On May 28 I spent the day at the annual conference of the National Space Society — giving a talk, enjoying the other talks, and generally absorbing the space-y energy. The conference was exquisitely timed, only a few days after SpaceX achieved the milestone of docking the first privately-owned spacecraft with the International Space Station.
Since I only attended one day of the four-day event, I can’t give a real conference review — but the parts I did see were interesting enough to write about!
Among other reactions, I came away with the clear impression that at some point in the not too distant future — in particular, once the cost of launch is decreased a fair bit — we’re going to see an incredible explosion of commercial, scientific and recreational activity in near space. And there’s no shortage of detailed technical ideas about how to achieve lower launch cost. As with so many other areas of advanced technology these days, the rate of progress is going to depend less on fundamental science and engineering issues than on funding decisions.
Old & New Eras of Spaceflight
In spite of all the cutting-edge technology, the Space Development Conference felt to me in some ways like a throwback to my childhood. Like most nerds of my generation (I was born in 1966), I grew up fascinated by space travel and outer space in general. Literally my earliest memory is watching Neil Armstrong bounce around on the moon, on TV. I was 2 years old at the time, and I guess I was amazed by the rapt attention all the adults around me were paying to what was on the TV. I also understood that it was pretty cool to be walking on the moon.
I remember being baffled by the words Neil said when he climbed off the ladder: “That’s one small step for man, one giant leap for mankind.” Because the “a” before “man” was omitted in his statement (either by his error or transmission noise) — and also because I didn’t know the word “mankind” at that time — I didn’t understand the nature of the contrast he was drawing between “man” and “mankind.” I incorrectly guessed that a “mankind” was some kind of miniature man, and figured that a small step for Neil Armstrong must be a big leap for one of those mini-guys. I scrutinized the TV screen carefully, trying to find some trace of the little mini-person, the “mankind”, leaping onto the moon from the lander. I think I was 5 or 6 by the time I realized my mistake.
Also like most nerds of my generation, for a period in my childhood I wanted to be an astronaut when I grew up. But at some point I realized this wouldn’t be possible, because astronauts seemed to be drawn from the military and I hated the military, and also because my eyesight was too terrible. So I decided to become an astrophysicist instead, and figured maybe I could get into space on some kind of research mission.
Somewhere in my elementary school years I hatched a clever, space-related plan for living forever and meeting super intelligent robots. I figured I could build a spaceship, leave Earth and fly in a big circle at close to the speed of light, and come back to Earth a million years in Earth’s future — but only a year or so older myself, thanks to relativistic time dilation. That way I would get back to Earth after immortality pills and super-robots and lots of other cool stuff had already been invented!
My interest in astrophysics and spaceship design lasted a while, and morphed somewhat into an interest in general relativity theory, cosmology and unified physics. But by my mid-teens I started to get more and more interested in AI, which has ended up being the focus of my research career. Among other things, I was attracted by the way it seemed like AI was something you could just do. With AI, I felt, if you had the right ideas you could just sit down at the computer and type some code into your programming language interpreter, and you’d get an intelligent machine. Or you could gather a small group of friends to help you, and make even more serious progress. Whereas space science required massive amounts of money, expensive hardware, coordination of large numbers of people, government agencies and so forth.
But that was then, this is now…. Fortunately — as SpaceX, the Google Lunar X Prize, and a host of other current developments are indicating — this prohibitive aspect of space development is becoming less and less dire each year! Space flight is becoming more and more something that ordinary people can “just do.” This was one of the main points repeatedly raised at the Space Development conference … and it’s a pretty darn exciting one.
Buzz Aldrin, the second human to walk on the moon — whom I vaguely remember seeing on TV at age 2 — was sitting in the front row at the conference, smiling enthusiastically. Like everyone else, he seemed especially pleased at the recent trend toward lower-cost, commercially-driven space flight.
The Google Lunar X Prize, the Mars Foundation, and the Cost of Launch
The first talk I saw at the conference focused on exactly this theme of low-cost commercialized spaceflight — it was by Alexandra Hall, who works on the Google Lunar X Prize, a prize that offers a total of US$30 million in prizes to the first privately funded teams to land a robot on the Moon that successfully travels more than 500 meters and transmits back high definition images and video. The prize expires in 2015.
Hall’s talk was exciting, and refreshingly frank. She noted that of the 26 teams currently vying for the prize, 46% have raised less than 25% of the funds they will need. A number of teams are expected to drop out or merge during the rest of 2012. But, it seems clear that a significant number of teams will raise the funds to send spacecraft to the moon.
She also observed, ironically, that to a certain extent the competition may end up to measure ability at fundraising, more than ability at engineering. A number of different teams appear to have designs capable of succeeding at the challenge. However, the cost of launch is proving a difficult issue for many teams. Potentially some teams may seek to cooperate and share the cost of a single launch vehicle. Elon Musk’s SpaceX promises to drop the cost of launch to $1000 per pound, 1/10 NASA’s typical cost in recent years, but this is still a steep price.
A similar focus on the cost of launch emerged in my discussions with the folks from the Mars Foundation , who have created an impressively detailed plan for a small settlement on Mars.
They reckon that they can keep a small enclosed ecosystem running on Mars, consisting of humans and plants, and using machinery to gently adjust the gas balance of the artificial atmosphere as needed. They seem to have worked out the science and engineering issues rather well. But the main issue they face is cost of launch — it would cost a lot to get the needed 250 tons of material off Earth and into space. Their current plan for doing this involves 10 smaller missions, each bringing a couple people and a lot of materials. At SpaceX’s $1000/pound launch cost, getting 250 tons off Earth would cost half a billion dollars.
An Amazing influx of Indian Students
Not everyone at the conference was frustrated by the high cost of launch though. Some of the younger participants, while acknowledging the issue, felt confident that the cost would come down dramatically in their lifetimes, enabling them to travel and perhaps even reside in space well before they got old.
The conference was particularly full of high school students from South India, who were all participants in a contest to design a space habitat, and gave 10 minute talks on their designs, as well as displaying posters in the exhibition area. Not all the contest participants were from India, but it seems a considerable majority were. The high school students brought a bright-eyed optimism and youth to the conference, which otherwise was relatively dominated by gray-haired space geeks from the government world.
Here are a few examples of the student space habitat posters, just for fun:
Synergizing Government, Commercial and Amateur Space Scientists
After the Google Lunar X Prize talk was a panel discussion on the creation of the organization hosting the conference, the National Space Society, which apparently was recently formed via a merger of two other societies: the National Space Institute and the L5 Society. As the tale was told, the National Space Institute consisted mainly of aging Washington insiders with decades of experience in government space projects, whereas the L5 Society was mostly a bunch of wild-eyed amateur space aficionados with science fiction on the brain. Everyone present seemed quite pleased with the merger, and felt that the resultant combination of vision, passion and practical know-how and networking had a lot of potential to help accelerate the path toward low-cost space-flight and eventual space colonization. There was also talk about ongoing collaborations with space societies located in other countries — but I sensed a bit of minor conceptual friction between folks concerned with space development as an international human enterprise, and others interested in making sure the US remains in the lead.
Lori Garver, a panelist and the Deputy Administrator at NASA — and former Executive Director of the National Space Society — spoke candidly about the typical attitudes that government employees have toward space projects, as contrasted with the more ambitious views of many amateur space geeks and space entrepreneurs: “Typically what you have in the government is people who want to do the project that exists, continue the contract….. We’ve been trying to get them to look out 25 years, and to look at different scenarios besides Apollo-like scenarios. One of these scenarios is settlement. But we’ve been having trouble getting them on board.”
Garver came across as an ambitious personality and a strong advocate of space colonization. But her view was that government, due to its conservatism, was unlikely to push hard toward this anytime soon. She placed more hope in commercial efforts, catalyzed by government-funded science advances.
One of the other panelists emphasized the same point: “We must take the $14 trillion economy of the US and figure out how to direct some of it to make a profit from space…. We probably spend more on astrology than we spend on astronomy. …. What NASA is doing is to take a very small percentage of our money and use it to catalyze the development of new technologies …” He viewed NASA as creating core tech, that would then need to be used by commercial ventures to get the really exciting stuff done.
Another panelist focused on the need for space amateurs to get involved, to prod the government and corporations in more ambitious directions. “We have to continue to bring together the professionals and the amateurs — the people who are enthusiastic about space…. We need to close these two worlds and bring them together – the imagination and the practicality. … Bring in people who are passionate about space, and bring them together with people who work in the field….. If we get up to 1% of the GNP in space, we will continue to go into space, we will go to Mars and the Moon — and in doing so we will become rich not only in material things, we will become richer spiritually ….”
These comments made me think of all the Indian high school students, and their beautiful space habitat designs. Talking about their hopes and expectations to live in space one day, some of them did seem to have an almost spiritual glow in their eyes. Of course, the majority of them will wind up in careers other than space — just like I did, in spite of my early enthusiasm for astronautics. But maybe a few will launch Indian space start-ups — I hope so.
AGIs and Uploads, Not Meat Humans, Will Colonize Space and Explore the Universe
My main purpose at the conference was to give a talk in the special session on Transhumanism and Space, which was organized by the session chair Karen Mermel and Humanity+ Chair Natasha Vita-More. My talk was focused on Artificial General Intelligence and Whole Brain Emulation in the context of space exploration. My main point was a simple one: humans aren’t made to live in space. Keeping the human body alive in space is a pain, and humans are bound to get bored while traveling for thousands or millions of years to reach distant stars or galaxies (which is what it will take unless someone invents a teleporter or something similar). The only way humans will get through those long space flights without going nuts, I suppose, is to connect themselves to virtual realities Matrix-style. Whereas, robots can operate better in space than on Earth — computers love cold temperatures, and they don’t need air, food or water, just electricity. And they aren’t stuck with human motivational and emotional systems; they can be programmed to enjoy the long isolation of interstellar or intergalactic flights. Heck, they can spend the time reprogramming themselves and improving their code, making themselves smarter and smarter along the way. The only way humans are going to the stars, I reckon, is in the form of uploaded humans — human minds extracted from human bodies and put into digital form. Even then, the uploaded humans will probably end up modifying themselves somewhat to improve their harmony with the deep-space situation.
Nobody in the audience really disagreed with my points rationally, but they didn’t like them emotionally. Most of the people at the conference seemed to badly want to go into space personally, as humans. I can’t blame them — it would be lots of fun. And humans will go into space. But colonizing space, and exploring the universe — that’s going to be mostly for AGI robots.
Paul Werbos on Politics, Stupidity and the Cost of Launch
Besides me, the Transumanism and Space session consisted of David Orban and NSF Program Director Paul Werbos. David gave a fantastic and well-received talk on distributed sensor networks, but I was particularly struck by Paul’s talk, insofar as it hit at the heart of one of the themes from earlier in the day: the high cost of launch.
He put the matter quite simply and directly: “If we want to put solar power satellites in space, $200/pound or less is what we need. It’s also what we need for humans — for business plans that require humans in space at a large scale.”
$200 per pound is only 1/5 of the $1000 per pound that SpaceX promises. But still, a factor of 5 additional improvement may potentially take a while to come about.
At $200 per pound, the Mars Foundation’s design for a startup Mars colony would still cost $100 million for launch costs alone (which they say are the vast bulk of the costs). But that’s definitely starting to move toward the price point where they might actually be able to get funds to execute their plans.
Werbos acknowledged my point that robots would have an easier time in space than humans — but expression his passion of human space colonization, even as he lamented that government bureaucracy was placing obstacles in the way of the realization of this vision. “The robots are doing pretty well. But I would really like to see humans settling space…. I think it’s of fundamental value whether humans can settle space — but we’ve got to get the cost down…. I think it’s possible, but we’ve screwed it up. Right now, it doesn’t look good. The technology is there, but the politics is keeping us from getting the technology.”
He talked a bit about the National Aerospace Plane , which was briefly and partially developed in the early 1990s but never completed.
As Paul describes it, this was “technology to get us into space for 1/10 of the shuttle or less…” — and remember, this was decades ago. He attributes the failure of the NASP project directly to bad bureaucratic decision-making: “Some people say, just get the program funded and then let them execute it. But that’s not true. It’s our job to make sure they don’t take a good idea and turn it into pork barrel bullshit. There is a tendency in Washington to take the best ideas and turn them into utter garbage. … The technology that we need to get into orbit cheaply was eliminated … these two lobbyists sold NASA on the idea that to get a spaceship back to Earth, what you need is active TPS, active fluid cooling. But active TPS basically doesn’t work. And they got rid of the part that did work.
“You need a high quality nonlinear controller. The skeptics said to me — nobody knows how to solve a complex nonlinear problem like this. But I told them, I know how to do it…. We solved the intelligent control problem, but there was still another problem… hypersonics.”
The discussion became more and more technical — but in short, he conveyed that, in his view, the problems involved with hypersonic flight had been on the way to being solved back in the early 90s, but funding was diverted to unworkable solutions such as active fluid cooling, and then the project was shelved. At this point, he suggests, the best approach to make something like the NASP is probably to use what is known as “hot structure technology” . This image shows some recent work in this direction — a carbon-carbon component of the X37 hypersonic vehicle, cooling from a test temperature of more than 2500 degrees:
“We have the technology right now, with off the shelf technology, to get to space for $200 a pound. It can be done. And the only reason it isn’t getting done is stupid politics of one kind or another….
“Hot structure technology is stranded and abandoned — because of stupid politics. Boeing lost the only test article. This year the Trans-Atmospheric Vehicle team was sent to the four winds; the people are still available but barely and not for long.”
TAV type technology is fascinating, and a somewhat different direction from what SpaceX is pursuing. I asked Paul how much it would cost to build a trans-atmospheric vehicle capable of inexpensive launch, as he was describing.
His reply: “For a totally modernized version, $150M. $300M for a global reach vehicle, which would save the technology…. And SpaceX lacks this technology. NASA R&D could help them all.”
Toward the end of his talk, Paul Werbos’s frustration with society’s unwillingness to fund what is, in his view, a very clear path to low-cost launch, led him to muse on the theme of improving the quality of human decision-making. “I’m afraid human intelligence — the mammal intelligence we have — isn’t enough to solve the problems we’re creating. If we were smarter we wouldn’t be involved in all this stupid politics. There’s something that I call quantum multi modular intelligence. There are two ways to achieve this — you can build computers, or you can get people to use their potential more thoroughly.” Glancing at me, knowing my research focus on creating AGI, he said pointedly: “I’m really more interested in the latter approach — getting people to use their potential.”
But improving human intelligence also, in his view, requires more of an innovative approach than the science and engineering establishment typically takes. “The neuroscientists on the own are not getting anywhere. The engineers are not getting anywhere. We need to put these different kinds of people together, if we want to understand people, or if we want to make robots.”
I know from Paul’s frequent comments on the mailing list of the Lifeboat Foundation, an organization of which we are both Advisors, that he is very concerned with the risks posed to humanity by advanced technologies. One of the reasons he is so eager for humans to colonize space, is as insurance against the risk of us destroying ourselves here on the home planet. With humans living on Mars or the Moon, and shooting in spacecraft toward the stars, humanity will still survive even if nuclear weapons or synthetic biology or rogue AI destroys human life on Earth.
He reiterated this point at the end of his talk. “It’s serious. There are risks. People say, ‘How can you work on this technology if we can’t trust ourselves to apply it responsibly?’ I say, ‘But we have to understand ourselves… If we don’t understand ourselves, we don’t have a chance — any chance.'”
In a hallway chat after Paul’s talk, an Indian student, about to start law school in DC, expressed his sympathy with Paul’s frustration with short-sighted bureaucratic decision-making. “There’s a lot of great science and technology in India also… But you have the same problems there as in the US — a lot of politics. It’s the same anywhere. But in India it’s probably even worse….”
Onward and Upward!
In many ways, it seems, space science is similar to my own field of AGI. At this space conference, just as at the AGI conferences I help organize, there was a wild and wooly mix of the old guard, new maverick researchers, and starry-eyed amateur enthusiasts. In space science, like in AGI, you have a lot of big ideas — and nowhere enough money around to test and realize them … and a lot of frustration at the apparently misguided funding decisions of the Powers That Be.
But overall, in spite of the frustration at NASA’s failure to create low-cost launch methods, the tone at the conference was enthusiastic. The successful SpaceX docking with the International Space Station … the wonderful space habitat designs from the smiling young Indian students … the buzz around the Google Lunar X prize … all of it added up to a real feeling of vibrancy and excitement.
I really got the feeling that if one simple problem — decreasing the cost of launch — were solved a little more convincingly, then commercial space flight would massively thrive. And we’d probably see start-up colonies on the Moon and Mars and so forth — and the Seasteading Institute would get complemented by a Spacesteading Institute.
Werbos estimated the cost of creating a vehicle capable of $200 per pound launch at a few hundred million dollars. That’s a hell of a lot of money from my point of view as an AGI researcher — in AGI you can do a lot with a million dollars, and ten million would be incredible bounty. Yeah, space is still expensive. But still – $100M isn’t that much money, by historical space-program standards. Any major government can spare that easily. And if one government does it, the others will copy.
And if the approach Werbos prefers doesn’t come about, probably some entrepreneurs will prevail with a different approach. SpaceX is already promising $1000 per pound launch.
I have a feeling we’re going to get to Werbos’s hoped-for $200/pound launch cost reasonably soon, via some chaotic mix of government and commercial funding — propelled by the enthusiasm of amateurs and professionals alike. Ultimately this work is going to pave the way for AGI robots to propel themselves from Earth throughout the universe. But, alongside that robot outflux, we old-fashioned meat humans may well have some fun exploring and inhabiting near space… the more the merrier!