path: root/transcripts/breaking-bitcoin/2017/solar-powered-space-pirates.mdwn

Solar powered space pirates: A threat to bitcoin?

petertodd

video: <https://www.youtube.com/watch?v=0WCaoGiAOHE&t=1m33s>

<https://twitter.com/kanzure/status/1005633917356007425>

Let's see. How do I make this fullscreen? Alright, anyone who can actually read French know how to make this fullscreen? My french teachers in high school would be very disappointed in me.

My name is Peter Todd. I do cryptography consulting. I've been involved in Bitcoin Core as well. I want to tell you about whether solar powered space pirates are a threat to bitcoin.

If you're a scifi and space nut to me, you might have heard about collecting solar power in space. You put up a bunch of solar panels up there, and you beam your power back to earth and run civilization. Or if you're crazy enough, you, apparently that's the sun on the left and the squiggly lines are charged particles getting collected. It's space- there's energy all over the place, and you don't have the EPA to worry about, and you could run nuclear reactors and just dump your radiation into space. There's a lot of it, after all.

The issue with this is the power transport. What are you going to do with these kilometer square arrays of solar panels? I really love how they say- the main technical deterrant to this proposal of collecting ions from the sun is getting the power back to the earth. I thought this was a hugely optimistic statement. The problem wasn't building it, it was getting the power back to earth.

Then I remembered we got this guy, and he's a rocket in more than one way. I've been watching expanse, the tv show. Obviously, everything is possible, besides having decent labor relationships between the poor and whatever.

Max Keiser suggested something clever. He suggested this in response to the Blockstream satellite thing- not sure he understood what he was talking about there, but if you do have this solar panel stuff, and you have power in space, then the easiest way to get the money back that it represents, is in terms of bitcoin. This is something that happens right now. You can get things called stranded hydropower or stranded wind power or stranded any power. You have some power generated in some place, but not the infrastructure to get it off the grid to where it can be used. This is common in China where you built a hydroplant 50 years ago for a mine that was never profitable. But why not load up some bitcoin miners and load them up and run the mining there? If you solve the challenge of solar power in space, adding the bitcoin aspect to it is actually pretty easy.

If we're going to talk about the effects of this, let's start with Blockstream satellite. You might have heard of it. It's being beamed through you right now. I shouldn't say repurposed, but rented geostationary satellites used for broadcasting TV back to earth. It's broadcasting blocks. Long story short is that, they have ground stations, they send blocks up to the satellite, and then they get bounced byack done. Anyone with a cheap sat receiver can get this.

So why not imagine our simplest use case? I'm a space pirate and I'm going to decide whether to put my solar panels up within that geostationary range, right? Well, the question for me is whether it's profitable. There's your geostationary orbit- and being a space pirate, you have to do the figures approximately, it's not like it's rocket science or anything. Well, it is rocket science, but the easy kind. Say 48000 km. The speed of light is set at 300km/sec... and if I go find a block, I want to get my block to everyone else before someone else goes and finds a block. Because after all if they find a block and I'm just some little guy with my little space barge without that much hashrate, chances are they will find a nother block after me. This latency number is very important. The latency from pretty much anywhere on earth and back up again is 0.3 seconds, which is quite fast when you compare it to fiber. The reason is two things- in space, you cna go in a straight line. Whereas if you look ta fiber maps of where they put the cables in the ocean, they squiggle around all over the places, because they have to go around undersea mountains. The speed of light is 70% slower in fiber, compared to vacuum. It's kind of competitive.

More to the point, let's do the math. How much money am I going to earn relative to a miner with a faster internet connection? We know that the blocks are found as a poisson process. I'm around a bunch of french people- I hope I pronounced that right. We could look it up on Wikipedia. When n=0 for this time interval out of the block interval which is T. My time for the latency is little t, my time for the block is big T. I'm just a space pirate, I don't know how to do this E thing, why not plug it into Wolfram Alpha instead? And there's actually a good reason to do this, which is that, intuitively, if I know the blocks are getting found once every 10 minutes, then surely if I take a tiny slice of that, then the amount of blocks I should find in say one second versus ten minutes should be kind of linear because the reason why the poisson process gets weird is because what's the probability of me not finding a block in 60 minutes? Thta's based on the probability of me not finding the block in the first 10. Whereas when we're talking about short intervals, we should expect it to be linear. When we plot e^x, and when the exponent is small, it looks like a straight line. You can approximate this pretty easily. You can feed in more values into Wolfram Alpha to check this. How can you approximate it and make it simple? Well now that it's so simple, and I'll tell you that when you do this, in this small interval, it's a pretty close result-- so 0.3 second, the block interval is 600 seconds, and now, this is saying, what is the probability of nobody finding a block during this time T from their block getting up to me and then back again? It's 99.94%. So that means 99.94% of the time that if I find a block in that interval then I wont be orphaned. This sounds pretty good, right? Chances are that I have othe rexpenses that matter more than this 0.06%, like maybe spcae suits are expensive and suits are expensive and I have to buy all that oxygen to breath.

But hang on a second. Why would I restrict myself to geostationary orbit? Why would I go put my mining so close to earth? I looked it up, and NASA has planned closest probe to the sun, that is actually a relatively accurately photo, it's about 10 diameters of the sun away. That's really close to the sun. If they can put a probe there, chances are tha tI can put solar panels there. Maybe with more radiation coming on them, I can mine for cheaper than the poor guy who has to put solar panels way out in earth's orbit. It'scertainly possible.

If that guy's operation is cheaper, what happens with latency? Earth to the sun is like 150 million kilometers. So I plug this in with speed of light delay- that's like 1000 seconds. That's a good deal more than the block interval. If you go put this curve a little further, Wolfram Alpha is saying I can't use such a simple approximation. It's so long that chances are that someone else is going to find a block. So surely my space pirate life baking outside the sun is not going to work out and I wont make any money.

But what if I'm some major organization- maybe someone who can put enough hashrate in one place that this all flips? Remember the assumption- the beginning equation assumed that I had negligble hashrate. But what if I'm Wakanda and instead of digging up bioweapons that spit out acid or whatever, I could go put up a majority of hashrate in space. In that circumstance, it's not me that has to worry about that latency, it's earth. It becomes their problem, not my problem. My hashrate is all in one place, so I don't really care.

The issue with that too is that again like, even if I'm the space pirate and I want to head off to the sun... well, that orbit, is pretty big. Even that close to the sun, it's like 12 million km. Let's look at the speed of light delay. They wont be happy if I position my equipment right next to them. What if I position them on the other side of the orbit? That's 80 seconds of extra delay. How likely is it for someone to get a block in 80 seconds? It's pretty likely. So I probably don't want to spend 10% of my income there. At the very least, I might be incentivized to stick my equipment closer to where they are... which is not good for bitcoin, because what happens if there's a solar flare? Do we want all the bitcoin hashrate being knocked out by one event? But that's what the incentive is.

So we have an interesting conclusion here. Which is that decentralized consensus only works if it's centralized. What do we mean by centralized? In terms of the light cone, if you will. Are you close enough to your othe rparticipants in this "decentralized" consensus to come to consensus in a reasonable amount of time? After all, we could go tweak the parameters and make the entirety of our solar system essentially be very close to each other in relation to the block interval. But I'm not going to do a hard-fork just because I want to do space mining. Myabe there's an issue here.

If I'm some new altcoin designer and I say well 10 minutes is a long time for a retail transaction and I turn the knob down to 15 seconds... maybe it's not space pirates I'm worried about, maybe it's people behind the great firewall of China that I'd be worried about. The equations don't care how far you are away- they only care about the ratios.

I'm going to leave you with that basic lesson on latency. Thank you.