Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Once we can see them, it’s too late

scottaaronson.com

Post reply on HN
Print view View original

51–60 of 114 posts

Re: Once we can see them, it’s too late

#51

Post by eloff »

ndjtjgkglf wrote: post #47
ivalm wrote: post #18

This kind of assumes that a “maxed out” civilization can expand at speed of light, but I don’t think it is given. It may very well be that meaningfully massive transport cannot exceed even a small fraction of c . After all, propulsion has to make energetic sense. 1. There is energy to get to speed 2. Mass of fuel to decelerate (and you have to accelerate that fuel in the beginning!) 3. Highly blue shifted CMB (that a…

In theory you can manipulate atoms and molecules at a distance with light. So you can use a laser beam to construct a 3D printer at a distance. Rinse and repeat, and this is asymptotically expanding at c.

A laser beam expanding out over light years is not a laser anymore and definitely not atom sized.

Re: Once we can see them, it’s too late

#52

Post by dsr_ »

btilly wrote: post #32
isoprophlex wrote: post #3

This is the most profound thing I've read in a while. I'm tempted to whip up some numerical simulation to verify this to some extent... > But here’s the interesting part: conditioned on all the steps having succeeded, we should find ourselves near the end of the useful lifetime of our planet’s star—simply because the more time is available on a given planet, the better the odds there. I.e., look around the universe a…

Want something scary? The form of reasoning used is exactly that in the https://en.wikipedia.org/wiki/Doomsday_argument for why humanity is unlikely to last long, and almost certainly will not have a future where we break out and colonize most of the Solar System, let alone the stars. If you believe both arguments, the most likely outcome is that there is one more step on the way to Robin's argument. And that outcome…

It's just as likely that we all greet our future AI grandchildren, as they will be us.

Re: Once we can see them, it’s too late

#53

Post by chronolitus »

I like to imagine that this is how things would look if we entered a Karadashev type III civilization's light cone:

One day, in a small quadrant of the galaxy, we'd notice starts "shutting off", simply going dark. After a few months, our astonomers would confirm their new model: the amount of stars going dark is increasing. Days later: the rate of increase itself, is increasing. One day we'd wake up, and a small portion of the night sky would be dark. A disk of darkness, right at the brightest part of the galactic plane. By then we would have figured it out of course: most stars in the galaxy were being turned into dyson spheres. They were spreading almost as fast as light, making it look almost instantaneous, even though the process had taken hundreds of thousands of years. In a few thousand years a large portion of our sky would be dark. And after that...

Re: Once we can see them, it’s too late

#54

Post by ndjtjgkglf »

eloff wrote: post #51
ndjtjgkglf wrote: post #47

Earlier quoted context omitted.

In theory you can manipulate atoms and molecules at a distance with light. So you can use a laser beam to construct a 3D printer at a distance. Rinse and repeat, and this is asymptotically expanding at c.

A laser beam expanding out over light years is not a laser anymore and definitely not atom sized.

You are assuming present day lasers. An advanced civilization will individually control the photons.

You also don't really need atom resolution anyway. You can build mechanical computers driven by rolling stones by abblating matter with powerful beams - lithography on continent size scale, etc.

Re: Once we can see them, it’s too late

#55

Post by hagmonk »

chronolitus wrote: post #53

I like to imagine that this is how things would look if we entered a Karadashev type III civilization's light cone: One day, in a small quadrant of the galaxy, we'd notice starts "shutting off", simply going dark. After a few months, our astonomers would confirm their new model: the amount of stars going dark is increasing. Days later: the rate of increase itself, is increasing. One day we'd wake up, and a small port…

This is approximately one of the opening ideas in the Commonwealth Saga by Peter Hamilton. Highly recommended!

Re: Once we can see them, it’s too late

#56

Post by throwaway-571 »

rossdavidh wrote: post #44
skybrian wrote: post #25

It seems like the argument gets a lot less interesting if the most-expanding interstellar civilizations don't spread at anywhere near the speed of light? It's easy to think of reasons why this might be the case. There is acceleration, deceleration, and replication time, which seems like it would be substantial for any civilization that doesn't want to put most of its resources into replicating as fast as possible.

Ah, the classic mistake in a game of "Risk"; expanding as fast as you possibly can, leaving too few resources (troops) at your border to prevent a rapid collapse. If a fast-as-you-can civilization meets a slowly-expanding civilization, it is not at all clear to me that the former would prevail.

In rsk everybody is at a roughly similar technological level. it really depends on things like technological level difference (ask the mayas and aztec) and philosophy, and if you're expanding as fast as you can, you can probably mobilize resources rather easily. Then you have philosophical difference: are you expanding to make friends, or just to gather resources for the hive queen?

Re: Once we can see them, it’s too late

#57

Post by eloff »

ndjtjgkglf wrote: post #54
eloff wrote: post #51

Earlier quoted context omitted.

A laser beam expanding out over light years is not a laser anymore and definitely not atom sized.

You are assuming present day lasers. An advanced civilization will individually control the photons. You also don't really need atom resolution anyway. You can build mechanical computers driven by rolling stones by abblating matter with powerful beams - lithography on continent size scale, etc.

Even if you individually control the photons, I'm not sure that's possible. Matter still behaves like a wave at quantum levels. I don't know enough to say with certainty. This is all highly speculative.

Re: Once we can see them, it’s too late

#58

Post by nojokes »

Gupie wrote: post #9
isoprophlex wrote: post #3

This is the most profound thing I've read in a while. I'm tempted to whip up some numerical simulation to verify this to some extent... > But here’s the interesting part: conditioned on all the steps having succeeded, we should find ourselves near the end of the useful lifetime of our planet’s star—simply because the more time is available on a given planet, the better the odds there. I.e., look around the universe a…

But we had a billion years left. We have plenty of time to do whether we are going to do before the "planets clock runs out".

and we just getting started.

Re: Once we can see them, it’s too late

#59

Post by TeMPOraL »

lisper wrote: post #41
ctdonath wrote: post #30

Earlier quoted context omitted.

Think “matter scanner/printer”. Scan an object, at atomic & energy vector level. Transmit that information to a destination’s printer. Expansion is then indeed at speed of light, capped only by need to physically move a small “seed” printer (rapid transport may have enormous energy cost, but that’s all that needs moving; everything else is just data). Having seen tech go from digital imaging to 3D organ printing in a…

But first you have to deliver the printer to the destination somehow.

Here's an idea.

Imagine you have a way to build von Neumann probes that can boost themselves to some high fraction of c (say, 0.9). You start pumping them out in your home system, and send them towards some stars - not necessarily the closest ones. You keep sending them. There's now a shell expanding at 0.9c that's shaped by your production capabilities. Of course, the probes will spend time accelerating and then decelerating towards their destinations. But even as they decelerate, other probes, going to further targets, pass ahead of them.

Any probe that lands in a system sets up a facility for building and provisioning more of itself, and sends those to further star systems, adding to the wave. Meanwhile, any technological advancement could be transmitted at the speed of light - so the probes receive improvement instructions as they fly, and the entire shell keeps upgrading itself as it expands.

What's the aggregate speed of that shell? Less than 0.9c, I think, but I'd say it's still a high fraction of it.

The main warning sign we could look for, assuming there are no crazy unknown physics allowing creation of silent drives, are deceleration plumes and associated thermal emissions. But then, civilizations like ours should be easy to spot from a distance. Don't bother with radio waves, just look at the spectral lines of the planets in a system; if you spot high concentrations of chemicals that don't have geological rationale to be there (such as oxygen in our case), this implies something is keeping the planet away from equilibrium - likely some form of life.

If we go by Dark Forest theory, the standard procedure upon detection of such "life signs" would be to send updated instructions to the von Neumann probe heading towards that system: "don't decelerate, ram the planet (or the star) at 0.9 c". If there's anything of use left there, another probe will come later and consume it.

So, by the time we spot the evidence of the von Neumann probes, we may just be about to get hit by a relativistic kill vehicle at the vanguard of the expanding shell.

(I'm tempted to build a toy model and play with the parameters to see how much of an advance warning we would have.)

Re: Once we can see them, it’s too late

#60

Post by tsimionescu »

Pxtl wrote: post #19
bena wrote: post #11

Earlier quoted context omitted.

Fermi's paradox. It assumes that technology will get to the point where it becomes trivial to explore space. That's not an easy assumption to grant.

Trivial and possible are the same. If colonization is possible it will eventually be mastered and then you get geometric growth. Covid has given us all a quick refresher course on the nature of geometric growth. On the timescale of the universe, geometric growth in colonization is functionally hyper rapid. So either colonization (and thus, geometric growth) is impossible or there's nobody else out there.

Well, if establishing colonies and moving from one star system to the next is possible, but takes a few hundred million years each time, then it wouldn't be surprising in any way if even the 'first' civilization to arise has barely colonized a handful of star systems. Perhaps in a few hundred billion years the universe will indeed be full of colonies.
Post reply on HN