This is a pretty thought provoking article. I've always really enjoyed Neal Stephenson's odd perspective on things -- in fact, I always want to smack people who complain about his stories' content and ending situations, and tell them "That's not the point, read the descriptions of things dammit.". The thing I really enjoyed was the bit about "the catch is it has to be the size and shape of a hydrogen bomb".
Anyway that aside, it really brings up the old infrastructure problem -- Large investment in infrastructure is a two-edged sword, bringing both benefit and lock in, and when it comes time to change there is lots of debate. I think Neal either misses or avoids a big part of the argument here -- infrastructure is turned into the bad-guy and the good guy. People don't see the benefits it has brought for a myriad of reasons - they have internalized it, they are not in the class of society that directly gets money for it (and the general improved life isn't apparent because their neighbors are in the same boat), they are afraid of tax increases (or lack of tax cuts), they don't understand the current tech, and figure that "it has problems so anything else we do won't fix it either", it has broken society in the following ways...
Really all of this though is just setup. The way I see it, large scale tech and infrastructure projects are hard to get at and harder to revamp because they are just too easy to target with populist attacks from all sides. The issue is usually complex, but easy to attack with a simple disingenuous quip. Doubly so when the alternative is something that sort-of works, because then the quip doesn't even need the effort of disingenuety, just a mean spirited "they are trying to change the perfectly good stuff we have just to take it away from us, and ignoring everything else to fix"
I have no idea of the solutions we could offer to these types of scenarios, but I do think that somehow we need to find a way to look at these piles of infrastructure we have an find ways to make them better. To do that we need to get around the "infrastructure problem".
Dr. Jordin Kare has an interesting proposal for cheap space launch. Use lasers to remotely energize (zap a heat exchanger on) lightweight craft carrying hydrogen. The hydrogen wouldn't burn. Instead, it would just act as a reaction mass of very low weight, thus producing very high exhaust velocities. Basically, this would give you the exhaust velocity (read: fuel efficiency) of the upper theoretical end of nuclear rockets, but without the heavy nuclear equipment onboard, making the craft even lighter and more efficient.
But that's not even the clever bit. The clever bit of Jordin's proposal is that the laser tracking/energizing system can be built modularly. You can build one prototype module that can launch one toy craft. Then you figure out how to mass produce it and build a whole bunch of these puppies that can lock onto and zap a much larger heat-exchanger carrying craft.
What you get is very cheap access to orbit without ungodly huge R&D and infrastructure costs up front.
Actually the systems which will allow to target the "heat exchanger" are not that easy to build and only recently we have seen big lasers mounted on planes which can barely hit a rocket.
So it's not that easy and anyway it will be weapon technology.
Actually the systems which will allow to target the "heat exchanger" are not that easy to build and only recently we have seen big lasers mounted on planes which can barely hit a rocket. So it's not that easy
I never said it would be trivial. There are two big differences, though. 1) The laser-launch craft wants to be hit. We can mount retro-reflectors on it, and it can transmit back helpful data. 2) The laser is going to be stationary, and the vehicle is going to be traveling on a known track.
and anyway it will be weapon technology.
This makes it much more likely that it will get funded and developed.
It's not just populist attack, it's attack by powerful entrenched interests. For the most part, those interests are employees of the current system.
It's often said that to lock in a program, you need to create a class of people who might lose their jobs if it is shut down. These people are highly motivated and willing to go to great lengths to protect that program. In contrast, people afraid of tax increases or people who do or don't observe the benefits have little motivation to do anything.
The people afraid of losing their jobs protect that program from ever being cut, but they also prevent any radical improvements that might make them obsolete.
Corporations have the same challenge, some times called the "innovator's dilemma" or "cash cow disease". For example, Microsoft struggled when the web (and thus IE) threatened Windows' dominance.
People tend to be most skeptical of infrastructure projects funded with tax dollars coercively taken from them. It takes a lot of bad publicity to begin a populist movement against private transactions.
I tend to oppose many government infrastructure efforts because the government has no incentive to do a realistic cost/benefit analysis of new projects. It is always in the legislator's and lobbyist's interest to start a new project, whether it be massive high speed rail lines, expensive green power installations, or large subsidies for classes of fuel and transportation that are uneconomical in the free market (hydrogen cars, corn ethanol). The only people paying a cost for new projects don't have a direct voice in the planning meetings.
The interstate highway system of the United States has long been the shining example of a "good" infrastructure project. But it has certainly subsidized the development of sprawling suburbs and a car-centric culture. Nowadays many expensive government infrastructure proposals (such as passenger rail) are aimed at mitigating the externalities from the earlier infrastructure projects. I wish we took a little more care at spending money wisely.
When you describe rail and green energy projects as "uneconomical in the free market", are you intending to compare these projects to existing options for energy and transit like coal and cars+highways+oil?
The light-gas gun launch + rocket hybrid is a very interesting concept. It consists of a large, but conventional light-gas gun propelling a g-hardened payload 2/3 of the way to orbital velocity, then the payload uses its own engines to accelerate the rest of the way. The scheme is advertised to cost around $400M, but we should consider that within an order of magnitude. Even at that, the opportinity to significantly lower launch costs should be embraced.
Halving launch costs would be significant, it would save the satellite industry many billions. Note that half orbital speed discount is much more than 50% in launch energy savings.
I'm surprised Stephenson didn't touch on SpaceX's potential as a game changer. The biggest "next" step to increased space access, is a vastly cheaper, more efficient, and potentially more reliable orbital rocket. And, SpaceX is doing exactly that!
As Musk is so fond of saying, the existing options are the Lamborghini's of launch vehicles, whereas he's trying to build the Honda. Safe, reliable, and cheap!
I have a lot of hope that we will have a new Space Renaissance in our lifetimes and I think the work that SpaceX, and other companies is doing will be instrumental in getting us to Mars and beyond.
I agree with you and disagree with Stephenson. From TFA: "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."
From what I understand of the rocket business, this is completely wrong.
Current rockets designed by the established players (LockMart, etc.) are optimised for performance, not overall cost. This comes from a variety of factors, starting with the basic philosophy of the designers, to the rocket fuels chosen, all the way up the design hierarchy.
I estimate that it is possible to reduce launch costs by at least 10x, and that is using 1960's rocket technology with modern computers and sensors. The key is looking at what is going to reduce overall operational cost. I don't have room to explain everything, but suffice it to say I'm a fan of big dumb boosters.
There are many, many political factors at work that have slowed down the cost reductions which we would see if this was some other industry. One of the biggest factors is that the major governments don't really want launchers to become cheaper. Again, it gets back to the space race, and the governments don't want just anyone to have ICBM capability. The big companies also don't want competition for their cash cows, so they lobby hard to make sure space access remains as expensive as possible.
> I estimate that it is possible to reduce launch costs by at least 10x, and that is using 1960's rocket technology with modern computers and sensors.
I couldn't agree more. They are already at least 3x cheaper based on my very rough calculations and wikipedia:
- Falcon 9: $56million
- Delta IV: $150million
- Atlas V: $187million
And they are just at the very beginning of recovering their research costs and systematizing the process of rocket building. I am amazed at what they have been able to accomplish thus far.
I think he touches on it in a very generally with this statement.
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.
Sure SpaceX is the first (almost) successful commercial venture of its kind and has been a long time coming. But they're still using essentially the same old ICBM technology the USA and USSR have been using for the past 60 years.
Stephenson is advocating more energy, research and money be devoted to the technologies on the horizon that would replace giant explosive tin cans.
Sealaunch got around some of the problems - such as a launch site but then got stopped by the same vested interests.
California (coincidentally home to all the current rocket makers) first banned the launchers from docking in Californian ports claiming they were weapons - then banned loading the satelites in foreign ports claiming they were exporting secret technology.
Well, SpaceX is still playing the "expendable rocket" game, at least for the moment. They're trying to recover and reuse first stages, but they say that they're set up to be profitable even if that doesn't work --- and while their prices are cheaper than others, it's not enough yet to change the way you'd build a satellite.
Besides, Stephenson at least used to have a consulting relationship of some kind with Blue Origin, the rival launch-system startup funded by Jeff Bezos, which is still awfully quiet about their vehicle, at least for the moment. (I was wondering if this was a preview, but even if it was, he'd be highly unlikely to say so.) If the relationship continues, or even if he just holds stock, that might be reason enough not to talk up the competition.
'Well, SpaceX is still playing the "expendable rocket" game, at least for the moment.'
As much as I am an enormous fan of non-chemical-rocket approaches into space (as hartror linked), they all fall into one of two categories: "Not proved feasible" (space elevator and all variations thereof), and "Contains the word 'nuclear' and not proved feasible". I break that second one out as a separate category on the grounds that making many of these proposals functional seems a lot more likely than the more far-out ones, but are Utterly Impossible because nobody will even fund research into a space technology that contains the word "nuclear", let alone permit it to fly. We barely get RTGs into space, because they are the only solution, and not without ignorant protesters nearly managing to stop it.
For instance, I'm a fan of the nuclear lightbulb [1]; I don't know if can be built or not, and I suspect I will never find out, because nobody within my lifetime will try, despite the fact it may require no actually new physics or technology.
If you're trying to make money in this field, best start with the things we actually know will work. Let us all hope the multi-trillion-dollar-market-cap SpaceX of 2040 see its way to funding some of these proposals, goodness knows our governments aren't going to until the word "nuclear" completely loses its fear factor.
Oh, and if we all die because we're hit by a meteor that could have been stopped with a nuclear lightbulb or even Project Orion, let the record show I hate you all, humanity.
That's sort of me explaining why; the next hill probably has NUKULAR and OMGRadiation!!1! written all over it. We're trying to jump to the next one after that, if indeed it even exists, because of our "nuclear" phobia.
There have been many attempts since the 80s to change the game just that way, repeatedly failing. The article tries to explain why. I no longer follow the field, and don't know about SpaceX -- hopefully it's found the way or the right time to break the jinx.
It's not mentioned in the article, but in addition to his work as a writer, Neal Stephenson once worked as an advisor to Blue Origin, the spaceflight company founded by Jeff Bezos.
Blue Origin is focused on suborbital space flight using rockets. But I remember Stephenson making some vague statements about his work there (at a reading of Quicksilver at the University of Washington) that included research into space elevators and other less-proven technologies.
If you enjoyed this piece, be sure to read "Mother Earth Mother Board", written by Stephenson and published in Wired in December 1996. It's a long and fascinating look at the world of undersea cabling, and it's chock-full of super-interesting facts.
Excellent, I was hoping that this article serves as a kind of precursor or preview of material that Neal is working on. Given that undersea cabling is discussed in some detail in Cryptonomicon (published three years later in 1999) then that means we can probably look forward to interesting space travel discussed in Neal's next tale. Anathem touched on a Project Orion type spacecraft, but this article is definitely covering a different slant.
This article reflects a constant theme in American innovation. Capitalism by itself does not demand innovation, but its militant thirst for resources does. What Stephenson calls lock-in is a product of a system that demands maximal efficiency of resources within an organization, but huge waste across the culture. Think of the empty flatbeds driven across Iraq to inflate cost plus contracts, drug companies competing to produce chemical analog drugs, or three drug stores selling identical products on a single corner.
The modern Capitalist attitude among modern business schools is that research and development is a cost center which must be minimized. This was a major element of Carly Fiorina's plan to cut HP to profitability. However, tremendous resources are thrown into marginal technologies in order to redundantly market and protect them.
Most great innovations in American culture seemed to occur due to a great existential need by our military. The Internet was designed as a communications network to survive nuclear attack. Rockets, as Stephenson pointed out, were improved for ICBMs. Most alternative fuel research is funded by the DoD to provide alternatives for tanks and jets in case our nation gets cut-off from its oil supply.
As someone who has run for office twice, I deeply understand that lobbyists exist largely to make money flow regardless of the suitability of a given contractor or product. I doubt our current military-industrial-congressional complex is independent enough to provide groundbreaking developments when incremental improvements suffice. Military lobbyists have become too powerful and too monetarily influential to the candidate that wins.
Your Wikipedia-based contention does not appear to be relevant. Paul Baran's original research does not match what the Internet actually became. On that point, we agree. I based my contention strictly on the primary source-- Paul Baran's RAND research note. The DARPA research and the RAND note were an investigation into the idea of having a network which would meet military objectives, possibly including communications after a nuclear strike. Much of the note focused on that. Example after example exists of commissioned military research which gets applied to a purpose other than the original subject of the research-- that's exactly the point I was intending to get across with my example. The structure of the Internet was designed to be decentralized in part because Baran envisioned this military need. Whether the RAND team and the DARPA team agreed on the direction of the research seems besides the point.
Ha: "[After all,] the modern petroleum industry is a direct outgrowth of the practice of going out in wooden, wind-driven ships to hunt sperm whales with hand-hurled spears and then boiling their heads to make lamp fuel."
>The above circumstances provide a remarkable example of path dependency. Had these contingencies not obtained, rockets with orbital capability would not have been developed so soon, and when modern societies became interested in launching things into space they might have looked for completely different ways of doing so.
>Before dismissing the above story as an aberration, consider that the modern petroleum industry is a direct outgrowth of the practice of going out in wooden, wind-driven ships to hunt sperm whales with hand-hurled spears and then boiling their heads to make lamp fuel.
Is he seriously citing the adoption of petroleum fuels as an example of path dependency?
It is, we had whale oil which needs handling in barrels and could be burnt in lamps, so we switched to crude oil which was then refined to the same grade as whale oil and could be handled in the same barrels and burnt in the same lamps.
Then we switched to tankers but the oil could still be burnt in the same lamps so no consumer change, then the consumer changed to using it in their cars but we shipped it in the same tankers.
Now we have a system where we can only use a fuel that is delivered in a certain quantity to a local gas station as a liquid that can be burned.
Compare this with the electricity grid where we can switch from coal to nuclear to wind to gas without anyone noticing.
Maybe I'm just ignorant, or blinded by cognitive biases, but I always imagine the combination of gas/diesel and internal combustion engines as an engineering optimum, reached iteratively after decades of wandering through parameter space. At least wikipedia's version [1] of the history of cars seems consistent with this, showing a history of dozens of engine designs and fuel choices (solids, liquids, gases, and electric batteries) long before the ICE came to dominate. (Maybe I'm unreasonably optimistic, but I'm unsurprised we've succeed in breaking free of our original "path", steam engines). Perhaps it's my lack of imagination, but I'm not aware of any practical fuel (chemical or otherwise) superior, in energy density and convenience, to liquid hydrocarbons. Nor am I aware that there was a historically superior way of obtaining liquid hydrocarbons to letting them gush out of the ground, which I understood was the least expensive source for them throughout modern history.
Am I simply rationalizing the history of modern transportation, and if so can you enlighten me as to the "paths not taken" that I'm blind to?
If oil hadn't been available do you think Karl Benz would have sat down and invented the entire drilling, refining and transport system before he started on the car?
The ICE was possible because oil infrastructure was there - the first gas stations were the pharmacists who already sold the oil
Nobody is claiming anything about the internal combustion engine. Uvdiv is claiming that oil is not an example of path dependence because it is actually optimal. That there exists a path to get to the ICE does not refute the argument.
I'm with uvdiv here: the idea that the modern oil industry exists due to path dependence ignores the massive fact that petroleum was by far the most concentrated, easily transported exploitable form of energy on the planet in the 19th century.
TL;DR version (though you should just read it): The space industry is a great example of path dependency and lock-in in innovation. The reasons why we use rockets to launch satellites are historical dating back to the days of Hitler and the H-Bomb. After trillions spent on developing ICBMs capable of crossing the world, our governments are 'locked-in' to using rockets to get things into space. Using rockets for this purpose is not nearly as efficient as other methods, but we have perfected that practice to the point of perfection. In order to increase space accessibility we need to "cross-the-valley" to another technology, but since it has taken so much money (path-dependent) we are locked-in, and it will be very hard to innovate in this space.
Not quite - we use very large expensive rockets to launch very large expensive payloads.
The rockets are large and expensive because the payloads are large and expensive - which they need to be because launches are large and expensive.
Imagine if Nasa had been in charge of developing aviation in the US - you certainly wouldn't have Cessnas and 737s.
There is a brilliant parody about Nasa being in charge of the old wagon trains going into the weat and their plans to make one giant wagon that would be able to make the journey and return safely.
Reading this article raises a question that I ask as devil's advocate (mostly): why do we need to get into space?
I can think of three reasons:
1) We need to put up comsats.
2) Military superiority
3) Because we can and it's cool.
To the first, as the article stated the sky's already getting a bit full as there are only so many comsat "slots"
To the second, SDI never really worked and that threat doesn't exist anymore.
To the third, well...is it worth the brajillions? I personally think it is because we don't know what's out there, but that's a really poor sales pitch. :)
Perhaps it's my limited imagination and understanding, but I'm unable to conjure the reward that offsets the risks/costs.
"For example, there are millions of asteroids of different sizes and composition flying throughout space. One category, known as S-type, is composed of iron, magnesium silicates and a variety of other metals, including cobalt and platinum. An average half-kilometer S-type asteroid is worth more than $20 trillion."
Well, it's not like you are just going to grab the whole thing and bring it back. But I agree, a new source of rare metals would change the pricing. But, platinum is an interesting metal -- engineers could do a lot fun things with it if it wasn't so expensive.
An 800m iron asteroid would make a rather nasty weapon.... (about 75000Mt, probably considerably more than all of the nuclear weapons that have ever existed).
Also: The Universe is awash in energy; it is young, and full of hydrogen, much of it pre-packaged into natural fusion reactors for us. Most of it is available only on not-Earth. If it were actually feasible to put things in space we could use more of it without the issues arising when we try to do it in our biosphere.
Space is dead. Anyone who gives a couple moment's serious thought will realize that there is really no such thing as environmentalism in space; at best, there are aesthetic concerns. (Many people are not willing to be even that thoughtful, but one can only try so hard.) Some of the things we do on Earth would be better done in space, where the byproducts hurt nobody.
Earth does not have enough of certain resources for a really green society, based on the fact that many green technologies require various rare earths. Getting more palladium or platinum from space will enable us to build more fuel cells and such things. For bonus points, combine with the above and refine in space.
The problem isn't that space is useless by any means. It's not even useless to "the poor". It's just that there is a bit of a gulf between where we are now (communications satellites) and where it becomes really useful, even putting aside the technological advances that will arise in response that will be useful even on Earth (a lot of environmental systems work and such).
There's also the not having all our eggs in one basket, which is getting increasingly important. I do fear in the race between anybody's brother being able to build a biological uberkiller and having humans in space the former will win by a decade or four.
Humans are going to space. Those that do are going to be awash in riches far more vast than the Europeans found in the New World.
America today is a great place for innovation, startups, and living comfortably. We got here by risk-taking, fighting, exploring (geographically and other ways). The people who continue to do that will be the new America. I'd like the old America to be the new America.
The massive amounts of resources in space are, of course, a plus... but I mostly like it for the long-term survival of the human race.
Setting up self-sustaining human presences in colonies on other planets/moons makes it vastly less likely that some random asteroid that we didn't see coming will wipe us all out. (Other potential disaster scenarios involve superviruses, grey goo, that giant volcano under Yellowstone, and so forth.)
4) so we can build a defence system against incoming asteroids
5) to provide a hedge against a planet-wide disaster (aforementioned incoming asteroids, global nuclear war, massive pandemic, runaway greenhouse, take your choice)
6) to get at the massive resources available in space - this one needs expanding. Yes, with current technology those resources simply aren't cost effective compared to what is available on Earth. But imagine a world where space-gun launches or a space elevator reduce launch costs down to tens of dollars per kilogram in orbit. That changes everything in the cost benefit analysis of space resources.
7) Science. We'll discover things that we never even imagined could exist. Who knows how those discoveries will aid humankind in the future?
Those are just some reasons that I could come up with quickly, but I'm sure there are others
8) Expansion (possibly covered by 5)
9) to preserve the Earth: if we can move the bulk of the human population off planet before we totaly destroy it, we could turn the Earth into a planet-sized nature reserve.
>But we are not making any serious effort as a society to cross those valleys. It is not clear why.
Because what we have works well enough for the task at hand. The problem, which Stephenson alludes do earlier in the article, is space just isn't proving to be as useful as we had anticipated. We're already doing pretty much everything that you can justify from an economic perspective. Activities like asteroid mining and space solar power aren't going to make sense for centuries even if we dramatically cut the cost of $/kg to orbit.
So sure, we could do the same things more cheaply after a huge investment in, say, tethers. But so what? How does it make sense to spend trillions on a new launch technology when you could use that money to buy all the conventional rockets you'll need for the next 50 years.
His points translate more or less directly to nuclear fission - Uranium-fueled light water reactors are the result of another "hill-climbing process" and other reactor designs (HTGR, LFTR, etc.) receive almost no commercial attention, in large part due to the regulatory, accounting, and insurance burden of proof that any new design has to meet.
It's worth noting that the article that always gets cited every time the Dvorak keyboard gets mentioned, ("Fable of the Keys", claiming that it's a myth that it is more efficient than Qwerty) is based on the near religious belief amongst certain groups of economists that path-dependency of this sort doesn't exist.
They believe we use rockets/Qwerty/Windows because they are the best and the all seeing market has chosen wisely, not because of a series of effectively random decisions and coincidences that occured in the past.
Great article but the modern oil industry did not grow out of whale hunting. In fact it bankrupted whale hunters because it provided cheaper oil.
Also we would only end up as "the Ottoman empire of the 21st century" if someone else creates a radically superior technology which we spectacularly fail to copy, reproduce, or simply license form them.
I think the sad truth is, there just isn't much to do in space if all you have is a cheap and easy way to escape earth's gravity.
Sure space tourism would be fun for a while. But at some point we need to either terraform something or decide to live in city sized spaceships, THEN you'd have a real incentive to innovate launch vehicles.
Anyway that aside, it really brings up the old infrastructure problem -- Large investment in infrastructure is a two-edged sword, bringing both benefit and lock in, and when it comes time to change there is lots of debate. I think Neal either misses or avoids a big part of the argument here -- infrastructure is turned into the bad-guy and the good guy. People don't see the benefits it has brought for a myriad of reasons - they have internalized it, they are not in the class of society that directly gets money for it (and the general improved life isn't apparent because their neighbors are in the same boat), they are afraid of tax increases (or lack of tax cuts), they don't understand the current tech, and figure that "it has problems so anything else we do won't fix it either", it has broken society in the following ways...
Really all of this though is just setup. The way I see it, large scale tech and infrastructure projects are hard to get at and harder to revamp because they are just too easy to target with populist attacks from all sides. The issue is usually complex, but easy to attack with a simple disingenuous quip. Doubly so when the alternative is something that sort-of works, because then the quip doesn't even need the effort of disingenuety, just a mean spirited "they are trying to change the perfectly good stuff we have just to take it away from us, and ignoring everything else to fix"
I have no idea of the solutions we could offer to these types of scenarios, but I do think that somehow we need to find a way to look at these piles of infrastructure we have an find ways to make them better. To do that we need to get around the "infrastructure problem".