http://www.slate.com/id/2283469/pagenum/all/

It's a long article, but the most relevant stuff is at the end, about how we're pretty much locked into the existing rocket technologies:

That is not, however, the most important way that rockets generate lock-in. In order to understand this, it's necessary to know a few things about (1) the physical environment of rocket launches, (2) the economics of the industry, and (3) the way it is regulated, or, to be more precise, the way it interacts with government.

1. The designer of a rocket payload, such as a communications satellite, has much more to worry about than merely limiting the payload to a given size, shape, and weight. The payload must be designed to survive the launch and the transition through various atmospheric regimes into outer space. As we all know from watching astronauts on movies and TV, there will be acceleration forces, relatively modest at the beginning, but building to much higher values as fuel is burned and the rocket becomes lighter relative to its thrust. At some moments, during stage separation, the acceleration may even reverse direction for a few moments as one set of engines stops supplying thrust and atmospheric resistance slows the vehicle down. Rockets produce intense vibration over a wide range of frequencies; at the upper end of that range we would identify this as noise (noise loud enough to cause physical destruction of delicate objects), at the lower range, violent shaking. Explosive bolts send violent shocks through the vehicle's structure. During the passage through the ionosphere, the air itself becomes conductive and can short out electrical gear. Enclosed spaces must be vented so that pressure doesn't build up in them as the vehicle passes into vacuum. Once the satellite has reached orbit, sharp and intense variations in temperature as it passes in and out of the earth's shadow can cause problems if not anticipated in the engineering design. Some of these hazards are common to all things that go into space, but many are unique to rockets.

2. If satellites and launches were cheap, a more easygoing attitude toward their design and construction might prevail. But in general they are, pound for pound, among the most expensive objects ever made even before millions of dollars are spent launching them into orbit. Relatively mass-produced satellites, such as those in the Iridium and Orbcomm constellations, cost on the order of $10,000/lb. The communications birds in geostationary orbit—the ones used for satellite television, e.g.—are two to five times as expensive, and ambitious scientific/defense payloads are often $100,000 per pound. Comsats can only be packed so close together in orbit, which means that there is a limited number of available slots—this makes their owners want to pack as much capability as possible into each bird, helping jack up the cost. Once they are up in orbit, comsats generate huge amounts of cash for their owners, which means that any delays in launching them are terribly expensive. Rockets of the old school aren't perfect—they have their share of failures—but they have enough of a track record that it's possible to buy launch insurance. The importance of this fact cannot be overestimated. Every space entrepreneur who dreams of constructing a better mousetrap sooner or later crunches into the sickening realization that, even if the new invention achieved perfect technical success, it would fail as a business proposition simply because the customers wouldn't be able to purchase launch insurance.

3. Rockets—at least, the kinds that are destined for orbit, which is what we are talking about here—don't go straight up into the air. They mostly go horizontally, since their purpose is to generate horizontal velocities so high that centrifugal force counteracts gravity. The initial launch is vertical because the thing needs to get off the pad and out of the dense lower atmosphere, but shortly afterwards it bends its trajectory sharply downrange and begins to accelerate nearly horizontally. Consequently, all rockets destined for orbit will pass over large swathes of the earth's surface during the 10 minutes or so that their engines are burning. This produces regulatory and legal complications that go deep into the realm of the absurd. Existing rockets, and the launch pads around which they have been designed, have been grandfathered in. Space entrepreneurs must either find a way to negotiate the legal minefield from scratch or else pay high fees to use the existing facilities. While some of these regulatory complications can be reduced by going outside of the developed world, this introduces a whole new set of complications since space technology is regulated as armaments, and this imposes strict limits on the ways in which American rocket scientists can collaborate with foreigners. Moreover, the rocket industry's status as a colossal government-funded program with seemingly eternal lifespan has led to a situation in which its myriad contractors and suppliers are distributed over the largest possible number of congressional districts. Anyone who has witnessed Congress in action can well imagine the consequences of giving it control over a difficult scientific and technological program.

Dr. Jordin Kare, a physicist and space launch expert to whom I am indebted for some of the details mentioned above, visualizes the result as a triangular feedback loop joining big expensive launch systems; complex, expensive, long-life satellites; and few launch opportunities. To this could be added any number of cultural factors (the engineers populating the aerospace industry are heavily invested in the current way of doing things); the insurance and regulatory factors mentioned above; market inelasticity (cutting launch cost in half wouldn't make much of a difference); and even accounting practices (how do you amortize the nonrecoverable expenses of an innovative program over a sufficiently large number of future launches?).

To employ a commonly used metaphor, our current proficiency in rocket-building is the result of a hill-climbing approach; we started at one place on the technological landscape—which must be considered a random pick, given that it was chosen for dubious reasons by a maniac—and climbed the hill from there, looking for small steps that could be taken to increase the size and efficiency of the device. Sixty years and a couple of trillion dollars later, we have reached a place that is infinitesimally close to the top of that hill. Rockets are as close to perfect as they're ever going to get. For a few more billion dollars we might be able to achieve a microscopic improvement in efficiency or reliability, but to make any game-changing improvements is not merely expensive; it's a physical impossibility.

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.

A temptingly simple explanation is that we are decadent and tired. But none of the bright young up-and-coming economies seem to be interested in anything besides aping what the United States and the USSR did years ago. We may, in other words, need to look beyond strictly U.S.-centric explanations for such failures of imagination and initiative. It might simply be that there is something in the nature of modern global capitalism that is holding us back. Which might be a good thing, if it's an alternative to the crazy schemes of vicious dictators. Admittedly, there are many who feel a deep antipathy for expenditure of money and brainpower on space travel when, as they never tire of reminding us, there are so many problems to be solved on earth. So if space launch were the only area in which this phenomenon was observable, it would be of concern only to space enthusiasts. But the endless BP oil spill of 2010 highlighted any number of ways in which the phenomena of path dependency and lock-in have trapped our energy industry on a hilltop from which we can gaze longingly across not-so-deep valleys to much higher and sunnier peaks in the not-so-great distance. Those are places we need to go if we are not to end up as the Ottoman Empire of the 21st century, and yet in spite of all of the lip service that is paid to innovation in such areas, it frequently seems as though we are trapped in a collective stasis. As described above, regulation is only one culprit; at least equal blame may be placed on engineering and management culture, insurance, Congress, and even accounting practices. But those who do concern themselves with the formal regulation of "technology" might wish to worry less about possible negative effects of innovation and more about the damage being done to our environment and our prosperity by the mid-20th-century technologies that no sane and responsible person would propose today, but in which we remain trapped by mysterious and ineffable forces.

New Comment
20 comments, sorted by Click to highlight new comments since:

The subtext here is almost more interesting than the article itself. Within the last few years, Stephenson has worked as a consultant for Blue Origin, a commercial spaceflight company that's made very little public about its plans and development; Wikipedia claims it's working on a single-stage vehicle similar to the DC-X but that's as much information as I've seen anywhere. It doesn't seem like too much of a stretch to suggest that that experience informed the opinions we see here.

We probably don't have enough information to say how, but there's enough odd specificity in those opinions that I'm pretty sure there's more going on than shows on the page. I'd love to know exactly what it is.

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.

This is plainly not true. I know of only three alternative schemes that might be viable, and when you look at what they are and what their caveats are, it becomes obvious why we're stuck on chemical rockets.

Then there's space elevators, which would require a major breakthrough in materials science, and might still be impossible even then.

Then there's the Orion spacecraft design - that is, an extremely large ship propelled by launching nuclear bombs out the back to ablate a shield on a spring. In addition to the political problems, it's also extremely inflexible (it has to be huge or it doesn't work), hard to iterate on (for the same reason), and environmentally damaging (as in "would cause human deaths", not to be confused with the more common "would anger irrational activists").

Finally, there's the space gun idea (as envisioned by Quicklaunch): a wide, neutrally buoyant gun barrel extending from the ocean surface to the sea floor. Unfortunately, the calculations I've seen indicate that the acceleration will be too high for humans, and it still needs a second stage rocket (for fine-tuning its orbit, if nothing else). So, even if this were built, we'd still need rockets for launching humans, and possibly also for launching equipment too large for the barrel diameter. I think this one should be pursued, but it's no historical accident that rockets came first.

Two other technologies that might play a role are Airship to Orbit from J.P. Aerospace, and the space tethers ideas of Robert Forward's company.

I agree, though, that unless ATO works out, chemical rockets are probably just about the best and safest technology we are likely to find for launch from earth.

But if Less Wrong ever undertakes some kind of collective brainpower exercise as an experiment, I'd love to see whether the smart people here can figure out how or whether the ATO idea might work. There is some weird stuff that happens when the atmosphere gets thin enough so that the mean free path becomes comparable to the scale height and the airship dimensions become comparable to both.

Wikipedia link with more ideas: Non-rocket spacelaunch. I like the idea of building a tower of balloons :-)

we started at one place on the technological landscape—which must be considered a random pick, given that it was chosen for dubious reasons by a maniac

Now that sounds intriguing! Dubious reasons by a maniac? Can you tell me more?

Godwin's law violation coming up:

He means Hitler putting R&D muscle behind V-2's even though he claims they weren't worth it economically. Do read the article, it's good.

It's interesting, yes, but I'm not sure rockets are all that much of a 'local' optimum, as it were. What plausible alternatives are there? The SpaceX and other recent changes seem pretty standard.

(Railguns? Seems like a hard argument since 60 years after rockets began putting things in orbit, I still haven't heard of a railgun anywhere near orbital capability. A staged ballistic space gun is more plausible but if Stephenson is complaining about rockets' accelerations...!)

The idea of space elevators have been around for a long time. The technology isn't completely available at present, but (as far as I know) they aren't too far off from reality. If there was some more effort and research put into that program, it could become a viable option fairly soon.

The concept of tensile* space elevators only date back to 1959, a decade before rockets put men on the moon. And they seem very far off to me. Do we even have inches of the necessary construction material? As far as I know, we don't. Fairly soon? We seem about as far away, R&D-wise, as Hero of Alexander from the steam engine & Industrial Revolution.

* I am aware that Wikipedia dates it to Tsiolkovsky in 1895. If that's a space elevator, I humbly suggest that the true date of the space elevator concept be pushed back by around 3000 years to the Tower of Babel.

You are correct in saying that the technology isn't here yet. I do think, though, that the Hero of Alexander claim is a bit hyperbolic. I would be surprised if we had inches of the necessary construction material, but I think part of the reason why it seems so far away is that there isn't a major, concerted effort to do it yet. I'd say it sounds about as far off as the proposal to go to the moon did, before the US had even achieved earth orbit. Or perhaps, as far fetched as the theoretical scheme that the matter in the nucleus of atoms could be converted into energy, creating an incredibly powerful explosive did, before there was a major push for that. A space elevator is a theoretical idea at present, but when there is the funding and the effort behind a technological development, it can happen faster than we typically think. I'm definitely not expecting a space elevator within the decade. But I'd be surprised if it wasn't possible in my lifetime.

I'd say it sounds about as far off as the proposal to go to the moon did, before the US had even achieved earth orbit. Or perhaps, as far fetched as the theoretical scheme that the matter in the nucleus of atoms could be converted into energy, creating an incredibly powerful explosive did, before there was a major push for that.

Both had, a decade or three beforehand, the basic technology proven. What is the difference in kind, and not degree, between putting a man in orbit and putting him onto the moon? Once you've gotten all the way to a reactor pile which can go critical, you've done most of the hard work.

If we could manufacture a few meters of space elevator material, that'd be one thing, and I might accept an argument that 'with a Manhattan project equivalent, we could build a space elevator in a decade or two'. If you can manufacture a few meters, then you can do it again and again and scale your processes up. But we can't even manufacture inches, putting us closer to the Curies or Roentgens of space elevator than the Fermis or Oppenheimers.

If we could manufacture a few meters of space elevator material, that'd be one thing, and I might accept an argument that 'with a Manhattan project equivalent, we could build a space elevator in a decade or two'.

I think you are looking at the wrong problem. Assume that you can easily turn coal into bucktube material suitable for building an elevator. Now, compute how many tons of the stuff you will need. And then, how many tons of chemical rocket fuel will be required to lift all that material up to GEO.

Oh, we may build an elevator some day. But I doubt that the material for building it will come from the Earth's surface.

Assuming von Neumann machines doesn't do much to strengthen the argument 'we could build a space elevator relatively soon if we really wanted to'. If anything, it weakens it...

By "von Neumann machines", I usually understand stored program computers. You are apparently talking about some kind of self-reproducing (nanotech?) robots. Assuming that such things exist doesn't change the rocket fuel requirements for building an elevator, but they might help to build the rocket-fuel refineries. So, I don't see how this assumption weakens the argument.

Oh, I assumed your last comment meant that the material would be coming from the moon and/or asteroid belt, and usually people aren't proposing sending humans out there to mine them but von Neumann machines.

Ok, so we send a pair of robots to an asteroid and let nature take its course ...

And then a few generations later we have thousands of robots heading back to earth to build an elevator for us. Yeah, that might work. And it might be cheap. But it probably won't be particularly quick. Maybe 40 - 100 years from first arrival of robots at asteroid, I'd guess. I still don't see how the argument is weakened by the existence of robots, but I agree it is left pretty weak.

I still don't see how the argument is weakened by the existence of robots, but I agree it is left pretty weak.

No, it's weakened by a variant of the conjunction fallacy, as it were. If you previously argued 'A ~> C' but have now changed your argument to 'A & B ~> C', then probablistically C has gotten less likely.

So one originally starts off arguing 'we may have elevators soon, since when we can create miles of nanotubes, then we can create space elevators quickly', and changes it to 'we may have elevators soon, since when we can create miles of nanotubes and we have also finally developed space robots to go synthesize it in orbit for us, then we can can create space elevators quickly'.

You have narrowed the possible routes to creating a space elevator by ruling out routes that don't involve von Neumann machines; that ought to reduce our probability.

Ah! I've got it now. The assumption that bots are available doesn't weaken the case for an early elevator. The assumption that bots are necessary does weaken the case.

I don't know why it took me so long to pick up on that. Sorry.

No problem. I wasn't sure I was being fair in inferring that the bots were necessary. If they aren't necessary, then by the same exact logic, our probability ought to go up - 'A v B ~> C' is stronger than 'A ~> C'. (The more independent pathways to a result, the more likely one will work within a certain time span.)