[codethief]

> The single most important assumption in this paper is that energy consumption will increase by 2% per year. […] How much energy will we consume in 1,000 years

I keep posting this link here on HN but, once again, it seems very appropriate:

> The upshot is that at a 2.3% growth rate (conveniently chosen to represent a 10× increase every century), we would reach boiling temperature in about 400 years.

https://dothemath.ucsd.edu/2012/04/economist-meets-physicist...

[jl2718]

When the nuclear apocalypse has passed, and the last remaining humans are living underground on mushrooms, somewhere an economist will claim growth using a model updated to include the work done by ants, and somewhere a bureaucrat will update the tax code to collect on it.

[casparvitch]

If HN was reading Murphy the front page would look very different every day

[istjohn]

A global cooling system is easily conceivable. The ISS uses radiators to dissipate heat into outer space. We could do the same. We could concentrate heat with heat pumps and pump hot steam, molten salt, or plasma up a space elevator and radiate the heat away.

Edit: The link points out that in 1,400 years we'd be using energy at the rate produced by the sun and in 2,500 years at the rate of the entire Milky Way. Even if we solved the heat radiation problem, it seems unlikely we'd be able to obtain fuel for our fusion reactors at a sufficient rate given the speed of light and the density of matter in the universe.

[amluto]

There’s no free lunch.

The sun radiates heat like a blackbody sphere with the surface area of the sun and a surface temperature of 5600K or so.

If the Earth were to have a fancy heat pump that radiated the same amount of heat into space, it would need the radiating area times the effective temperature to the fourth power equal to that of the sun. The real killer problem is that your heat pump is subject to the Carnot efficiency of pumping heat to that same temperature.

By the most straightforward Second Law calculation, entropy transferred to hot side >= entropy removed from cold side. So Q_hot / T_hot >= Q_cold / T_cold. Q_cold is the heat removed from the pleasant ~300K place where the humans are. At 5600K on the hot side, 18.6 times as much heat needs to be radiated out, so for every bit of useful work done on Earth, at least 17.6 times as much energy is consumed in cooling. If you want the radiator to be only half the surface area of the sun, multiply that by about 16.

What future humans really need is a Dyson sphere with absolutely enormous radiating area. :)

[anon____]

Well then maybe we should do this at night, i.e. always on the dark side of the Earth, not the one that's facing the Sun. Basically passing through the radiation that reaches us.

[pyinstallwoes]

Like space itself

[naikrovek]

we can (and do) radiate heat into space already. there's a specific infrared wavelength that passes out of our atmosphere unimpeded.

building things which receive sunlight but which emit IR at this wavelength would have a cooling effect, though I don't know to what scale.

no need to pump anything up to space, lol. just put hot stuff inside vessels which are painted with a material which radiates heat at this specific wavelength.

[AstralStorm]

Uh, paint doesn't radiate heat not does it convert IR frequencies. (except for very unusual, usually fluorescent materials) It reflects incident IR.

[naikrovek]

not normal paint, no.

thanks for nitpicking; always appreciated

[bmitc]

> A global cooling system is easily conceivable.

That sounds like a line from Dr. Strangelove by the titular character with ol' Bucky following up with "Mr. President, we should not allow a heat pump gap!".

[nvm0n2]

That's not a great argument by the physicist, the economist definitely won that debate. The physics guy only seems to realize that though when he wakes up the next day and has calmed down a bit.

Although the link is useful for the thermodynamic calculations, there are two major problems with the argument as presented:

1. Right up front the physicist arbitrarily bans space travel. The economist, being an agreeable man who'd probably rather be making smalltalk with a pleasant member of the opposite sex rather than defending his whole profession to a bolshie physicist, accepts this limitation, but he shouldn't have done. Nothing in economics is predicated on a space travel ban. We are already obtaining economic growth from space via satellites and that era has barely got started. None of the physics arguments work if you make the relatively small leap to putting factories, power plants etc on moons, asteroids, space stations or other planets. This doesn't require colonization assuming progress in robotics.

2. Much more seriously, the physicist doesn't understand what growth or wealth mean in an economic context. The economist tries patiently to explain this to him many times, and he just doesn't get it. This is a very common problem when talking about economics because people aren't used to the expansive definition of wealth economists use, so often conflate it with other things like money or (in this case) energy.

You can increase wealth indefinitely even with a stable population and stable energy/resource usage. This isn't controversial or weird, it's just part of how wealth is defined. The VR example is one attempt, dessert another attempt to explain this to him, but he just doesn't get it until the next day when he suddenly has an epiphany but decides it wasn't his fault because he personally distinguishes between "growth" and "development". No such distinction is recognized by actual economists for valid reasons. But you don't get to claim there's a problem with economics just because you failed to understand the lingo of the field.

[dTal]

If that's "wealth", then "wealth" is worthless.

When you have a capped number of humans, and every last one of them is jacked into the Matrix and has everything they could possibly want - what is there left to grow? If the growth in "wealth" doesn't actually reflect an improvement in human experience, then it's a pointless term. Sure you can have two computers trade virtual tokens at an ever-increasing rate until you saturate the network cable, but calling it "wealth" is an exercise in semantic subterfuge.

It's like saying economic growth is unbounded because you can always print more money. Eventually it has to bottom out in something real.

[nvm0n2]

I think you're arguing at cross-purposes with me. Nobody has argued that bouncing virtual tokens around creates wealth. That's the "wealth is money" fallacy that I just criticized.

Wealth as used by economists just means all the goods and services that we provide to one another, sometimes with the addition of "services" like a clean environment. It's very broad and includes things like cultural wealth, and doesn't even have to involve selling something.

That's why printing money doesn't create economic growth and nobody claims otherwise. In fact in the debate they specifically agree to discount inflation to avoid it getting in the way.

[dTal]

The point is that there's no such thing as a (meaningful) good or service that doesn't require some amount of a tangible and finite physical resource - human time, if nothing else.

The economist's position - and yours, apparently - is that "growth", whatever it is, can be sustained literally infinitely, on a finite rock amongst a finite group of hairless apes where nothing about the actual situation is infinite. So you can either have a highly abstract definition of "growth" that allows this to be true, or a definition that most people would recognize as meaningful or positive, but not both.

[nvm0n2]

Be careful here - the economist's position about a finite world was something the physicist picked. The economist agreed to it early on, probably to try and seem agreeable, but it's not a reasonable concession to make. Space exists, we put things there today, presumably will put more things there in future. So there is no finite world.

But let's make the same assumption for a moment. For economic growth [of wealth] to stop requires two things to stop: population growth and productivity growth. As GDP is roughly population * productivity.

Clearly, populations can stop growing or shrink. We can also assume a finite population limit. So this is an argument that productivity growth is also finite. But, why should that be the case? Take the example of computers. Modern computers are much more productivity enhancing than older computers, but they are also smaller (i.e. less physical resources needed) and more energy efficient. Even if the Earth had reached carrying capacity, smarter chips would continue to be designed and the productivity boost of computers would keep increasing. That's just one example, there are many others.

But again, the finite world assumption doesn't hold. So the whole debate is a bit of ivory tower silliness anyway.

[dTal]

1) Space doesn't help you, it only postpones the inevitable. Eventually the ravenous maw of your exponentially growing civilization will run out of stars to harvest, trapped as it is inside a light bubble growing at 1c, its volume expanding with mere cubic growth.

2) You've just relabeled the discussion from "wealth" to "productivity". In any case, scoring some efficiency wins here and there is once again a game with physics-imposed limits.

I repeat my question: when all humans are fully fed, housed, maximally entertained and satiated, every dopamine receptor firing on all cylinders - and you can't make more humans - what exactly is there left to grow? At all, let alone exponentially?

I confess I find the notion of infinite exponential growth of anything, let alone things humans value, so patently and obviously physically unsustainable that I am deeply perturbed by these earnest efforts to defend it. I engage in the spirit one might engage Flat Eartherism, as an intellectual exercise in probing the manifestly absurd - except that this dogma apparently pervades mainstream economic thought, an observation that should terrify anyone who wants civilization to survive.

[joak]

This boiling temperature conclusion makes the assumption that we continue using thermal power (steam engines, etc...), where waste heat is around 60%.

However photovoltaic and wind does not produce much waste heat. Arguably solar and wind cannot scale 1000x but then you could have non thermal fusion like Helion's https://www.helionenergy.com/technology/.

Btw, thermal power is already showing limits (rivers overheating in summers), we don't have to wait 400 years to see its failure.

[codethief]

> This boiling temperature conclusion makes the assumption that we continue using thermal power (steam engines, etc...), where waste heat is around 60%.

No, it doesn't. The waste heat, at the very end of the day, is always rather close to 100%. I.e. we use all that electric energy we generate to power computer, fridges, and many other machines, all of which – sooner or later – convert that electric energy to heat. (And, well, maybe a bit of chemical binding energy, depending on the application. But then, a few decades later, those products of our work will usually fall apart and/or are being burnt or torn down.)

> you could have non thermal fusion like Helion's

No, you can't. As the author says:

> this statement is independent of technology. Even if we don’t have a name for the energy source yet, as long as it obeys thermodynamics, we cook ourselves with perpetual energy increase.

Free free to have a look at his other article (the one he links in the paragraph above) for some more details -> https://dothemath.ucsd.edu/2011/07/galactic-scale-energy/

[XorNot]

100% of the heat we get from the sun is currently lost as waste heat, by one means or another. So you're not boiling the oceans by any reasonable projection of energy usage increase of solar energy.

(yes yes some of it is reflected back into space as non-IR light, but you can also lose IR-emissions back into space without heating the Earth as well).

[codethief]

Most of that incoming heat is reflected / radiated off actually. It doesn't "stay" on Earth.

https://en.m.wikipedia.org/wiki/Stefan%E2%80%93Boltzmann_law

[XorNot]

That is literally what I said.

[codethief]

My apologies then, I understood your comment differently!

[midasuni]

Where does the energy go then? EM radiation leaving the planet?

[joak]

Chemical energy. Like the plants, they take sun energy and store it as chemical energy. We could do the same, take CO2 and water and make CH4 and oxygen (electrolysis + Sabatier reaction) and make plastic. Or take ore, aluminum oxide and iron oxide, and "unburn" them releasing again oxygen (we already do that) etc...

We mostly use energy to manufacture things.

Life takes energy and lowers the entropy of the planet. Why shouldn't humans able to do the same? Any technical reason?

[AstralStorm]

There are limits to doing this, which is why we find it really hard to, say, remove CO2 from the atmosphere.

The technical reason is called entropy. Diffuse heat is hard to concentrate and use, much like gas outside of a container.

[codethief]

> Life takes energy and lowers the entropy of the planet. Why shouldn't humans able to do the same? Any technical reason?

Reducing entropy necessarily produces waste energy + entropy elsewhere. So the technical reason this cannot be done at the planetary scale is called Second Law of Thermodynamics.

Yes, we could try to put all that excess entropy & heat into space but there are limits to that (Stefan-Boltzmann law, among other things).