Slashdot Mirror
Neal Stephenson On Rockets and Innovation
Dr. Gamera writes "Science-fiction author Neal Stephenson gives us his perspective on the history of the development of rocketry. He uses that history to illustrate the phenomena of path dependence and lock-in."
It's a neat article, as usual with Neal, but the ending is odd. He says the current state of rocketry is at a local maximum, it's not going to get appreciably better, and there may be other ways of putting stuff in orbit that are better, and then he says he doesn't know why we aren't trying those other better things. This, after spending the previous twenty paragraphs writing about how the US has spent four trillion dollars to get to the top of this local maximum, and the old USSR spent about the same, and in the process we've established a huge military-industrial complex based on the money still flowing into that development path, with lots of political inertia greased by manufacturing and administrative money going into congressional districts... and he wonders why we're not considering spending another trillion dollars on a different, unproven system that would probably involve taking money from the people who are now getting it? He's already answered his own question, and that's surprising because he's a very bright person and does a good job of analyzing the subject.
Nostalgia's not what it used to be.
Stepheson makes this point late in the article:
"There is no shortage of proposals for radically innovative space launch schemes that, if they worked, would get us across the valley to other hilltops considerably higher than the one we are standing on now—high enough to bring the cost and risk of space launch down to the point where fundamentally new things could begin happening in outer space. But we are not making any serious effort as a society to cross those valleys. It is not clear why."
It's somewhat clearer why, to me.
I want to buy a more fuel-efficient car, and keep my current, less-efficient car. My current car is useful for many things, but commuting to work could be done by another, more efficient one. Here, however, is the rub. Despite the improvement in fuel economy, it is still a net increase in cost to me for a fairly long time. Acquisition, insurance, and upkeep consume most of the fuel savings. Yes, it would be better for he environment also, but that doesn't immediately or directly impact my costs very much. So I put off buying that car.
Our current methods of delivering object into space work well enough, and the alternatives are both unproven and not sufficiently advantageous to warrant immediate adoption.
However, as we re-enter manned space exploration, we will be looking for heavy-lift options that don't actually exist today, and those present the opportunity to develop new methods. Avoiding the vertical portion of a rocket launch also avoids the need for massive thrust to overcome gravity that directly. Stephenson alludes to this, and 'space planes' are the current focus, along with some multi-mode concepts. NASA'a failing Ares program is a fair example of lock-in that Stephenson is writing about. Being more open to the development of ultra-high-speed vehicles and their engines might offer both better alternatives and true advances. But that takes ingenuity and a willingness to risk that NASA doesn't seem to possess right now. Bad climate to propose trillion-dollar space programs, though we've been willing to propose trillion-dollar stimulus packages for more mundane projects, such as propping up failed financial institutions.
Imagine the impact of a trillion-dollar space plane project. Would US students consider a career in engineering if they saw both the opportunity to be part of a cool new future, and the employment options as well? Would this give US aerospace companies something else to sell instead of weapons systems, and is that a good thing? Would it spur international competition, and is that good? Would it divert China's resources into something besides crushing the world's manufacturing competitors? Does that matter? Would a trillion dollars given to this project do more good than giving it to the bankers? Will the bankers also flourish in the glow of this project?
deleting the extra space after periods so i can stay relevant, yeah.
Yes, we've seen it, as well as the refutations pointing out that the lack of rail standardizations proves it a total fabrication.
Do try to keep up.
Suppose you accept his premises that our current state of rocket technology evolved in part due to key improbable events. As a result, we've continued that technology, to "climb to the top of that hill" as he puts it. That doesn't, by itself, automatically mean there must be higher hills to climb. We may have purposefully or accidentally climbed the highest hill we are currently capable of climbing. Perhaps we would have been further along with some other technology if we hadn't climbed this hill, but it might not have been better overall. Right? I mean, it could have turned out like our quest for magnetically confined fusion.
Blind people develop superior hearing to sighted people. I'd still rather have my vision, and I don't think that's entirely due to path dependency.
Same mistake with the combustion engine. Yes, we are getting close to maxing out the technology. But it's not clear that, if we had not developed it in the first place, we would have come up with something more effective in its place. It's not even clear we would have come up with something *as* effective. It's not even clear we even have anything plausible *yet* that would be as effective.
The fundamental mistake in this article seems to be an assumption that the grass is greener in the counterfactual, but he presents no evidence to persuade us that this is actually true.
If you don't know where you are going, you will wind up somewhere else.
May I introduce you to the International Space Station - an artificial satellite of a mass greatly exceeding capabilities of any launcher (and before the inevitable: no, it's not simply a fiction of rocket limitations - we build even ocean going ships in segments nowadays; modularization and, eventually, mass production, is simply a very good idea)
And FYI, the new toy of USAF, X-37, is launched by "dumb rocket" (with Russian main engine...); it's a "spaceplane" mostly because of its envisioned niche usage scenarios, so it can afford wasting most of its mass for airframe.
One that hath name thou can not otter
I think if he looks a bit more deeply it has very little to do with lock in and everything to do with the fact all the wonderful SF ideas out there simply can't be built with our current level of technology.
"Seven Deadly Sins? I thought it was a to-do list!"
That doesn't, by itself, automatically mean there must be higher hills to climb. We may have purposefully or accidentally climbed the highest hill we are currently capable of climbing. Perhaps we would have been further along with some other technology if we hadn't climbed this hill, but it might not have been better overall.
It might be actually slightly the other way around - did we already forget the absolute dominance of "spaceplanes" in scifi of 30s, 40s or 50s?! (even design attempts - Silbervogel, or early winged visions of von Braun) Flying saucers even, at some point...
... and we know how that ended, it didn't deliver on any of its main points as advertised; not a lot of flying boats around, too)
No doubt fueled by rapid advances in aircraft technology at the time. What almost everybody wished for. And we still do, it's easy to remember and relate common experiences of air travel, while forgetting how it's "supposed to" look like (airplanes from "our" times as envisioned ~130 years ago, no doubt influenced by rapid advanced in marine technology), when approached in the same style as "spaceplanes" (actually, I wonder how much the Shuttle was influenced by designers and decision-makers growing on spaceplane scifi
One that hath name thou can not otter
Either way it's on the level of "human tools discovered to be close to their creators in magnitude of size!" Well duh...
One that hath name thou can not otter
I think the article is ignoring some basic physics that has driven us to these outcomes, both his rocket and his oil dependency example.
To get anything into orbit needs a very good weight/energy ratio. The only thing that can provide this itself are your typical rocket fuels. There's two other options:
- Atomic: this usually goes out the window when you consider manned vehicules due to the weight of shielding, and for unmanned vehicules the environmental effects.
- Cheat by leaving a significant part of your mechanism on the ground. Space cannons, magnetic rails and the like. The problem here might indeed be one of technology. even a very fast car (Thrust SC2), might go about at the speed of sound. Sounds pretty fast? It's still nowhere near enough what you'd need. The escape velocity is about 11 km/s, the speed of sound about 300 m/s. Now we need to think in energy, so we need to use E = 1/2 mV^2. Or in other words we need to compare the square of the velocities. 300^2/11000^2 = 0.00074 or about 0.075% of the energy required.
Going much faster and the friction with the atmophere melt your vehicule.
So to get anywhere with a space cannon type system, it needs to be on a very high platform, probably 10km or more, and then be big enough to accellerate a payload to 10-20 times the speed of sound.
When you look at the basic physics, you very quickly end up with rocket-like devices.
A similar thing holds true for our dependency on oil. It again boils down to weight/energy ratio, but with much bigger safety, usability and logistics constraints.
The math is not as straigthforeward, as it's mostly economics, but only rocket fuels give much more power to weight ratio then the conventional fossile fuels.
RogerWilco the Adventurous Janitor
Ironically none of your examples are correct.
- The heat shield tiles were originally developed for ICBM warheads (using composites technology developed in other fields)
- Velcro was a commerical invention having nothing to do with the space program
- Hydrogen fuel cells were invented in 1838. The ones used by NASA were invented by a commercial company, G.E., in 1959.
So it really is more of a case that increasing technological development enabled the space program, not the other way around.
If you don't know where you are going, you will wind up somewhere else.
I like how one of the things Stephenson blames in his article for the rocket lock-in is, "engineering culture," that is resistant to change. I often find that nontechnical folk (and no, sci-fi writing does not count as a technical pursuit) use terms like, "engineering culture," or "scientific elitism," to describe phenomena brought about by actual technical details. In other words, that engineering culture doesn't develop simply because we engineers are resistant to change. It develops because we engineers crunch the numbers and have to deal with reality.
Anyone who thinks that engineers working in the space launch industry are resistant to change just for the shits is pretty misinformed. When it comes right down to it, we're the ones who would love to find a new Pandora's box technology that could get us into space faster, cheaper, and safer. Hell, we have devoted our lives to pursuing the development of the space industry. If anyone wants to see men and women living on Mars, manufacturing in orbit, and fucking onboard inter-galactic colony ships, it's us. Unfortunately, we don't have the luxury that sci-fi authors have of writing about some great new idea and just assuming it will work. We have to test material strengths. We have to plot thermal loads. We have to damp harmonic oscillations. We have to produce enough energy to overcome gravity. Those aren't trivial tasks. And we don't get to defy the laws of thermodynamics and gravity with some hand-wavy bullshit about, "couldn't this idea totally work in theory?!"
So yeah, there are lots of proposed theories and ideas on how to get to orbit. Great, congratulations Mr. Stephenson, you have an imagination. And, awesome, you can see sunnier hilltops across the valley that reach higher than the one we are standing on now. That's a great fantasy land. I hope you enjoy living in it. But while you draft up clever metaphors based on cherry-picked "facts" and unrealistic assumptions, those of us working in the industry, you know, the ones doing the math, actually have to look at the numbers. And those distant, high hilltops you see, well they might not be as high as you think. And all those, "innovations," on how to get to space, well they might not be as Earth-shatteringly ingenious as you think.
I'm not saying there's not room for improvement, there definitely is. But until someone shows me some numbers that prove a space-elevator, a launch loop, or a space fountain can be built, today, without unobtainium (in the form of some material, or some epic power source), I am going to delegate those ideas strictly to fantasy-land for now. And as for things like space planes, hypersonics, multi-propulsion-type vehicles, and so on, we are trying them, to an extent. And, believe it or not, just like rockets, they are still fucking difficult to get right. That's why it takes a long time to develop them. In the end, chucking something out of our gravity well is no easy task, no matter what method you take. And it is expensive, in both time and energy, no matter what technology you utilize. So stop lamenting about how poor off we are compared to where we could be. We're doing everything we can with what we've got. If that's not good enough for you, vote to give us more money or design a small, portable power-plant that can produce a proper metric fuckton of thrust.
In the end, engineering culture is just a term being used to say, "technical shit that I don't understand well enough so I'lll use it as a scapegoat to justify my preconceived notions"
Motorcycles, Robots, Space Gossip and More!
Check put Pegasus rocket - it does to a large degree what you want. And is one of the most expensive, per kg, launchers around.
The general problem is how "enthusiasts" forget about physics, about rocket equation, about how majority of the acceleration must happen outside the atmosphere, how there's a square attached to speed in kinetic energy (which comes from the energy of propellant). Read about HOTOL or Skylon, too. When rigorously looked at, ending not better (in best case scenario!) than a normal rocket using similar materials (which we don't have yet, and which are required for a spaceplane to even barely work - all those fancy, complicated flight sequences accomplishing... to lift structures... which are necessary for said sequences)
One that hath name thou can not otter
That's definitely a possible method to achieve orbit. However, you need to realize that the big hurdle in getting to orbit is not altitude, it's velocity. Yes, you can hoist a big ass rocket up to 100,000 feet and launch it from the back of a hypersonic plane. It turns out, however, that you still need a really big ass rocket (I know, lot's of technical details there) to achieve orbital velocity no matter how high you launch from. That said, it tends to be a lot cheaper and easier to launch a big ass rocket from a stationary, land based platform, than it is to launch it from a flying, attitude-subject-to-weather-conditions plane.
So is it possible? Yes. Are we trying it? Yes, some companies/organizations are. Will it be the final answer to space access? Well, maybe, but it will probably prove to be just as complicated and expensive as rocket launches. Then again, maybe not. We'll find out in another couple decades when the technology comes to fruition.
Motorcycles, Robots, Space Gossip and More!
Yeah, I was really disappointed by this from Neil. It's extremely poorly done.
1) "Without the Nazis, rockets wouldn't have happened" -- the Nazis merely accelerated something that was already ongoing. All major nations were working on rocketry. There were two primary purposes: sounding rockets, and aircraft. This was, you'll recall, before we knew that jet engines would win out over rockets for airplane propulsion, and all sides were working on rocket-propelled craft. Even if rocket-plane propulsion were to stop, sounding rocket development would have continued to advance to V2-scale. WWII just accelerated things.
2) "A-bombs were too expensive and militarily ineffective" --really? Taking out an entire city and its mass production capability isn't worth the cost to purify some uranium? Perhaps if you divide the number of bombs dropped on Japan by how much we spent on the Manhattan project, maybe, but most of that was a sunk cost. The world was terrified of atomic bombings.
3) "Without A-bombs, rocket development would have ceased." -- ignoring the issues in #1, after WWII, rocketry had already captured the public mind. In fact, even during WWII, Von Braun had already been talking up, and getting military interest in, orbital space bombers that would stay in orbit and drop their (conventional) payloads on enemy targets at a moment's notice (plus taking spy photographs, etc, all without risk of being shot down) during WWII.
4) "All payloads are sized to be like A-bombs" -- not in the least. There's a huge range of payload profiles and lift capabilities of modern rockets. Just because the stacks were originally designed for a specific load doesn't mean that all of their descendants are.
Probably the most disappointing line, however, was:
5) "Rockets are as close to perfect as they're ever going to get." Oh really? Scramjets? Nuclear thermal? Strained-bond chemicals? Cryogenic solids and hybrids? The dramatic materials enhancements we're starting ti get (which has a profound effect on rocket performance)? Advanced heat shields? And on and on. Plus, just ignoring radical changes, look at how much of a difference design approaches have toward launch costs -- compare the Space Shuttle to SpaceX, for example. Rockets are nowhere close to being completely optimized.
There's a lot of legacy that could be criticized with the space industry, esp. the government space industry. Nobody would insist on keeping on reusing as many shuttle components as possible for a next-gen stack if it wasn't all the jobs on the line. Even the "radical", ground-up redesigns, such as SpaceX's Falcon, still uses some legacy parts. So there is a lot of legacy stuff to criticize. But Neal only skimmed over these things :P And he skipped the most important part of such an article: proposing alternatives. So you don't like rockets -- fine. Let's talk alternatives. What do you like -- skyhooks, space elevators, launch loops, ballistic launch, what?
I guess I just let my urge to spawn soldiers for Satan's dark army get the better of me.
Everyone in the rocket business thinks of the alternatives all the time.
And then looks at the laws of physics, and the laws of economics, and goes for the solution that gets the job done with minimal waste and effort.
This isn't to say there's no waste or extraneous effort, but the main theme of the project isn't based on a fantastic boondoggle.
And if there's more than one way to skin a cat, it will get tried eventually as someone realizes they do have the resources to attempt it.
But while it may work for a niche, eventually you come back to the science of rocketry and the equations of motion and you decide that your rocket is going to look and act like a lot of others before it.
Just another example, regarding the "globalization effect" he's talking about: Thorium based reactors that the US doesn't give a rat's ass about, but may well play a major part in solving the energy crysis and how China is thinking about building them.
This is the crux of it, I think:
I don't agree that Hitler choice of rocketry for the V2 was random. I think he went to his not-yet-rocket scientists and said, "How do I deliver X kilograms of payload to England with such and such circular error probability, and oh yeah, it can't be intercepted?" And rockets were the answer. And for good reasons. I'm not sure what other technologies of the 1930s and 1940s could have performed the task: submarines with huge artillery built-in (susceptible to torpedo planes unless you could do some kind of shoot and scoot); they did try the bomber thing but that wasn't a winner; balloons don't seem like a possibility. We *still* don't have something better than rockets and missiles for mass producing corpses (whether you agree its a good idea or not): perhaps the Navy's upcoming railguns are different enough to be considered a "change".
It was an interesting article, but there were a couple of parts that I thought were really weak. One problem area:
What? Surely step 2 was more or less a direct result of step 1 - there's nothing improbable about that at all. I omitted the rest of the list, but the sequence didn't seem all that improbable to me. In fact, I really can't even figure out how the concept of probability even applies to historical events. They're all unique.
Another problem:
*crickets*
Ok, Neal, care to explain to us what these radically innovative schemes might be? In addition to the parent's theory that institutional inertia is the reason (which is probably a big part of the answer), there's also the very real possibility that none of these other "radically innovative" schemes have any chance of competing with rockets on the basis of cost-effectiveness - because all the R&D on rockets is essentially done and paid for, whereas to "cross those valleys" would be simply enormously expensive. And there's no guarantee your radical new technology would work. I think the risk factor here is even more important than the problems of inertia.
Come now, do you really expect me to do coordinate substitution in my head while strapped to a centrifuge?
The reality is, of course, that there aren't really any significant forces acting on you at all. You're travelling in a straight line when in orbit. It's space that is bent.
I guess I just let my urge to spawn soldiers for Satan's dark army get the better of me.