Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

[P]
NASA Conclusion: "One Line of Code"

By rusty in News
Wed Mar 29, 2000 at 11:48:50 AM EST
Tags: Science (all tags)
Science

You probably recall our other story about this issue, but NASA's official findings about the loss of the Mars Polar Lander are out now, and the Washington Post reports that the most probable cause was indeed spurious signals produced upon deployment of the landing legs. Some more info, and a bit of editorializing, below.


The basic problem was that deployment of the landing legs produced signals that made the lander believe it was safely on the ground. It therefore shut down it's decent thrusters about 130 feet above the ground, and hit the surface of the planet at 50 MPH, exploding, in the words of JPL's John Casani, "like... one of those Bouncing Bettys."

NASA management is taking the public blame for the failure, saying they pushed "faster, better, cheaper" too far, making the project understaffed, underfunded and overstressed. The problem, they claim, could have been avoided with "one line of code," presumably to test with some other criteria when the lander received a signal that it had touched down.

"One line of code" makes for good news copy, but I do have to note that I've heard this phrase from management many, many times, and so far, it hasn't once turned out to be true. It's rather unfortunate that this is the way it was expressed, because it makes the problem seem like it would have been trivial to fix. Alone, it would have been. But the reason it wasn't found is that deploying an interplanetary probe is a huge task, with literally millions of "trivial" problems to fix, and, as seen in this case, if even one of those is overlooked, the results can be disastrous.

The real issue here isn't one line of code. The issue is, how committed is the United States to pursuing a space program? Better, faster, cheaper is all well and good, but sending a craft to Mars isn't the same as building a car, or even an airplane. "Cheaper" must be understood relative to other space programs. NASA has finally seen exactly how far they can push this philosophy, and it's to be hoped that future programs will be sufficiently funded to achieve their goals.

To paraphrase a well known programming rule, space exploration should be exactly as expensive as it needs to be.

Sponsors

Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure

Login

Related Links
o our other story
o official findings
o Washington Post
o most probable cause
o Also by rusty


Display: Sort:
NASA Conclusion: "One Line of Code" | 13 comments (13 topical, editorial, 0 hidden)
Actually, the first thing that ran ... (3.00 / 1) (#4)
by pwhysall on Wed Mar 29, 2000 at 08:39:59 AM EST

pwhysall voted 1 on this story.

Actually, the first thing that ran through my head when I saw the words "one line of code" was the old saw:

Every program has at least one bug, and at least one line of code that can be removed.

Therefore, by induction, every program can be reduced to one line of code that doesn't work.

:)
--
Peter
K5 Editors
I'm going to wager that the story keeps getting dumped because it is a steaming pile of badly formatted fool-meme.
CheeseBurgerBrown

Re: Actually, the first thing that ran ... (none / 0) (#9)
by eann on Wed Mar 29, 2000 at 04:44:49 PM EST

Yeah, that's 'cos I used it yesterday as an aside in the UNIX "security" article. :)

Our scientific power has outrun our spiritual power. We have guided missiles and misguided men. —MLK

$email =~ s/0/o/; # The K5 cabal is out to get you.


[ Parent ]
Re: Actually, the first thing that ran ... (none / 0) (#11)
by rusty on Wed Mar 29, 2000 at 11:53:16 PM EST

Indeed. I recalled that as well, and believe it or not, that's the first time I'd heard that. Great line.

By deduction, I'm pretty sure that the one, non-working line of code is in fact:

while (1) {do everything }

I'm not aware of a language in which that would work. Except maybe perl4. But only if 'everything' was a subroutine defining a lot of other stuff, and it'd still be a bareword, and wouldn't work when you used -w. You did use -w didn't you? ;-)

____
Not the real rusty
[ Parent ]

I must confess (none / 0) (#13)
by pwhysall on Fri Mar 31, 2000 at 02:40:58 AM EST

I only read the first paragraph or so of that, and then work beckoned...
--
Peter
K5 Editors
I'm going to wager that the story keeps getting dumped because it is a steaming pile of badly formatted fool-meme.
CheeseBurgerBrown
[ Parent ]

Don't blame it on "better faster ch... (none / 0) (#5)
by marlowe on Wed Mar 29, 2000 at 08:58:05 AM EST

marlowe voted 1 on this story.

Don't blame it on "better faster cheaper". It can't be that simple. An expensive Mars probe has failed too. Blame it on bad management at NASA. And accept no excuses from them. They need to quit screwing up. Throwing more money their way right now would be to reinforce the wrong behavior.
-- The Americans are the Jews of the 21st century. Only we won't go as quietly to the gas chambers. --

Re: Don't blame it on "better faster ch... (none / 0) (#6)
by analog on Wed Mar 29, 2000 at 12:36:52 PM EST

Don't blame it on "better faster cheaper". It can't be that simple.

Actually, looking at the big picture, I think it can. When Congress first started handing NASA the big budget cuts, Goldin told them that more probes would be lost; as the sum total of the losses would still be much less than the loss of a single probe under the old way of doing things, it was deemed that this was acceptable. Although they've had the unfortunate circumstance of two in a row going bad, AFAIK they're still within the original parameters of the Mars program.

Spaceflight is inherently risky. If you're going to send up spaceships, you will lose some. If you're going to make it a primary design goal that they be inexpensive, you're going to lose more than if you concentrate more on other aspects.

The people designing these things are human, just like you and me. They can't cover everything, especially when 90% of what you're dealing with is unknown by definition. Can you point a finger in hindsight and say "you should've done blah?". Of course. The trick is knowing what to look for ahead of time, and every time Congress cuts NASA's budget, they add to the list of things that can't be accounted for.

[ Parent ]

Re: Don't blame it on "better faster ch... (none / 0) (#7)
by rusty on Wed Mar 29, 2000 at 12:48:13 PM EST

The preceeding is all true. The lesson from this episode, I think, is that a lot of people have learned that when they said "better faster cheaper" they really just meant "cheaper." Congress, and the US public, still expects NASA never to lose a probe. And that is just not possible. So we need to evaluate-- are we willing to accept lost craft, or are we willing to spend what we need to to lower the number of potential losses? We've now lost $360 million worth of spacecraft trying to get to Mars. So the current cost-benefit equation is:

$360,000,000.00 = Some lessons about how not to run a space program

Would it not have been better to have it look like:

$1,000,000,000.00 = Lots of valuable scientific data about Mars

Basically, we spent 360 mill to learn that 360 million will not buy you any successful space missions. I hope we don't keep spending "small" amounts of money to learn this over and over.

____
Not the real rusty
[ Parent ]

Re: Don't blame it on "better faster ch... (none / 0) (#8)
by Anonymous Hero on Wed Mar 29, 2000 at 03:14:17 PM EST

Congress, and the US public, still expects NASA never to lose a probe. And that is just not possible.

The saddest thing about this is that these losses have usually been planned for. I don't think anyone considers the loss of a probe truly 'acceptable', but that we will have losses is built into the program. I read an article once about the original planning for the Space Shuttle program; if we had actually had the losses they expected, we'd have lost at least three of them by now.

If you look at NASA's track record, they have had some fairly big losses (in terms of press coverage if nothing else), but overall it's unbelievably good. If you compare them to other space agencies (and in particular to the only other one of their size, Russia's), they have been spectacularly successful. I think in many ways they're a victim of their own success; they've been so good at minimizing the inevitable losses that we take it for granted.

[ Parent ]

That's not really right (none / 0) (#12)
by pwhysall on Fri Mar 31, 2000 at 02:37:01 AM EST

The Mars probe was not expensive.

Galileo was expensive (>$1.5B?) and it still works.

MPL was cheap 'n' nasty.
--
Peter
K5 Editors
I'm going to wager that the story keeps getting dumped because it is a steaming pile of badly formatted fool-meme.
CheeseBurgerBrown
[ Parent ]

I think the basic problem is that C... (none / 0) (#3)
by analog on Wed Mar 29, 2000 at 10:53:09 AM EST

analog voted 1 on this story.

I think the basic problem is that Congress is telling NASA what it needs to do the job, rather than the reverse.

I was going to get off on a rant here, but it's just too early. ;) Suffice it to say that we appear to be electing our representatives on the basis of the quality of their soundbites, and we're getting what we deserve. As you pointed out, it appears that NASA is fighting back in kind; I'm sure they're lobbying the "if we had just a little more funding" angle.

The perfect qualifications for being elected to Congress these days appear to be being a rich moron with a big mouth. NASA is just another in a long line of symptoms of a much bigger problem.

How expensive was it to send the ea... (none / 0) (#1)
by hattig on Wed Mar 29, 2000 at 11:20:01 AM EST

hattig voted 1 on this story.

How expensive was it to send the early ships to other continents? Sure, to an island 1 mile out from the shore they might have fared well, but it was very hard to take these early ships out into the ocean - the technology wasn't advanced enough. Yet people managed to get from Africa to South America and back over 10,000 years ago on reed boats (the equivalent of these cheap probes and current space travelling technology). One day we might get the spaceship equivalent of a tall ship, then even bigger ships, then someone will build the spaceship equivalent of the Titanic (this space ship is so big and strong it can fly through planets and suns and survive - but we better not graze them) and then, one day in 2000 years time, we will have the equivalent of luxury cruise liners and we will do our duty free shopping on Mars with a 4 hour journey time with the equivalent of a ferry.

End ramble.

Hmm... here's the missing line... ... (3.00 / 1) (#2)
by Emacs on Wed Mar 29, 2000 at 11:41:45 AM EST

Emacs voted 1 on this story.

Hmm... here's the missing line... If (landing_gear_sensors != NULL) If anyone at NASA is interested in hiring me please send me an email :)

Re: NASA Conclusion: "One Line of Code" (3.00 / 1) (#10)
by your_desired_username on Wed Mar 29, 2000 at 05:43:39 PM EST

#include<lines_of_code_are_always_misleading_bullshit_rant>	 

'One line of code'? The MPIAT (Mars Program independent Assesment
  Team) does not support this notion. It cites poor communication,
  inexperienced management, and insufficient funding (with poor
  communication being the biggest problem, and probably the cause of
  the insufficient funding).

As for the spurious signal problem, the MPIAT report views this as a
  symptom of the root causes; they believe that more experienced
  management would have led to better pre-launch testing, which
  (probably) would have discovered the problem before the launch.

I suspect Washington Post came to the conclusion that an article blaming
  software bugs would draw more readers than an article blaming poor
  communication. 


NASA Conclusion: "One Line of Code" | 13 comments (13 topical, 0 editorial, 0 hidden)
Display: Sort:

kuro5hin.org

[XML]
All trademarks and copyrights on this page are owned by their respective companies. The Rest 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!