Hacker News Reader

Comment by danieltillett | Hacker News Reader

The big argument against intelligent life being common (and us not being unique) is the existence of stars. No civilization more advanced than us is going to let stupendous amounts of energy go to waste in the form of star light. If you assume von Neumann probes and Dyson spheres are possible, then star light is incompatible with the existence of advanced technology.

While life appeared very early in the history of the earth, technologically advanced life took billions of years to evolve. You require a planet with an extraordinarily stable climate to allow a technological society to evolve and I predict this is the filter.

Just one factor contributing to this stability is the need for a very large moon to keep the axis of rotation stable over billion of years. If the Earth didn't have its moon the Earth's axis would tip over creating massive climatic change. The more we look at the universe the more unlikely the stability of Earth's climate looks.

View on HN

What if they've got something more-convenient or powerful than going around messing with each star?

I just imagine an intelligent beaver: "There can't be any other intelligence out there, just look at all these rivers and streams!"

Meanwhile, humans are enjoying fusion-power and will take a little while to notice suspiciously-well-engineered beaver dams showing up everywhere.

Exactly. I like this analogy especially because our needs as a more advanced civilization are drastically different from the function performed by beaver dams. It's not even about energy per se. We could dam all the streams, but that's not a meaningful pursuit given our interests. Likewise, wondering about damming all the stars assumes that solar energy requirements are both the bottleneck and the ultimate purpose of advanced civilizations.

"We could dam all the streams, but that's not a meaningful pursuit given our interests"

Could you please explain that to the World Bank and governments in Africa, SE Asia, China, and Oceania that insist on damning the last free-flowing rivers on Earth?

Rivers are bigger than streams.

When it comes to hydroelectric power and titanic concrete structures, we look for very different water-situations than beavers do.

Humans dam lots of small streams too for agricultural and recreational uses. They don't get much news, but if you look at aerial surveys of places where people live you will see lots of small dams.

Live in the universe is necessarily finite, because the useable negentropy is finite. It is a plausible assumption that an advanced race would want to keep on living as long as possible. This means managing the negentropy as well as possible, which means dismantling stars and using them. So unless you disagree with my assumption, or add magical energy sources outside of known physics, starlight seems to indicate that there is no race capable of harnessing all available energy.

But now you're hypothesizing Aliens + some extremely convenient energy source.

In order to justify ignoring all the burning stars, it needs to be easier to tap than solar and much much more energy than all the stars will ever generate. I think that requires new physics.

Really the energy source must be infinite. If you can see that it will run out in a trillion years and you can survive for another billion by harvesting suns, somebody will.

We have nuclear power but still dam rivers, and I think beavers notice.

OTOH If there is a source of unlimited free energy, maybe some of those bright spots in space aren't stars at all.

Aren't you hypothesizing aliens + making human whims seem like universal goals? I don't think there needs to be an easier source of energy than solar to justify not tapping stars, just lack of desire.

This is kind of like how we currently have 8tb drives, but many systems are shipping with 128 or 256gb drives. In that case, the trade-off is for the speed of SSDs.

Or lack of ability if FTL is completely impossible relativity puts a limit on the size of a civilization. And the universe can easily out a time limit on it as well because extinction events can happen.

Even if we move onto mars a GRB would cook all of us.

And even if it's something considerably more boring then expanding to other planets or even star systems might not save a civilization from collapsing.

I hate to break it to you, but we have already dammed most of the rivers. Any beaver who thinks that there are no dams around is just one that hasn’t tried to swim far downstream :)

The argument is really more why is there only one technological species on the planet - the answer is the first one to arise will displace all the other potential competitors. This is exactly what has happened on Earth and why any other species that has a chance of creating a technological civilisation is not doing so well.

The same argument applies to the universe where the first civilisation that chooses to take over the universe will displace any competitors that choose not to. All it takes is one civilisation to spread and the whole universe goes dark.

Population of apes on Earth:

     Orang-Utan: 50,000
     Chimpanzee: 200,000
     Bonobo: 35,000
     Homo Neandertalensis: 0
     Homo Habilis: 0
     Gorilla: 100,000
     Homo Sapiens: 7,500,000,000

Yup. "The first civilisation that chooses to take over the universe will displace any competitors that choose not to."

How many zeros would you need to use to count the ants in the world?

Just one zero who makes many appearances.

I'd look at ant species rather than ants.

Note that ants don't directly compete for a niche with humans.

And megacolony species such as the Argentine ant are displacing other ant species.

1. Ants are not, technically, apes.

2. Ant biomass is about the same as human biomass on Earth.

3. Typical human mass is 60kg, typical ant mass is 3mg.

4. Do the math.

No species the size of ants can ever smelt steel. They are not competitors for humans to become Earth's technological primary species.

Have you read City? (Not that it's seriously making an argument the other way.)

Arguing that smelting steel is the key thing is the typical, we're good at this therefore this is what matters argument we've been making since time immemorial.

Humans can't sneak into other organism's bodies and take over their DNA (yet, to be sure) but viruses can, so humans just don't have a real shot.

No, for power we only bother to dam huge ones which no beaver could possibly take on.

Or dammed a few stars and decided it was ugly? Maybe watching other life emerge is their version of national parks?

> The big argument against intelligent life being common (and us not being unique) is the existence of stars. No civilization more advanced than us is going to let stupendous amounts of energy go to waste in the form of star light. If you assume von Neumann probes and Dyson spheres are possible, then star light is incompatible with the existence of advanced technology.

What about undetected energy aka dark energy.

> Just one factor contributing to this stability is the need for a very large moon to keep the axis of rotation stable over billion of years. If the Earth didn't have its moon the Earth's axis would tip over creating massive climatic change. The more we look at the universe the more unlikely the stability of Earth's climate looks.

I have heard this argument but I don't buy it compared to the need for asteroid protection. Things can get various like climate living underground/water. One of the critical things about our solar system that made life happen with out having a mega earth planet with crushing gravity is the presence of Jupiter and Saturn.

Jupiter and Saturn were in our solar system at the right place and right time. This theory is called the "Grand Tack" [1] and was recently covered in Scientific America.

Needless to say Jupiter continuously protects us from asteroids and various other objects from colliding with us on a continuous basis and not our axis. [2]

Continuous asteroid collisions vaporize oceans and destroy atmospheres along with other various debilitating things.

[1]: https://en.wikipedia.org/wiki/Grand_tack_hypothesis

[2]: http://imgur.com/gallery/DCLp8yc

Yes the presence and size of Jupiter and Saturn are another unlikely factor contributing to extreme stability of the Earth’s climate. Another one is earth seems to have escaped being exposed to a supernova going off very nearby [1].

1. https://en.wikipedia.org/wiki/Gamma-ray_burst

> The big argument against intelligent life being common (and us not being unique)

Those are two different things. "Common" is something up for interpretation, especially with the huge numbers concerned. Intelligence being uncommon could still mean hundreds of civilizations active in the galaxy right now. The term "uncommon" allows us to have somewhat meaningful discussions about probabilities. "Unique" on the other hand is an unsubstantiated, almost supernatural, claim of us being the only intelligence in existence anywhere.

> No civilization more advanced than us is going to let stupendous amounts of energy go to waste in the form of star light.

That is also a big claim. Yes, a civilization pursuing a maximum energy strategy would probably harness stars, at least from our current technological perspective that seems reasonable. However, very few people looking for signs of this would actually claim that all civilizations more advanced than us must do this. So originally at least, this hypothesis was proposed under the more reasonable premise that for purposes of detection it would be enough if one such civilization pursued this strategy. Again, the claim that all of them would necessarily do this seems needlessly extremist.

> You require a planet with an extraordinarily stable climate to allow a technological society to evolve and I predict this is the filter.

There is a good chance you're at least somewhat correct, but then again I can easily imagine planets with a way more stable history than Earth had.

A sample size of one is always going to be a problem, especially if you lack the means of even knowing what to look for. Making absolutist claims a priori seems to me like an argument to stop looking. Historically, claims of uniqueness and specialness have had a tendency to be wrong. It seems to me like a good default stance is to assume that whatever we see in front of us has a chance of occurring that is significantly higher than zero.

You don’t need most advanced civilisations to want to spread, just one. The first one that choose to spread will occupy the universe and capture all the energy that the stars are currently emitting into space.

The more we learn about other planets the more unusual Earth looks from a climatic stability perspective. There really is a good chance that Earth is unique in its level of stability - at least within the visible universe.

It's plausible that such civilizations are up against an unsurmountable hurdle. Perhaps they can't spread in an economically viable way?

If you look at the rocket equation, a slightly different planet would make launching rockets extremely difficult.

You have to postulate that all civilisations fail to spreed, not just some. On the economics once you can make the first von Neumann probe the cost of taking over the universe is zero.

I don't get this argument.

Let's just posit that every single advanced civilization ends up building lots and lots of Dyson spheres (which I'm totally unconvinced of). Now, can you prove that there are no such spheres out there? No, you can't. You can only prove that a whole lot of stars have no sphere, but maybe we suck at detecting them and there's one right next to us.

The only conclusion the argument draws if that advanced aliens are not yet nearly finished enclosing all the stars in spheres, and you might try to determine an upper limit to the rate of stars with Dyson spheres without us noticing.

The argument is once you have one civilisation that can create von Neumann probes [1] then that civilisation will spread through the universe wrapping each star in a Dyson sphere [2] that will capture all the light.

It only takes one civilisation in the universe to do this and all stars will cease to shine. The fact that we can see stars in the night sky means that they don’t have Dyson spheres around them hence there are not civilisations ahead of us - at least within the visible universe.

1. https://en.wikipedia.org/wiki/Self-replicating_spacecraft

2. https://en.wikipedia.org/wiki/Dyson_sphere

Sure, but are we sure these probes are instantaneous? What if it takes more than zero time to build a Dyson sphere? As I said earlier, you're only proving that they have been building spheres for less time than it takes to finish up the whole universe, which they say is big.

The probe’s spread and and dyson sphere building does not need to be instantaneous, just that it had to have happened in the past. For example, if we were to create von Neumann probes and have them spread across the galaxy, it would take less than 100,000 years for the probes to reach every star. When you compare this to how long it took for our civilisation to evolve on Earth it is a blink of an eye.

An interesting thought is if a new von Neumann spewing civilisation arose in the galaxy we would not have much warning before it reached us. The probes would be travelling just behind the light front and so would reach us only a little before the information that the stars between the probes origin and us went out. About the only way you could detect such an event is to look at very distant galaxies for examples that had spherical chunk taken out of them.

I should get around to writing this all up in a blog post because it is such an interesting topic :)

There was a quote yesterday in HN (either here or through wiki) saying it would take 500,000 years for von Neumann probes to spread throughout the Galaxy. Not meaningfully different on cosmic timescales, but still.

Also, your thought on the probes travelling behind the light front implies that they are going to be travelling at relativistic speeds. Is that realistic?

Yes they could be travelling just under the speed of light if the probes are very small and they are being accelerated by an energy source (say laser) at the origin.

But then it needs to break away (and actually brake) at the next system, replicate at outrageous speeds, and fire a bunch more at a heap more stars

If they are sent as a swarm you could bounce the laser off one to slow a second. This way you would not need to transport fuel. Also since you are aiming at stars you could use a hub and spoke model which would speed up the takeover process since you could bypass the build stage for most stars.

Cool. Interesting. Thanks for your thoughts

> About the only way you could detect such an event is to look at very distant galaxies for examples that had spherical chunk taken out of them.

There have been actual studies searching for the waste heat of Dyson spheres[0]. You basically image an "empty" part of the sky in the visual spectrums and look for infrared waste. (The studies turned up nothing but were also limited in scope.)

[0] - http://arxiv.org/abs/1408.1134

Yes I am aware of this study. From memory the temperature limit was 50K which no civilisation capable of building a Dyson Sphere would let leak out. Also the study was limited to our galaxy. The final issue is that if the von Neumann probes are moving at close to the speed of light we would not be able to detect these Dyson spheres anyway.

It would be really interesting to do a survey of galaxies looking for these spherical chunks. It should be possible to just use existing data and some machine learning to scan for possible candidates. If anyone is interested in doing this with me my contact details are in my profile.

This is a completely speculative argument.

It's like arguing about building a computer that can simulate reality. It just keeps getting bigger and more complicated and consuming more power and running slower as you learn about the new things it has to do, and soon the "possible in theory" becomes larger than the observable universe.

I agree with you and I also don't think that existence of intelligent life will necessarily lead to all stars being surrounded by Dyson spheres. Maybe there are other sources of energy that are easier to harvest for an advanced civilization.

However, in case there actually is a Dyson sphere around a star, it will probably emit energy in the form of infrared radiation and we should be able to detect it. Especially if there are billions of stars like that.

You can only detect a Dyson sphere in the infrared if the Dyson sphere is relative close into the star. If the outer sphere is at the orbit of Pluto then the emission would be close to the cosmic microwave background [1].

1. https://en.wikipedia.org/wiki/Cosmic_microwave_background

Where are you going to find the material to build a sphere around a star the size of Pluto's orbit? How are you going to arrange this material into energy harvesting equipment in a way that results in a net positive energy outcome?

The energy collector sphere only needs to be one atom thick and spaced just below the wavelength you want to capture. There is more than enough matter in our solar system to do this at the orbit of Pluto if you use a series of shells.

And you do this why? You're making the assumption that capturing all of a star's energy is somehow a good thing. Something worth devoting goodness knows what effort to. And whoops another comet destroyed a ludicrously big section of it.

There will, of course, be a fleet of space faring repair robots with infinite delta-v that can zip around the solar system and patch these things up!

Doesn't a pluto orbit sized energy collection unit leave a lot of energy on the table? Something about inverse squares..?

No. A. Pluto sized Dyson sphere is actually 99.5% Carnot efficient [1].

1. https://www.tillett.info/2016/08/03/carnot-efficient-dyson-s...

Dyson spheres would have a distinctive IR signature, which hasn't been observed. Particularly not at galactic scales.

Recent news on this via studies.

The study was limited to looking for inefficient Dyson spheres that leaked a lot of energy in the infrared. Efficient Dyson spheres would be emitting at just above the cosmic microwave background level and would thus be undetectable.

Evidence of absence is awfully hard to come by.

Significant evidence of insufficiently advanced Dyson Spheres seems somewhat compelling.

What's the internal temperature of an efficient Dyson sphere? Or conversely, what's the minimum size of such a sphere around various star classes to maintain a goldilocks-zone internal temperature? What are the mass requirements of such a sphere? How do the nonrotating elements of that maintain themselves against stellar gravity?

At some point in the thermal life of the sphere, it's got to maintain a steady-state flux of energy out to energy in. At a galactic scale, that's probably going to be detectable.

It's also quite possible that there are astral objects around which one would prefer not to build Dyson spheres. Black holes, say, or neutron stars, or binary or multiple star systems (a sphere around a single star is likely complex enough). Etc.

I've never found the Dyson sphere concept particularly compelling myself. Interesting concept, but challenging to pull off.

But then, what do I know, best I can claim is to have blown up a single planet. And that was with the assistance of alien technology.

>What's the internal temperature of an efficient Dyson sphere? Or conversely, what's the minimum size of such a sphere around various star classes to maintain a goldilocks-zone internal temperature? What are the mass requirements of such a sphere? How do the nonrotating elements of that maintain themselves against stellar gravity?

I think the confusion might be you are thinking about a classical Dyson sphere and not what would be built by a civilisation trying to most efficiently use the resources available. An efficient system would be composed of a set of independent (but co-ordinated) energy collectors arranged in a series of shells (for lack of a better name lets call this structure a Tillett Onion). The inner shells will be quite hot, but the outer shells will be very cold. The emission spectra of the Onion will be determined by the outermost layer, which if it is out near the orbit of Pluto, will be close to the background temperature of interstellar space.

If these collectors are coordinated they can use some of the energy they capture to maintain their orbit. These energy collectors don’t need to use much matter (they can be one atom layer thick) since all they are doing is capturing the energy released by the inner shells.

This really needs a more detailed write up than a post buried on HN :)

So, using some values off the Web, notably 0.0077g/m^2 for the weight of graphene, and computing the mass of a spherical shell at Pluto's radius, roughly -- 40 AU from a central star, I get (using GNU Units):

    You have: 4 * pi * 40au^2 * 0.0077g/m^2
    You want: million gigatonne
            * 86.618649
            / 0.011544858

How's that compare to the mass of one typical rocky planet?

    You have: earthmass / 86.6 million gigatonne
    You want:
            Definition: 68986.619

So, 1/70,000th an Earth Mass.

Turns out that there's not all that much carbon on Earth -- it's 730 ppm, but that's still a total abundance of about 50x what would be needed for our Pluto-diameter shell.

On the other hand, that means that at best, using an entire Earth's worth of carbon, you could create a Pluto-orbital-radius graphene Dyson sphere 50 atoms thick.

But Pluto's kinda way out there. What if we make the sphere the size of Earth's orbit instead. Surely that's going to save us a few orders of magnitude of matter, right?

Not quite. About 1.5 OOM. We go from ~90 million gigatonne to 2 million gigatonne. The difference between squares and cubes matters. So, on the positive side, you get about 2,000 layers of graphene, but you're still using up all of Earth's available carbon to do that.

There are more abundant minerals -- silicon or magnesium or iron, for example. But you'd still be using planetary-mass quantities for any substantial structure (or multiples of structures orbiting independently and forming a sphere).

Which is what Dyson's also proposed. And my thought is that a Dyson "sphere" comprised of, say, multiple flat sheets of stuff orbiting a star, would tend to have a few issues, such as solar wind, and occasional emissions from incomplete closure. You've also got the question of how you're going to move all that captured energy from where it's been captured to where you plan to use it.

And you still need a cold end to your hot end.

https://en.m.wikipedia.org/wiki/Abundance_of_the_chemical_el...

http://www.geek.com/science/geek-answers-what-is-graphene-15...

http://www.mathopenref.com/spherearea.html

http://hyperphysics.phy-astr.gsu.edu/hbase/solar/soldata3.ht...

> And you still need a cold end to your hot end.

That is interstellar space @ 2.7K :)

I actually just went through the calculations on my blog and to achieve a 99% Carnot efficient Dyson sphere (emitting at 58K) it would need to have a radius of ~7 AU - somewhere between Jupiter and Saturn [1]. There is plenty of mass to achieve that in the solar system and it would be also be undetectable in the infrared.

1. https://www.tillett.info/2016/08/03/carnot-efficient-dyson-s...

>No civilization more advanced than us is going to let stupendous amounts of energy go to waste in the form of star light.

Isn't this implying that any civilization more advanced than us must be able to take advantage of much more of a star's energy? Like say... humans in 6 months?

I see no reason for the band of possible civilizations between current humans and Dyson sphere building ones must be narrow at all. In fact, the vast majority of all possible civilizations might lie in that band, simply because it might be objectively impossible to actually build a Dyson sphere.

I did assume that it was possible to build a Dyson sphere, or some structure functionally equivalent. If you have von Neumann probes then Dyson spheres are by comparison simple.

Even a functionally equivalent system will have problems. For one, if you're absorbing a significant portion of the sun's radiation, that's going to generate a lot of heat that you've got to dissipate somehow. Then you've got to transport that energy in a way that's more efficient than the way the sun is already doing it.

Then you've got the logistics problem: it's easy to build a modern CPU (for example) if you're assuming all the infrastructure already exists. If you're assuming it doesn't, it takes much more energy to produce the first one than any subsequent ones. And it may be the case that in order to build an energy-positive Dyson sphere/von Neumann probe network you'd need to already have consumed a couple stars prior. And to do anything on that scale, you're already talking about coordinating activities across multiple planets or solar systems, which comes with it's own massive overhead.

There's a lot of ground to cover between burning fossil fuels and consuming stars, and little reason to assume a successful civilization must do the latter. You could potentially go quite a while syphoning fuel off a gas Giant, for example.

From what little I understand, a Dyson Sphere would require so much material in terms of heavy elements that you'd have a great deal of trouble building a significant number of them even if you were raiding the entire universe for these materials. But like I said, I understand very little (and would love clarification).

There are alternatives to a classical Dyson sphere (Dyson bubble is one) that can be constructed to capture all the energy released by a star and which do not require much material [1].

I actually think you would use a series of nested Dyson bubbles to capture the energy via a cascade much like the electron transport chain [2] uses a multi-protein redox cascade to capture the energy of metabolism.

1. https://en.wikipedia.org/wiki/Dyson_sphere

2. https://en.wikipedia.org/wiki/Electron_transport_chain

> No civilization more advanced than us is going to let stupendous amounts of energy go to waste in the form of star light.

Unless it is not worth the trade off. Material is rare and might be more useful for data storage than for harnessing energy. It's not difficult to imagine a civilization that consists of a single Dyson swarm, perhaps using the star as both an energy source and propulsion system. When they use up the star, they can move on to another star.

Detectable, but just barely, and only if we happen to look in the right direction.

Why can't they be brilliant and have no interest in harnessing that amount of energy for whatever purpose.

99.999% of all civilisations could have no interest in doing this, but all it takes is one in the universe and all the stars will be converted (over time).

"No civilization more advanced than us is going to let stupendous amounts of energy go to waste in the form of star light."

Any civilisation advanced enough to harvest dark energy isn't going to bother with backwards, inefficient starlight.

And any civilisation that has found the loopholes in the Second Law of Thermodynamics isn't going to waste time dealing with barbaric resources like dark energy.

If you assume the laws of physics hold even for advanced civilisation then they are not going to be able to find a loophole in the second law of thermodynamics :)

Even if you assume that dark energy can be harvested then any advanced civilisation is still going to be energy constrained - Bremmermann's limit far exceeds any energy that can be harvested from dark energy [1].

1. https://en.wikipedia.org/wiki/Bremermann%27s_limit

Advanced civilizations with a constructive proof that P=NP don't need all that much compute power.

> the need for a very large moon to keep the axis of rotation stable

First I've heard of this, do you have any more information?

Yes there is lots of information on this out there [1-3]. Really the moon is very, very, very unusual.

1. http://www.nature.com/nature/journal/v361/n6413/abs/361615a0...

2. http://www.astrobio.net/news-exclusive/the-odds-for-life-on-...

3. http://www.space.com/12464-earth-moon-unique-solar-system-un...

Could Black holes be Dyson spheres?

No. Black holes could be tapped as a source of energy though if you have an advanced enough technology.

Reply on news.ycombinator.com