Jump to content


  • Posts

  • Joined

  • Last visited


72 Excellent

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. Define generations*. Also, we wouldn't use fMRIs. Too low resolution. The approach that might work to get there would be first to make connectome or synaptome. (the mapping of every single connection in a preserved brain or even the individual synapse strengths - a big huge project that would require billions of dollars and warehouses full of slicing machines and multiple electron beam microscopes) Then we'd develop what I called a "scaffolding AI". This thing would be able to look at the synaptome and connectome and essentially convert it from a morass of billions of connections to a color coded visual map based upon the patterns common to each area. Then you'd build other AIs that essentially are a program, probably written in tensorflow, that can take a region from this brain map and write a program in tensorflow that mimics the connectivity patterns of that region. Then you'd spend many years developing ever more sophisticated systems that have ever more complex internal models and internal representations of the world. Having stolen heavily from the human brain your overall machine would be vaguely human like (eventually the machine would have many internal memory regions where it has a concept of it's own relationship to a model of the world, where it models possible courses of action against that model, and so forth). But yeah, if alphaGo is a few dozen systems working together, this would be thousands of them. A vastly larger and more complex system. You'd need moar simple AIs just to detect failures and help you debug this thing... Nevertheless the basic concept is to take what we have - simple but powerful tools that use a form of artificial neural network to be able to perform many well defined tasks better than humans - and to use it to build the scaffolding that would eventually let us make a fully sentient AI. The fully sentient AI would be a massive machine composed of thousands of artificial neural networks, with rules and connection patterns borrowed from a once functioning brain. Sort of how before people could create Python they needed to write assemblers and then Pascal/C programming languages in order to write Python in. And create a whole support library of algorithms and an OS to provide an environment so it would run on multiple computers. Once you realize the scale of the task - thousands of interdependent systems each as complex and hardware hungry as alphaGo - you realize that various people who thought they could invent AI decades ago were smoking the good stuff. They had no inkling of the scale of the task. That's why there is a false perception that the problem is intractable. That's one approach. But even a far simpler one where you start with a defined set of tasks would be useful. For example, a machine that uses a kinematic model (similar to an autonomous car) and can manipulate components in a factory. So it would internally know the shape and weight distribution of components. It would be trained to recognize said components reliably with just a few reference photographs. It would be able to pick the components up and manipulate them to the final state in a mechanical design for a part. So essentially it would let you design a machine in a CAD program and the factory robots would be able to build most designs with minimal training. (just like a human, it might need some hinting as to correct assembly order) Not as sexy as Hal9000 but achievable in the near term. You could build robots able to repair other robots, so long as the failed component is in an isolated, replaceable module with a simple type of attachment, by swapping parts. I'm saying if we had a rational society we'd dump the lion's share of non essential GDP into things like this. * In conclusion, true general artificial intelligence is closer than "generations" if you meant 60-90 years, because the road to getting there is to combine various advanced general tools like tensorflow and recent advances in brain scanning technologies to speed up the process of getting there. That process is an exponential one, it might seem like we're only 10% of the way to AI when it's only 10 years away as a result. As for the definition of it : all it really means is that any useful task a human can do, it will be possible to use a collection of AI tools to make an AI that does the same thing at at least average human levels of competency. Not one omni-being that has emotions and a voice and all that jazz.
  2. You know, we don't actually need to do that, either. The solution to this problem isn't to collect resources from asteroids. Not yet. The solution is to develop either some form of self replicating factory (either via a bunch of really compact and flexible robots or nanotechnology) or some form of useful AI (by improving tensorflow and studying the synaptic mapping of the human brain and developing simpler scaffolding AIs that can read a synaptic map and extract the pattern of how the brain's regions are organized). If we were a rational society we'd spend all 18 billion that would go to NASA, and maybe another 200 billion that would go to the military, and maybe another few hundred billion that would go to medicare (we'd start freezing hopeless medicare patients instead of wasting money to extend their lives a few weeks) into this effort. Maybe we'd even tax heavily our richest members. Do we as a society need more mansions, or should we spend instead spend a couple trillion a year developing the technology to defeat our species's biggest problems? Anyways, obviously with some form of AI, some form of self replicating factory, some form of digital immortality (a form of AI but it would mean our society would have members who have a longer view than the next 30 years) - with some or all of things like that, most of the problems of space would plummet in difficulty and become almost trivial. And all that stuff can be developed on earth, using massively scaled up versions of what we already have. Don't see any reason we need metric tons of platinum and rare metals to do it, either...
  3. Nibb, have you discussed this in depth anywhere else or have a link to an in depth discussion of this architecture? I'm rather curious to hear what NASA did wrong or how they might do it better. I mean, I take it that doing 30 falcon heavy launches instead would let you skip ISRU and probably save money vs SLS launches?
  4. The basic concept of life is you have self replicating molecules that obtain energy and resources from the environment and copy their patterns before they are destroyed. All life is like that. The trouble with Venus is it's so hot that it's very very chaotic at the molecular level. Anything that tries to have a pattern gets ripped apart by the heat before it can successfully copy itself. Most substances are an energetic gas at those temperatures. So Venus is basically impossible*, while Titan is so cold that life would be very slow - but it could work. * "life" that would function on venus, maybe, would have to be devices that have insulated shells and a way to eject heat using pumps. Inside the shells the machinery would be a saner temperature. Basically self replicating robots if they were very large* might function on venus. *the reason the machines have to be large is because insulation depends on shell thickness. So there is no way for microorganisms to live on Venus.
  5. What I was proposing was that the advantage of the asteroid belt from a "how do we get there from here" perspective is that we could tow an asteroid to high earth orbit. Then, we could practice living in a habitat tethered to it and practice mining. We could do this for years and years and also simulate the actual communications lag the real colony would face. Once the test colony has survived for a decade or so and we have fixed technical problems so that the required supply of spare parts from earth is small, then the test colony gets boosted to the asteroid belt to live there permanently. Since there's no large accelerations required, the fragile test colony station hardware would probably be able to be reused. I'm not looking at long term really, just how do we organize and develop the capability to live somewhere else at all. To me the asteroid belt seems easier because it's possible to "cheat" and simulate virtually all of the conditions the colonists would be under a few days from earth.
  6. I like it. It's a very grim and depressing setting - but it also would make an awesome setting for a video game. Maybe right after the primary fighting has stopped. The reason is that video games need a setting where the amount of total art and assets needed is feasible for a studio of a practical size. That's why the whole Earth, undamaged, with all the people and all the cities can't be in a feasible game - it would take too long and too much money just to model all that stuff statically. Much less try to predict how geopolitical politics would go once the game starts. What is feasible is if the earth has been pounded into a smoking crater oozing magma. A few successful alien attacks have eliminated all the mega-group of survivors, leaving tiny outposts with a few hundred people in each scattered throughout the asteroid belt. Something you could actually practically model, about the size of Diamond City in FO4. It would be a derivative of KSP - you'd start out somewhere and have access to various scrap parts and have to device a spacecraft to get around in the asteroid belt. Instead of science to unlock stuff you'd go find wrecked equipment leftover from the fighting and have to scavenge for components... Naturally in the plot of the game there's a surviving group of aliens, and they are willing to talk, and your character gets a choice to either wipe them out or work with them.
  7. The problem isn't destroying the enemy ship. If it's made out of matter and barring some unknown to current science and physics principle, that part is easy. The problem is if the aliens are competent enough for interstellar travel, they can probably rebuild their ship - it'll take your Orion weeks to months to get there - to have a honking gigantic space laser with incredible range. It would then proceed to snipe each torch missile from a million kilometers away.
  8. https://en.wikipedia.org/wiki/Black_hole_starship Unlike wormholes, black hole engines violate no known principles of physics. It is true that science is a bit uncertain about certain critical facts* as we've never seen a black hole, so using one as an engine may not be possible - but from current knowledge it is possible. All it is is E=mc^2 in the opposite direction. The black hole is acting like a perfect converter of matter to energy, and then your ship just has to reflect the released gamma rays (doable, using methods demonstrated on earth) backwards. The black hole must be able to hold an electric charge so that it can be kept attached to your ship's engine bell via magnetic or electric fields. I find is reasonable plausibly that you can do this. What isn't plausible is wormholes because they require negative energy, which current science has never observed anywhere. We've observed black holes and theorize that a small artificial one would do what we want. Those theories may be wrong - but maybe fusion drives are also impossible. Maybe AI's impossible. And molecular nanotech. Essentially you can assume that anything that doesn't exist right now this very week is impossible due to undiscovered laws of the universe. I would call that "motivated cognition" - you disagree with me on plausibility and thus want everything I want to be impossible. * the 2 major critical facts are : can a black hole hold an electric charge, and how easy is it to feed a tiny black hole via a particle beam? TLDR, a black hole engine is : 1. Ram together metal rods at close to the speed of light to form a black hole using an apparatus that spans across a solar system 2. Feed the resulting black hole from a particle beam so it remains the same size. Size determines power output, it would emit gigawatts to petawatts in the form of gamma rays so it burns matter constantly. Feed the hole a diet rich in either positively or negatively charged particles so that it retains a charge, even if said charge is actually matter clumped at the event horizon and the hole itself is neutral. 3. Using magnets and electric fields, install it in a starship, which is a gigantic parabolic dish of gamma ray reflective lensing material that bend the gamma rays to all travel backwards. 4. Ride that baby to wherever you want. You can even collect interstellar hydrogen to feed the black hole, unlike a fusion ramscoop you would actually gain velocity doing this. You have an ISP of something like 5C...and an acceleration somewhere between a few centimeters per second^2 to m/s^2. (limit is how well you can reject waste heat from the gamma ray reflector and how much of the gamma rays are absorbed) An infinite improbability drive is a purely fictional contrivance and it has no mechanism described whatsoever. I'm sure you can see the difference now between the 2.
  9. No. Wormhole generators are not known to current science. If you had typed pion propulsion into wikipedia, you'd get this : https://en.wikipedia.org/wiki/Antimatter_rocket TLDR, if you combine antiprotons and protons you get charged particles called pions. You can redirect them out the back with magnetic fields. People have worked out how strong the magnets would need to be and what geometry they would be in. All feasible with current science. The missing piece is mass production of antimatter but there are plausible methods to do that. A ship decelerating from interstellar flight headed for an Oort cloud object would be interceptable as it would be going far slower at the end of it's journey. There isn't stealth in space. All weapons use thrusters to course correct during flight, casablanca howitzer rounds would have onboard control and thrusters. There's no reason to assume that the aliens could "hack" or "fool" human systems they have never encountered, especially since the hard laws of physics means their ship must emit a signature flare that is incredibly bright in order to have the ISP for interstellar flight. Yes, if the aliens plan to ram the earth we are screwed. I'm assuming they recognize the earth, as a naturally occurring biosphere, is in itself valuable and they intend to seize it without destroying it's value. Their plan, if they got that far, would be to dock with oort cloud objects. Self replicating factories would rapidly expand, and in a few years they'd have an unstoppable fleet of thousands or millions of warships. They'd be able to conquer earth by sniping all the military forces with laser fire from orbit and then sending ground robots to the surface to root out any remaining resistance. As they would have self replication and artificial intelligence, casualties would be meaningless.
  10. Ok, let's assume the aliens don't have force fields or antigravity or anything else that known science doesn't say is possible. So what do they have? Well, they have interstellar flight. And they must have used anti-matter pion propulsion to do it. (or maybe a black hole engine). So there's actually one big advantage we'd have on Earth. Warning time. Their ship would take decades to decelerate, and leave a flare of gamma rays as it does so. Both engine types emit much of their exhaust energy in the form of gamma rays. Ok, what else do the aliens have? They probably have molecular nanotechnology, and they probably can copy themselves very rapidly given the needed resources. But their starship itself is probably very small and light at the end of an interstellar voyage, having burned off most of it's mass during the trip as propellant and expended stages. If that's true - again, I'm assuming plausible limitations based on current scientific knowledge - then it's not hopeless. Basically you have to assume that the alien ship is higher quality than anything we can make on earth. It can probably reconfigure itself and repair damage very rapidly. So you need at least 100 : 1 odds. If the alien ship weighs 1000 tons of payload*, you need to send 100 kilotons of warships to intercept it. And you have to reach it before it can dock with an oort cloud object and grow rapidly into an unstoppable fleet. What would the warships be? Orion nuclear pulse battleships, armed with lasers and nuclear howitzer rounds. Obviously. There's no stealth in space so you just have to get into range. Realistically you'd probably want to try to ram the alien ship. You've got to assume that long before you get into range, the alien ship would probably reconfigure itself into a gamma ray laser with incredible range or something. Maybe 100 : 1 odds isn't enough... * you calculate the payload mass of the alien ship with spectroscopy to analyze their engine flare, then you look at how fast it is decelerating. It certainly sounds like an interesting scenario.
  11. You could put an asteroid in high orbit around earth (I said LEO before but yeah it has to be in a permanent orbit above any significant atmospheric drag) and set your whole colony up. Test the life support equipment, the mining equipment, everything. The only difference between this colony and the one you'd really use is the sun is brighter and the asteroids are small because they had to be hauled a long way. You empirically test your whole plan for self sustainment. How many parts does the colony actually need from earth in practice? How well do the colonists cope being isolated from Earth? (you'd simulate the communications time lag unless there is an emergency). How well do things work over years? It just seems so much more achievable to be able to practice everything a few days flight from earth than to hit Mars and it's do or die... Why do you think it's easier to stay alive? What specific environment conditions are worse out in space and more likely to kill your colonists?
  12. TLDR, the reasons the asteroid belt might be easier are : 1. No stress of the aerocapture/aerobraking and the landing burn on your spacecraft. Freed from the design constraints of that (and the need to undergo such a high stress maneuver months away from the technicians who assembled and inspected the vehicle) you could have a very different spacecraft architecture. Since you'll never see atmosphere, you could make a bunch of rectilinear modules that can be bolted together in orbit many different ways. Every module would have an internal aluminum honeycomb structure and would be able to tolerate the stress of connectivity in almost any direction under the very low accelerations of the maneuvering needed to reach the asteroid belt. 2. You could deal with the radiation by finding a naturally occurring cleft in an S-type asteroid 1 km across or greater. 3. Colonists would live in inflated balloon style habitats, and inside the habitats would be carousel centrifuges providing the actual habitable space. I've realized a simple open framework of aluminum struts and floor gratings (very similar to what's used in skylab) makes a ton more sense than spinning the whole habitat. There would be 2 counter rotating carousels, each with a number of rooms and an interconnecting ring corridor all the way around. They way you board or get off is you get into an open cage at the hub and press a button after closing the door. Motors sync the cage to spin at the same speed as the carousel. You open the other door and done. Any time you want to maintain the carousel, you stop it and work on it in shirtsleeves. It's assembled IKEA style from a rocket load of pre-cut carousel parts and it's all bolted together by astronauts working in shirtsleeves inside the inflated habitat module. Only minor issue I know of is that the sound of the air flowing over the carousel as it spins might get very annoying after a while. Also, it would continually draw power as the carousel loses energy from air friction. 4. The colonists aren't stuck at the bottom of a gravity well. If they want to come home, they bolt together some of the modular pieces of their station/habitat and start a slow burn for earth orbit. Engine failure is not a big deal - as long as you still have enough propellant (tank it in separate modular tanks with safety valves so a leak won't cause the contents of more than 1 tank to be lost) and at least 1 engine remaining you can probably make it home. 5. You can test this whole system, full up, in low earth orbit. The only difference between the colonists living in orbit and living in the asteroid belt is the microgravity of the asteroid and the diminished sunlight. Capture an asteroid and move it into orbit, and the astronauts could try out their various mining technologies and test out living there and test their closed loop life support and everything. 6. Much of the mining would have to be done by teleoperated (or full AI) robots. Apparently you also need a C-type and X-type asteroid. Hopefully you can find all 3 within a few light-seconds of each other so teleoperation is possible. Ore hauling vehicles would take months to get between asteroids. 7. There's immense surface area, making mineral prospecting easy. Just build a hopper probe and send one to each candidate asteroid. It lands on a part of it, drills a sample and uses spectrometry, then hops to a different spot with a puff of hydrazine. Sure, the ESA had a bit of an issue with their first try but it's doable... Anyways, what makes Mars easier, other than the fact that it admittedly looks a lot more like a livable place in photographs? The core of my argument is that looking at this as an engineering problem from a high level view, the asteroid belt is a far more tractable problem than Mars. Being able to do full up testing, being freed from many causes of sudden catastrophic failure, and so on all make it far more doable.
  13. In a conversation I was having with a guy, he was going on about how the USA might have secret orbital anti-ballistic missiles. I thought it was highly unlikely, but I can't prove the USA doesn't secretly have them, so I tried to argue it was basically unfeasible. I only have a rough idea. ICBMs from the main semi-unfriendly nation (russia) with a lot of them launch from various bases and come over the North pole on the way to the USA. The ballistic trajectories have a very high apoapsis and the missile doesn't have the energy for orbit, at least not while carrying it's warhead. As near as I can tell, your interceptors would have to be in low earth orbit, in some kind of polar orbit, flying the opposite direction from the tracks the ICBMs would take. Essentially the parabola of the ascending and descending ICBM is going to intersect the orbital altitude of the interceptor twice. So the interceptor needs to be there when it happens. Err, maybe not, maybe the interceptor needs to use an energy minimization algorithm where it considers a number of points on the missile's ascending and descending arcs and the missile would need to use a gradient descent to find the point that requires the lowest energy for an intercept. Essentially solving this problem with realtime numerical integration of different strategies to find the one to use this time. That means you'd need a lot of interceptors, enough that one is passing by a given point every few minutes. And they'd need high thrust engines and a lot of dV. Basically an entire rocket stage, in orbit. I'm not quite sure roughly how much dV you'd need - I want to say it would be around a kilometer per second but I'm not sure how to go about calculating this. How would you calculate the requirements?
  14. Wait, what? No. There's studies on this for the Daedalus project. Impacts are rare, the problem is they damage the shield. And every now and then, black swan impacts can go through the shield and hit other places. That's why your ship has to be able to fix itself and also not have any key parts. Your logic doesn't even follow - if erosion is that fast, a thicker shield won't help. The reason for self repair is because if all the impacts hit a particular part of the shield by sheer chance, you can move mass from spares or the rest of the shield to shore up the damaged area. For THAT matter, there's other approaches. Damping layers can cause the fragments of an impact to stick inside the shield so you don't lose matter. You would need to restore these damping layers after each hit. Pristine shield components would be far stronger at tolerating impacts that damage components, losing less matter per impact, so being able to rebuild the shield is also critical. You could also use electric charge and other more complex mechanisms to aid things. And collect interstellar gas, fusion it to heavier elements, and use it to restore your supply of matter.
  15. The reason why chemical reactions change mass is simple. Chemical reactions that release energy form more stable chemical bonds. This means the electrons clouds are more compact - the electrons have a smaller distance to cover. That means the electrons travel at a lower velocity, which in terms means their relativistic mass is smaller. So the electrons are very slightly lighter for the products of chemical reactions that release energy. The example of the spring above, given by sharpy, is similar. The unstrained metal spring, the cloud of electrons that tend to travel around metals has slightly less distance to cover and thus they each weigh a little bit less.
  • Create New...