Faster, NASA, Faster

http://www.nytimes.com/2009/12/21/opinion/21lu.html?partner=rss&emc=rss

by vorador • 17 years ago

67 51 17 years ago

djcapelis2• 17 years ago

Innovation comes at a cost. One has to be willing to accept failure.

NASA isn't at the moment. I'm not sure any nation is willing to accept high enough failure rates in their space programs to fly by the seat of their pants.

Increasing launching is maybe a start, but until the space program tolerates more failure true innovation might be tricky. High risk high reward work loses in that environment, no matter the lipservice paid towards that work in overviews.

idlewords6• 17 years ago

NASA missions have proven very innovative - and very tolerant of risk - when astronauts' lives are not at stake. The various missions to Mars are a good example.

Contrast this to the kind of extreme conservatism needed to operate manned missions, where every mission-critical system needs to be massively overengineered and multiply redundant.

djcapelis2• 17 years ago

From what I understand of the original rover designs they were originally designed to be extremely intolerant of risk and it's only after an intern was dicking around with a robot that actually worked that they saw the light and realized they had to go with the "bump around a bit and see what happens" approach.

Even then, I would argue that NASA's strength is more in finding solutions to extremely constraining problems (i.e. all their big saves) more than it is to finding solutions when nothing's the limit. Though it's not my field and what they do is pretty amazing, so really this is all just coming from me and my armchair here.

I don't really believe you that the missions have been at all tolerant of risk. I see very very little to back that up.

Look at how much of a stink got raised when units were converted wrong and we lost one probe. Do you really think NASA's a risk embracing kind of place?

IIRC the current reaction that particular issue is all schematics involving every single component must be in US units. Sure sounds like a fast and loose environment to me!

anamax• 17 years ago

> Contrast this to the kind of extreme conservatism needed to operate manned missions

Needed? By whom and why?

It's not the test subjects.

idlewords6• 17 years ago

This is a very good (if unintentional) argument against manned space flight.

There is no way that firing an elaborate terrarium full of large primates into space is ever going to be cheap, and there is no scientific basis for continuing to do it (other than the circular justification of learning more about how space affects the people we fire into it).

dmfdmf3• 17 years ago

I agree. Manned space programs are a waste of money. Space is a dangerous place inimical to life. There is nothing out there, no places to go, no economical resources to claim. The analogy to seafaring implicit in the term spacefaring is an emotional appeal to adventure and discovery. But going to the moon is not comparable to the discovery and colonization of America. I am happy we went to the moon, it is a great achievement but nobody can live there or on Mars. Though I oppose govt stimulus spending, if we must spend a fortune on an "adventure" program let's colonize the oceans or learn to sea-farm or drill to the center of the Earth or build giant pyramids in the Nevada desert. Such projects would at least have the virtue of being honest goals.

encoderer3• 17 years ago

The thing is, nobody has compared the moon and North America.

Instead.. the moon is analogous to the inhabitable rocky island just off the coast, just past the horizon.

Getting there was relatively easy. Vikings did it. It took a decade.

No, going several orders of magnitude further, several centuries later, that's the New World.

But nobody would've ever gotten there if they first hadn't made it to the inhabitable rock 20 miles off the coast.

stcredzero11• 17 years ago

This is a very good (if unintentional) argument against manned space flight. There is no way that firing an elaborate terrarium full of large primates into space is ever going to be cheap

One should only say something like that with some physical principle as its basis, otherwise, it's just pointy-haired boss flying by the seat of one's pants.

Some prominent physicist once published a "proof" of the impossibility of heavier-than-air flight in the New York Times. (One which treated atoms as billiard balls and ignored fluid dynamics.) I think it was Kant who once gave "the chemical composition of the stars" as an example of something we'd never know. Two very smart gentlemen who had good sounding arguments, but a mistaken physical basis. From this thread, I haven't any idea about your acumen nor the basis of your argument.

I think if someone told a subject of Queen Elizabeth, that someday teenagers could idly write a screed, to be read by ten-thousands or millions of others, and effectively pay only a pittance to do so, they'd be dismissed as a lunatic.

EDIT: energy to get a spacecraft to orbit? It's comparable to sending a 747 over the Atlantic.

idlewords6• 17 years ago

The physical principle is basic inertia. You need to accelerate a container large enough to hold a human being plus the systems needed to keep that person alive to 17,000 mph, and then decelerate them again without cooking them alive or smushing them into a paste against the inside of the vessel.

Your 747 argument is specious. The energy in a stick of firewood is comparable to that in an equivalent mass of dynamite. I invite you to set both on fire and report back on the difference.

gvb2• 17 years ago

"[E]nergy to get a spacecraft to orbit? It's comparable to sending a 747 over the Atlantic."

[citation needed] I have not done the math, but I find this quite improbable.

The space shuttle has two SRBs with one million pounds of solid propellant each plus more than 500,000 gallons of super-cold liquid oxygen and liquid hydrogen. http://www.nasa.gov/centers/kennedy/about/information/shuttl...

The 747-8I has 64,225 U.S. gallons of jet fuel. http://en.wikipedia.org/wiki/Boeing_747#Specifications

According to http://www.convertunits.com/from/gallon+%5BU.S.%5D+of+naphth..., one gallon of jet fuel contains 133.92e6J, so a full fuel load on the biggest 747 is 8.6e12J.

According to somebody google found, the energy to get the shuttle into orbit is 1.17E11J, but that is just the potential and kinetic energy, neglecting the losses due to air resistance, etc. http://www.physicsforums.com/showthread.php?t=229835

http://xkcd.com/669/

rgrieselhuber• 17 years ago

Seems like we would be better off with android-avatars. There is not really a need to send actual humans to just do the kind of early exploration we want to do at this stage.

chancho• 17 years ago

But Avatars shaped like ourselves are still a few decades off, and needlessly complicated to execute. Avatars of our cars have worked out well.

stcredzero11• 17 years ago

As we go further and further out, the lightspeed delay will become a bigger and bigger inconvenience.

rbanffy• 17 years ago

I see a problem with the article: while launching a dot-com or a program is somewhat cheap, testing a new rocket engine concept frequently involves a couple years of work before you can even build a small prototype.

The fact failing integration tests ruins a million-dollar piece of machinery and not infrequently kills a dozen people also doesn't help much.

What NASA does is the impossible, bordering the very difficult. What we see is the artifacts that fall outside the impossible space.

gaius2• 17 years ago

I think I speak for every sci-fi fan when I say, while we're not reckless, if NASA said hey, you may die doing this, but either way humanity will be a little bit closer to becoming a starfaring species, we'd be like, hell yeah.

noonespecial• 17 years ago

Or, failing that (and fearsome moral dilemmas aside) you could just have a lottery open to the public where the "winners" only had a 1 in 10 chance of coming back alive, but if they did, they would be millionaires and heroes. You would need a separate agency just to handle all of the applicants.

Don't even get me started on the nightmare element "reality TV" would bring to it.

bitwize• 17 years ago

"Who Wants to Be An Astronaut?"

Hosted by William Shatner.

aarongough2• 17 years ago

Absolutely. While in a lot of cases I would say that safety and the value of human life should be the first consideration, space is not one of them.

Tens (hundreds?) of thousands of people lost their lives trying to sail across the oceans to discover new lands, but the world as we know it could not exist without their sacrifice. Space is a frontier. Frontiers are dangerous. Trying to make them safe requires either a short period of intensely dangerous activity or a very long period of less dangerous activity. But space is too important to be taken slowly.

Personally, if NASA told me I could be an astronaut but there was a 30% chance I would die, I would still do it. And I am sure there are thousands more out there like me.

Additionally, what NASA should be doing: http://en.wikipedia.org/wiki/Big_dumb_booster

joubert• 17 years ago

Why doesn't NASA use Soyuz rockets then?

RyanMcGreal2• 17 years ago

NIH?

gaius2• 17 years ago

Or fund the development of Skylon:

http://en.wikipedia.org/wiki/Reaction_Engines_Skylon

(Altho' arguably ESA or even the British MoD ought to be, whatever anyone else thinks)

aarongough2• 17 years ago

That is awesome! I had never heard of it before either, which is an incredible shame. I personally think that spaceplanes like this are a great idea and are likely to be the way we finally get into space for real...

physcab2• 17 years ago

Well, they do. They buy seats fairly regularly and when the space shuttle is discontinued next year the Soyuz will be the primary rocket for doing space missions. The problem is that the rocket has a very limited payload capacity compared to the shuttle and it can only carry 3 people at a time.

RyanMcGreal2• 17 years ago

The author seems to be conflating two very different things both called "launching".

stcredzero11• 17 years ago

I don't think he's actually comparing a company launch to a rocket launch. I think he's likening a development iteration to a rocket launch.

anonjon2• 17 years ago

This is an awful analogy because silicon valley has a huge failure rate for companies. Even an established company can fail quite rapidly. Imagine launching the shuttle and 90% of the time it blows up spectacularly. That would not do anything good for the space program.

stcredzero11• 17 years ago

That's not the author's analogy, only a straw man resulting from your misplaced referent. A shuttle launch is not analogous to a company launch. A shuttle launch is analogous to a version or a feature deployment of one particular company.

A company that iterates a lot, and deploys a lot, gets good at iterating and deploying. An organization that preps, launches, and retrieves spaceships a lot gets good at that too.

More effort on proving the null hypothesis is suggested.

idlewords6• 17 years ago

The parent's point remains completely valid. Rapid iteration only works if you are allowed to fail. In the case of manned space flight, failure means you lose the crew.

anonjon2• 17 years ago

The way I read the article was that we would cut NASA up into a number of rapidly iterating teams, each analogous to a small company. If you take this model and comparison to heart, and realize that there are actually /two/ levels of 'looping' that make silicon valley effective.

One is competitive attrition, the other is rapid iteration. The attrition factor should be obvious, you have a large number of companies vying for success, and only the top few are even moderately successful.

There would have to be a large enough margin that it doesn't cost NASA much to fail horribly on quite a few of its projects.

Iteration is good because if you iterate rapidly enough, you are allowed a larger margin of error, because any errors in your previous version will be fixed relatively quickly in the next version.

But here's the unfortunately reality: Sending anything into space is horrendously expensive. It is not like programming where releasing and distributing a patch is a tiny part of the overall cost of the program.

When you screw up with a space ship, bad things happen (shit falls out of the sky, there is pollution to clean up), it costs money into the millions, and you have to fill out a TON of paperwork.

Its not like a major error in a rocket you are launching can just be 'pulled from the website' after the rocket has exploded.

Even if you /do/ go with cheap unmanned probes, you still have the massive pr disincentive of potentially failing every day for a month, and losing tons of money.

NASA would become a joke.

eoyola• 17 years ago

I don't think it's a straw man. Just maybe the closest to what you're thinking is a slippery slope, but it didn't seem like the parent was trying to refute the author's argument.

The author does deal with the concern:

"I recognize that NASA cannot push a system to launch more frequently than it is capable of, because this could mean overrunning the budget or, worse, cutting corners on safety. Instead, future systems should be designed so that they can be rapidly prepared for launching by small teams.

This would not only increase NASA’s ability to send up innovative payloads but also make launching systems more reliable. After all, the more a rocket is flown, the better it can be understood and the safer it becomes. Frequent launchings would also reduce costs per flight in the long run."

moron4hire• 17 years ago

This article is waaaay off base. First of all, NASA is trying to get into faster launch cycles, that's the entire point of the whole reusable rocket projects. Second, nobody can complain with the shuttle software defect rate: the last 11 versions had a total of 17 defects (http://www.fastcompany.com/magazine/06/writestuff.html). Seventeen. If anything, the rest of the world could learn a lesson from NASA, especially in the realm of medical technology.

stcredzero11• 17 years ago

I have a couple of friends who worked for NASA, and I've met and spoken with others who worked there. The part of NASA associated with the shuttle is a bit ossified. One of my friends ported the software that tracks the shuttle's trajectory, and determines abort trajectories. He did that either right at the beginning of this century or the very end of the last. In any case, it was very recent, yet the use of version control software was considered a radical and risky move on the part of his group!

dandrews• 17 years ago

So what specific problem in NASA's development methodology do you think would be fixed by your favorite VCS? From the few references I've bothered to google the existing development process is heavily regimented - if expensive - and their record is pretty good.

oofoe• 17 years ago

See also http://www.fourmilab.ch/documents/rocketaday.html - "A Rocket a Day Keeps the High Cost Away", written in 1993...

physcab2• 17 years ago

What's interesting is that despite the consevatism, the space shuttle program, I think, has the worst mission success record. They've had what--14 deaths in the past 20 years?

From what I learned on my recent trip to Kennedy Space Center, the Space shuttle program is actually incredibly risky compared to other and past rocket designs. It's got solid rocket boosters that once lit, cannot be turned off. It's got no abort sequence until those SRBs are finished firing. Hell, the rocket itself is asymmetrical and rocks back and forth before liftoff.

Contrast all this with the Soyuz and our very own Saturn V designs. Saturn V I believe had 100% mission success during all our trips to the moon. It's liquid engines could be shutoff almost instantaneously during an abort sequence AND it had a blasting cap on top that would fire if the crew needed to abort during launch. Oh and the rocket did all this while being something like 95% efficient.

Why NASA abandoned this program is beyond me.

supahfly_remix• 17 years ago

The space shuttle does have the ability to snatch satellites out of space and bring them back to Earth. No other craft does this, and this was probably an important feature during the Cold War. NASA doesn't talk much about the military uses of its spacecraft.

Faster, NASA, Faster

Related Stories