problem and could cause fatal damage.
Even though we SEE a piece of debris bounce off leading edge of
wing, we WON'T spend a few thousands of dollars to have a Baker-Nune
camera in Hawaii take a photograph of the wing, just to be sure.
Ignore engineers expressions of concern because the calculations
say that there wasn't enough energy in the blob to cause damage.

Lanch Hubble space telescope without a full aperature test of
mirror figure because is would cost a few 10's of thousands of
dollars to do the test.
Spend millions on a 'repair' misson to correct the fault.

Mars Polar orbiter. Mix up metric and english specifications,
but NOT notice it in spite of dozens of engineering reviews.
A simple test firing of the engine would have revealed the error.

Huygens. Forget to include instruction sequence to turn on precision
oscillator so comm channel will work.
Would it have been THAT hard for the on board firmware to have
checked that it was enabled and turned it on as part of the
descent protocol? Of course no one thought to have this failsafe
in spite of innumerable reviews.

> Certainly. Much much much 
> more capable in optimising and maximising information collection and 
> processing. You bet.

At -180C I think not.
Oxygen deprived, fatigued, whatevered, I think not.
Humans may be more adaptable to unforseen cirumstances but I'd take
a digital photo over a word description of a scene.

Robots have done and will do a FAR better job of exploring space
than man himself ever will. Their instrumentation does not halucinate,
or otherwise become corrupt, not to mention that it makes far
more precise measurements, and for a tiny fraction of what a manned
mission would cost.

Two mars rovers have returned far more useful data about possible
life on mars than an manned mission could hope to.
And for far far less cost.

> "Hey Jim, the rovers stuck against a rock, again. Walk over and give it 
> a kick, will you"
> Sure, but why don't I just spend a few hours picking up a few hundred 
> potential samples and examine them and keep the most promising ones? And 
> I can split open ones which look interesting, and I'll get some chemical 
> assays done on the promising ones tonight - then we won't have to wait 5 
> months for it do a small part of the same thing?"
> 
> "Hey Sam, there's no signal on channel 2 - can you check the receiver 
> please?"

And if the problem was a bad connector in an inaccessbile location
the signal path is irretrievable, regardless of how clever the humans
are.

> "Houston, we have a problem" "Now, where's my duct tape ...".

You actually make my point for me.
Apollo 13 had a problem that was caused by a manufacturing
defect that was documented, but never noticed or
acted upon (excess current in heater because of damaged insulation).
When the heater was later turned on, the tank exploded.

There will ALWAYS be failures. At least with robots, nobody dies.
And you can send 100 robot missions for the cost of one manned one.

> "Are you SURE about those thruster figures? They look suspiciously high 
> to me. Y' sure they were converted correctly - I'm surely not risking my 
> life without quadruple checking all the data. Lets do another simulator 
> run - after all, we've still got another 3 months to go."

But the simulator is NOT the same thing as the 'real' thing
because humans will NOT think of everything. A simulation is only
as good as it's inputs.
The IEEE article on the Cassini/Huygens data link and the
negative impact of doppler shift proves this. And JPL KNEW that
this would be a problem, but because the Italian radio company refused
to discuss the details of the receiver design (competitive knowledge)
it was not uncovered until a very persistant engineer PROVED that
there was a problem, but only AFTER the mission was launched.

> "The high gain antenna's jammed. Again. Looks like you get to do an EVA 
> after all."

If it jammed once, what's to prevent it jamming again and again
and again? Obviously a design problem if it jammed at all.

> "Out of focus / broken / bent / too big/small/ ...? Guess we better get 
> busy asap ..."

Ahhh, yes... humans. Like the one who panned the Apollo 14? camera
across the sun, burning the vidicon, rendering all further television
transmissions too dark. Even his hammering the camera didn't fix
an intrinsic design flaw, no manual mode. It's HUMAN's who are
making the mistakes. The machines just do what we humans tell them to do.

> Challenge: Run an autonomous vehicle Barstow to Vegas with well defined 
> route markers and conditions. Successful entrants - none. Got more than 
> a mile - one. Year 2004. Now try it with 10 year olds on trail bikes.

No, try it with 10 year olds driving the SAME vehicles (no driving 
experience), with blinkers on so that they can only see straight ahead.
Given enough experience and resources any system can be made to work.


The real problem, as I see it, is that missions always run over budget,
and to save money, testing is the first thing tossed. And invariably 
it's the missed tests that would have revealed the flaws in all of
the above disasters.

Robert

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist