Is it useful to think about the difference between ‘physically possible’ i.e. obeying the laws of physics and possible to engineer? In computer science there is something like this. You have things which can’t be done on a turing machine (e.g. halting problem). But then you have things which we may never be able to arrange the atoms in the universe to do, such as large cases of NP-hard problems.
So what about in physics? I have seen the argument that if we set loose a paperclip maximizer on earth, then we might doom the rest of the observable universe. But maybe there is simply no sequence of steps that even a super brilliant AI could take to arrange matter in such a way as to say move 1000kg at 98% the speed of light. Anyway, I am curious if this kind of thinking is developed somewhere.
We know, and can formalize, what the upper limits of computing are. We have no idea what the upper limits of physics are. Even if we assume we know all the things there are to know about physics, that doesn’t mean clever engineering can’t come up with a way of doing it.
Even if you only move at a fraction of the speed of light, you could reach all the stars in a galaxy in a few million years. In the 1960s engineers came up with a way of travelling 1-10% of the speed of light using nuclear bombs. There was another proposal IIRC that used a giant collector in the front to grab hydrogen molecules, and could somehow reach 70% of the speed of light.
If you have nanotech, you can shoot millions of very tiny, pin sized or smaller, spacecraft at significant speeds. Even if only a tiny fraction reach their destination, the nanotech on board could self replicate, and create trillions more.
To correct one thing here, the Bussard ramjet has drag effects. It can only get you to about 0.2c, making it pretty pointless to bother if you have that kind of command over fusion power.
Colonizing the galaxy is a political problem, not a question of engineering possibility. A sufficiently zealous world government could jump-start asteroid mining with an Orion Heavy Lifter, construct mirror arrays near the sun, and start lobbing around interstellar VNMs, all with relatively simple refinement and application of existing technologies. Problem is, the best way of putting a complete industrial base into orbit runs afoul of certain atmospheric nuclear test ban treaties.
Without a reactionless drive there’s no point sending a colony ship faster than about 60% of the speed of light. Gotta save some remass to decelerate.
Is it useful to think about the difference between ‘physically possible’ i.e. obeying the laws of physics and possible to engineer? In computer science there is something like this. You have things which can’t be done on a turing machine (e.g. halting problem). But then you have things which we may never be able to arrange the atoms in the universe to do, such as large cases of NP-hard problems.
So what about in physics? I have seen the argument that if we set loose a paperclip maximizer on earth, then we might doom the rest of the observable universe. But maybe there is simply no sequence of steps that even a super brilliant AI could take to arrange matter in such a way as to say move 1000kg at 98% the speed of light. Anyway, I am curious if this kind of thinking is developed somewhere.
We know, and can formalize, what the upper limits of computing are. We have no idea what the upper limits of physics are. Even if we assume we know all the things there are to know about physics, that doesn’t mean clever engineering can’t come up with a way of doing it.
Even if you only move at a fraction of the speed of light, you could reach all the stars in a galaxy in a few million years. In the 1960s engineers came up with a way of travelling 1-10% of the speed of light using nuclear bombs. There was another proposal IIRC that used a giant collector in the front to grab hydrogen molecules, and could somehow reach 70% of the speed of light.
If you have nanotech, you can shoot millions of very tiny, pin sized or smaller, spacecraft at significant speeds. Even if only a tiny fraction reach their destination, the nanotech on board could self replicate, and create trillions more.
To correct one thing here, the Bussard ramjet has drag effects. It can only get you to about 0.2c, making it pretty pointless to bother if you have that kind of command over fusion power.
Colonizing the galaxy is a political problem, not a question of engineering possibility. A sufficiently zealous world government could jump-start asteroid mining with an Orion Heavy Lifter, construct mirror arrays near the sun, and start lobbing around interstellar VNMs, all with relatively simple refinement and application of existing technologies. Problem is, the best way of putting a complete industrial base into orbit runs afoul of certain atmospheric nuclear test ban treaties.
Without a reactionless drive there’s no point sending a colony ship faster than about 60% of the speed of light. Gotta save some remass to decelerate.