4 ms·

Really cool thought experiment but it's hard to imagine making significant investments on a project that wouldn't produce any results until ~250 years. Any proj

Really cool thought experiment but it's hard to imagine making significant investments on a project that wouldn't produce any results until ~250 years. Any project on that time scale and that distance from "home" would likely need to be completely self sufficient so I would image the project would need a fully capable AI and self replication or at least some sort of very advanced manufacturing capabilities. Autonomous self replicating drones like in universal paperclips are likely the answer.

cerved 5y ago
You'll probably be less disappointed if you read this as a paper about VRP using genetic algorithms as opposed to space exploration
- MisterBiggs 5y ago
  Agreed. I'm actually doing a VRP using genetic algorithms for a final project right now. That said I feel like I know enough about the problem to know that I don't know enough about the problem to contribute to HN about it. Space missions are much closer to my wheelhouse.
  - jhgb 5y ago
    > I'm actually doing a VRP using genetic algorithms for a final project right now Don't forget to compare it with simulated annealing to find out if Skiena was right about GA in The Algorithm Design Manual when he said that he "[has] never seen any computational results reported using genetic algorithms that have favorably impressed [him]".
semi-extrinsic 5y ago
Yeah, the time scales involved are my go-to counterargument to the people who say other intelligent creatures should have conquered the galaxy by now. I suspect that for all realistic lifeforms, given travel at a few percent of lightspeed max, the ROI on interstellar travel (both on the financial and the individual-sacrifice-vs-benefit level) is so bad that you'll essentially only do it when your solar system is doomed. And that takes you from exponential growth down to essentially zero growth.
- ceejayoz 5y ago
  250 years hardly sounds outrageous. https://en.wikipedia.org/wiki/Sagrada_Fam%C3%ADlia https://en.wikipedia.org/wiki/Sagrada_Fam%C3%ADlia has been under construction for nearly 150 years; "construction passed the midpoint in 2010".
  - WalterBright 5y ago
    Some of the European cathedrals took centuries to build. The people who started them knew they'd never see the result.
    - jhgb 5y ago
      St. Vitus cathedral in Prague: constructed between 1344-1929. That's not "centuries" -- that's a better part of a millennium.
- TooKool4This 5y ago
  I think the time-scale argument lacks a little bit of imagination. Time scales are very relative. There could easily be organisms that have lifespans in the thousands of years due to a variety of unique selective pressures (or no concept of a lifespan at all). Or species that have technology that can augment their lifespans to be thousands of years long. I can certainly see humans getting to that point in the next 10-thousand years. Given a thousand year lifespan, what is a few hundred years in terms of commitment to a project? It would be something akin to working a job for 10 years or so, which is not entirely unheard of. The ROI argument is more interesting! Although there are way too many factors to consider to have a definitive answer one way or another.
- iso1631 5y ago
  Once the technology exists to allow cults and misfits to build self sustatining asteroids and stick an engine on them, then it will. However as there would be no meaningful communication outside of a solar system, a single civilisation wouldn't spread to the stars

cerved 5y ago

You'll probably be less disappointed if you read this as a paper about VRP using genetic algorithms as opposed to space exploration

MisterBiggs 5y ago
Agreed. I'm actually doing a VRP using genetic algorithms for a final project right now. That said I feel like I know enough about the problem to know that I don't know enough about the problem to contribute to HN about it. Space missions are much closer to my wheelhouse.
- jhgb 5y ago
  > I'm actually doing a VRP using genetic algorithms for a final project right now Don't forget to compare it with simulated annealing to find out if Skiena was right about GA in The Algorithm Design Manual when he said that he "[has] never seen any computational results reported using genetic algorithms that have favorably impressed [him]".

semi-extrinsic 5y ago

Yeah, the time scales involved are my go-to counterargument to the people who say other intelligent creatures should have conquered the galaxy by now. I suspect that for all realistic lifeforms, given travel at a few percent of lightspeed max, the ROI on interstellar travel (both on the financial and the individual-sacrifice-vs-benefit level) is so bad that you'll essentially only do it when your solar system is doomed. And that takes you from exponential growth down to essentially zero growth.

ceejayoz 5y ago
250 years hardly sounds outrageous. https://en.wikipedia.org/wiki/Sagrada_Fam%C3%ADlia https://en.wikipedia.org/wiki/Sagrada_Fam%C3%ADlia has been under construction for nearly 150 years; "construction passed the midpoint in 2010".
- WalterBright 5y ago
  Some of the European cathedrals took centuries to build. The people who started them knew they'd never see the result.
  - jhgb 5y ago
    St. Vitus cathedral in Prague: constructed between 1344-1929. That's not "centuries" -- that's a better part of a millennium.
TooKool4This 5y ago
I think the time-scale argument lacks a little bit of imagination. Time scales are very relative. There could easily be organisms that have lifespans in the thousands of years due to a variety of unique selective pressures (or no concept of a lifespan at all). Or species that have technology that can augment their lifespans to be thousands of years long. I can certainly see humans getting to that point in the next 10-thousand years. Given a thousand year lifespan, what is a few hundred years in terms of commitment to a project? It would be something akin to working a job for 10 years or so, which is not entirely unheard of. The ROI argument is more interesting! Although there are way too many factors to consider to have a definitive answer one way or another.
iso1631 5y ago
Once the technology exists to allow cults and misfits to build self sustatining asteroids and stick an engine on them, then it will. However as there would be no meaningful communication outside of a solar system, a single civilisation wouldn't spread to the stars