Explain Like I'm Anti-Federalist: What is Bitcoin mining?

[ClaytonB]

From the image thread:

I have tried to wrap my mind around it, but I just cannot comprehend how one "mines" cryptocurrency.

OK. Here's a non-technical summary, and I'll answer some objections below.

Imagine you want to keep a single ledger, distributed between 1,000 people who live in different countries all around the world and none of whom trust each other. Imagine there is some black-box device that acts like a lottery machine, except that, every time you pull the handle, you either win a jackpot or nothing. The odds are set so that you have to pull the handle hundreds of thousands of times before you win a jackpot. Whenever one of these black-boxes hits a jackpot, all the other machines notify their owner that the jackpot was hit, and who hit it. Whoever hit the jackpot is given the right to write something into the ledger. Whatever he has written into the ledger is also broadcast through the black-box to all other other black-box owners, so they can update their copy of the ledger, too. The black-box itself is unhackable, meaning, it's impossible to "reverse-engineer it" and devise some strategy to trick/hack the system.

The more black-boxes you own, the more lottery tickets you can win, through sheer volume of lever-pulls. If you can make 10,000 lever-pulls per day with one black-box then, with two black-boxes, you can do 20,000 lever-pulls per day, and so on.

---

Objection: I was asking about miners and you started talking about black-box lottery machines? How are these related?

Answer: What we call "mining" is not like ordinary mining except in the sense that you have to expend a lot of real resources in order to find a very valuable resource. Rather, it is more like a lottery. Everybody around the world who is mining is just pulling the lever on their lottery-machine (their miner), zillions of times per second. Out of all the miners in the world, the first one to get a winning combination is the one who will get to add the next block to the blockchain (make the next entry in the ledger). The network has a sliding-scale of difficulty to keep the rate at which miners are mining to about one block every 10 minutes, on average.

Objection: Can't I just build an insane number of black-boxes and earn all the new money in the Bitcoin network?

Answer: Yes, you could earn most of it if you bought/built enough miners, but the real-world price of a miner is never zero. In addition, competition ensures that the market price of the miners will tend to go to their breakeven value. So, you can buy as many as you want, but the more you buy, the higher the market price will go, and the more unprofitable it will be for you. Remember, the first miner you buy will only just breakeven... everything else is just downhill from there.

Objection: When I get a winning lottery ticket, what prevents me from making an entry in the ledger that just transfers all Bitcoins to myself?

Answer: Every transaction in a blockchain block is independently checked by all miners and all full nodes (who are observers on the network that help with maintaining unprocessed transactions). Each transaction must be "signed" with a cryptographic key that acts just like a signature on a bank check -- if you try to fake a signature, the fraud will be detected, and the block will be rejected by the network, so all of those lever-pulls you did in order to get a winning lottery ticket will have been wasted for nothing. The miners have maximum interest in ensuring that the blocks they mine are fully compliant.

Objection: Isn't there an easier way to do all of this? Do we really have to burn a third-world country's worth of electricity just to maintain a simple ledger?

Answer: It is conceivable that there are other, better solutions to the problem that the blockchain solves (distributed trust without a central enforcer), but we don't know of any, as of 2022. The value of the blocks in the blockchain are being arbitrated against their true market value through the costs of mining (including both equipment overhead and operational costs).

Objection: Is mining really that hard? Couldn't Tesla just build some kind of mega-mining computer that is 1,000x more efficient than any other alternative and just soak up all the mining in a single monopoly?

Answer: It is mathematically possible that mining is easier than cryptographers believe. If it were to turn out that Bitcoin's proof-of-work function is flawed, there are other fallbacks such as Litecoin and Ethereum, which use different functions. These functions have computational properties that today's academic cryptographers do not believe to be cracked. What this means is that they believe there is no faster way to compute these functions than to simply perform the function "naively", which is what forces miners to make countless "guesses" at the correct answer to the cryptographic puzzle, thus performing "work". Quantum computers capable of operating with large numbers of ideal qubits (100+ ideal qubits) could endanger some aspects of these networks and would force an upgrade to "post-quantum" cryptography. Another threat is that a brilliant mathematician could just prove that P=NP which would be a nuclear strike not only all crypto-currencies, but all secure communications, including banking, finance and government. In short, no, Tesla cannot gain any significant advantage that the existing miner manufacturers are not already leveraging, because the nature of the proof-of-work function simply requires a certain number of bit-flips to be performed (many quadrillions of them, on average) in order to mine a block.

---

Try the following as a visualization of the mathematical mining function in Bitcoin. Below is a Galton Board. Just play the video and you will understand what it does (easier than explaining).



Notice how most of the balls fall in the center of the board. You can think of a single miner on the network as doing exactly this. Every second, it is pouring zillions of little balls down a completely randomized transfer-function (similar to the quincunx at the center of the Galton Board). In order to "win" the lottery, however, it has to get a ball into the furthest-left (or furthest-right, just pick a side) slot. We can imagine the board as being much, much wider than this board, so that the probability of getting a ball to bounce waaaay over to the furthest end is extremely tiny -- not zero, but extremely close to zero. And no, as far as we know, there is no way to "tilt" the board or otherwise rig the system. You just have to pour lots of little balls and hope to get one that bounces into the furthest edge slot.

[acptulsa]

Selling computer time and capacity to people who run big, fat calculations.

[Indy Vidual]

Selling computer time and capacity to people who run big, fat calculations.

Exactly, "mining" simply helped Bitcoin grow the "digital gold" branding.

[Anti Federalist]

I'm assuming that someone, somewhere "wins" real money.

Where does that "jackpot" come from?

[acptulsa]

Selling computer time and capacity to people who run big, fat calculations.

I'm assuming that someone, somewhere "wins" real money.

Where does that "jackpot" come from?

Who's doing the calculations? Presumably, they're doing them for a reason. Are they doing virtual wear and tear testing on a new product? It'll be on the market. Are they doing research? The money is in their grant.

[ClaytonB]

I'm assuming that someone, somewhere "wins" real money.

Where does that "jackpot" come from?

From Bitcoin's creation in 2009 until about roughly 2140, there is a scheduled payout of "free" Bitcoins to miners. This is called the block reward or block subsidy. At first, it was 50 Bitcoins per block (hard to imagine nowadays) and this gets cut in half roughly every 4 years -- 25, 12.5, 6.25 (current payout), 3.125, etc. In 2140, the final payout block will be awarded and there will be a grand total of exactly 21 million Bitcoins. Here is a chart showing historical payout versus price.

Another way to think of it is that there is just one "mega-Bitcoin" which has been sliced into 21 million pieces and is being paid out little-by-little with each block mined. As you can see here, the vast majority of Bitcoins have already been mined (about 90%), and the remaining mining is just targeting that last 10% of Bitcoins that will ever be paid out to miners. So the "inflationary phase" of Bitcoin is already in the rear-view mirror. From now until 2140, your Bitcoins will devalue a grand total of 10% as a result of the block reward. So, compared to any fiat government currency, Bitcoin is hard money.

The block subsidy incentivizes mining while the network grows. As the subsidy rapidly approaches zero, the miners will be working for transaction fees, rather than the reward. Here you can see a chart of transaction fees as a percentage of total block benefit (reward + fees). Right now, it's hovering around 2%, but has gone as high as 45% in the past. At some point, we'll reach a crossover point where the transaction fees will be higher than the block reward.

[Anti Federalist]

From Bitcoin's creation in 2009 until about roughly 2140, there is a scheduled payout of "free" Bitcoins to miners. This is called the block reward or block subsidy. At first, it was 50 Bitcoins per block (hard to imagine nowadays) and this gets cut in half roughly every 4 years -- 25, 12.5, 6.25 (current payout), 3.125, etc. In 2140, the final payout block will be awarded and there will be a grand total of exactly 21 million Bitcoins. Here is a chart showing historical payout versus price.

Another way to think of it is that there is just one "mega-Bitcoin" which has been sliced into 21 million pieces and is being paid out little-by-little with each block mined. As you can see here, the vast majority of Bitcoins have already been mined (about 90%), and the remaining mining is just targeting that last 10% of Bitcoins that will ever be paid out to miners. So the "inflationary phase" of Bitcoin is already in the rear-view mirror. From now until 2140, your Bitcoins will devalue a grand total of 10% as a result of the block reward. So, compared to any fiat government currency, Bitcoin is hard money.

The block subsidy incentivizes mining while the network grows. As the subsidy rapidly approaches zero, the miners will be working for transaction fees, rather than the reward. Here you can see a chart of transaction fees as a percentage of total block benefit (reward + fees). Right now, it's hovering around 2%, but has gone as high as 45% in the past. At some point, we'll reach a crossover point where the transaction fees will be higher than the block reward.

So, what we've got here are groups of propeller heads all around the globe, burning up googleplexes of computer time and the electricity needed to drive those servers, in order to "pull" trillions of virtual slot machine handles trying score a jackpot of a declining slice of the initial bitcoin basket startup?

Why do I feel less confident in this than I do investing in Zimbabwean money market accounts?

Until all this came up, my understanding of Bitcoin was that it was nothing more than a virtual "debit card" type of system, albeit with better security and privacy.

I think it was @dannno that mentioned this once to me when I was asking the same questions...

My understanding is also that the only thing of value these "miners" create, is the calculating network that BitCoin has piggybacked onto to provide security?

[ClaytonB]

So, what we've got here are groups of propeller heads all around the globe, burning up googleplexes of computer time and the electricity needed to drive those servers, in order to "pull" trillions of virtual slot machine handles trying score a jackpot of a declining slice of the initial bitcoin basket startup?

For a declining slice of the initial Bitcoin block reward (aka "coinbase"), plus transaction fees. Right now, the blocks are paying miners about 98% reward (subsidy) and 2% fees. But that ratio will shift over time. After 2140, it will be 100% fees only (no new Bitcoins being "printed").

Why do I feel less confident in this than I do investing in Zimbabwean money market accounts?

Well, there's no doubt it's a risky investment. I encourage people who don't understand it to just stay away. But even if you don't understand it, I would encourage you to consider the possibility that Bitcoin is your friend (a real force for liberty.)

Until all this came up, my understanding of Bitcoin was that it was nothing more than a virtual "debit card" type of system, albeit with better security and privacy.

Bitcoin did for money what BitTorrent did for file-sharing -- by splitting up the Torrent index across all active clients in the Torrent network, there is no central server that the authorities can shut down (as they did with Napster). The closest thing to that (for a while) was PirateBay, and that's why the Fedgod dropped the hammer-of-god on its creators. Ross Ulbricht was, similarly, made an example of, for the same overall reason. The idea is that, even though the Fedgod knows they cannot stop these networks, if they hit any central index operators to medieval extremes with maximum sentences, etc. then others who are tempted to operate such indices will be dissuaded from their foolish ways and the network will die out because there will be no online index to search, and casual users will give up. That will leave a few hardcore holdouts that can be torpedoed over time using search&destroy methods.

That strategy didn't work, however. BitTorrent is alive and well and doing better than ever, proving that Bram Cohen's original design was sound. And Bitcoin has taken this same concept and added proof-of-work (mining) to make sure that network operators have a dollars-and-cents incentive to keep the network operating. You can't (directly) make money operating a BitTorrent node. The Torrent network has no built-in payment system (although Cohen has toyed with that idea post-Bitcoin.) They couldn't shut down BitTorrent which operates pro bono... how much harder will it be to shut down a for-profit, distributed network with no central server/index??

I think it was @dannno that mentioned this once to me when I was asking the same questions...

My understanding is also that the only thing of value these "miners" create, is the calculating network that BitCoin has piggybacked onto to provide security?

Well, yes, the computations themselves are utterly useless. There are some cryptos that have toyed with harnessing mining for useful purposes (and I think this idea could even work). But for the purposes of Bitcoin, it doesn't matter whether the mining does anything useful or not. All that matters is that mining isn't free. This solves the so-called Sibyl attack, which is like dead Democrat voters (in a "voting" system, just fabricate as many identities as votes you need). Instead, Bitcoin mining is "vote by kilowatt". You can be the lucky person who adds the next block to the blockchain... all you gotta do is burn enough kilowatts of electricity and you will eventually win the lotto and get to add your block. Simple enough.

[dannno]

So, what we've got here are groups of propeller heads all around the globe, burning up googleplexes of computer time and the electricity needed to drive those servers, in order to "pull" trillions of virtual slot machine handles trying score a jackpot of a declining slice of the initial bitcoin basket startup?

Why do I feel less confident in this than I do investing in Zimbabwean money market accounts?

Until all this came up, my understanding of Bitcoin was that it was nothing more than a virtual "debit card" type of system, albeit with better security and privacy.

I think it was @dannno that mentioned this once to me when I was asking the same questions...

My understanding is also that the only thing of value these "miners" create, is the calculating network that BitCoin has piggybacked onto to provide security?

It is secure decentralized money that can't be inflated by governments.. and the amount of electricity used is far less than what banking institutions use.

The whole point is to put the central banks out of business.

[ClaytonB]

It is secure decentralized money that can't be inflated by governments.. and the amount of electricity used is far less than what banking institutions use.

The whole point is to put the central banks out of business.

Not to mention the collateral damage created by central bankers...

[jmdrake]

Objection: Isn't there an easier way to do all of this? Do we really have to burn a third-world country's worth of electricity just to maintain a simple ledger?

Answer: It is conceivable that there are other, better solutions to the problem that the blockchain solves (distributed trust without a central enforcer), but we don't know of any, as of 2022. The value of the blocks in the blockchain are being arbitrated against their true market value through the costs of mining (including both equipment overhead and operational costs).

Ethereum, the worlds second biggest cryptocurrency (I think), has moved from proof of work to proof of stake. It's certainly less resource intensive. But I'm not sure which is better.

[ClaytonB]

Ethereum, the worlds second biggest cryptocurrency (I think), has moved from proof of work to proof of stake. It's certainly less resource intensive. But I'm not sure which is better.

It depends on how you think about it. Ethereum requires "gas" in order to work, and gas can only be created by mining (just like Bitcoin, except Ethereum uses a different mining algorithm). So, who owns what ETH can be called proof-of-stake, but how you transfer ETH from one account to another still requires proof-of-work. As far as we know, you can't get around proof-of-work for any network that aims to solve the Byzantine Generals' problem. If some genius invents a better solution, that solution could potentially replace the Bitcoin blockchain as the de facto cryptocurrency root-of-trust. But, for now, PoW is the state-of-the-art.

Note that CBDCs and even fiat money also have a PoW mechanism, it's just that the "work" in this case is the government's credentials. Those credentials require maintaining a military (and burning all the fuel required to do that), government agencies, and so on. This is all part of the "work" implied in the govenrment's proof-of-work scheme for fiat/CBDCs. So the assertion that crypto is "bad for the environment" is simply false. I mean that it is false without respect to any other considerations such as green crypto, and so on. Bitcoin performs the same function as a government-administered banking system -- creating distributed trust between untrusted peers -- with 1% or less of the IRL costs (meaning, with 1% or less of the pollution and other detrimental effects). 1% here is poetic, not a hard number... but it could be even lower than that. The point is that the blockchain is virtually costless and pollution-free compared to the government-based monetary system.

[jmdrake]

It depends on how you think about it. Ethereum requires "gas" in order to work, and gas can only be created by mining (just like Bitcoin, except Ethereum uses a different mining algorithm). So, who owns what ETH can be called proof-of-stake, but how you transfer ETH from one account to another still requires proof-of-work. As far as we know, you can't get around proof-of-work for any network that aims to solve the Byzantine Generals' problem. If some genius invents a better solution, that solution could potentially replace the Bitcoin blockchain as the de facto cryptocurrency root-of-trust. But, for now, PoW is the state-of-the-art.

I'm not the expert. But it seems that Ethereum's Casper model is designed specifically to reduce the energy needed for Bitcoin like POW. And yes there are miners in Casper, but they play a much diminished role. On the flipside the Hive blockchain (which is where all of the legit Steemit users went after it got overrun by the Chicoms), doesn't seem to use POW at all.

See: https://www.coinreview.com/ethereums-proof-of-stake/

See: https://hive.io/whitepaper.pdf

Note that CBDCs and even fiat money also have a PoW mechanism, it's just that the "work" in this case is the government's credentials. Those credentials require maintaining a military (and burning all the fuel required to do that), government agencies, and so on. This is all part of the "work" implied in the govenrment's proof-of-work scheme for fiat/CBDCs. So the assertion that crypto is "bad for the environment" is simply false. I mean that it is false without respect to any other considerations such as green crypto, and so on. Bitcoin performs the same function as a government-administered banking system -- creating distributed trust between untrusted peers -- with 1% or less of the IRL costs (meaning, with 1% or less of the pollution and other detrimental effects). 1% here is poetic, not a hard number... but it could be even lower than that. The point is that the blockchain is virtually costless and pollution-free compared to the government-based monetary system.

No debate from me on that point.

[GlennwaldSnowdenAssanged]

Rumor has it that Satoshi Nakamoto was heavily invested in oil, gas, coal, Solar, wind, and electricity.

[GlennwaldSnowdenAssanged]

Climate change might be the death of BTC.

[ClaytonB]

I'm not the expert. But it seems that Ethereum's Casper model is designed specifically to reduce the energy needed for Bitcoin like POW. And yes there are miners in Casper, but they play a much diminished role. On the flipside the Hive blockchain (which is where all of the legit Steemit users went after it got overrun by the Chicoms), doesn't seem to use POW at all.

See: https://www.coinreview.com/ethereums-proof-of-stake/

See: https://hive.io/whitepaper.pdf

Unless I'm overlooking something, I think you will always need a certain amount of PoW because PoW is how the network filters out spam transactions/blocks. Proof-of-stake could be used to reduce the need for PoW but, at least with a Bitcoin-style network (Wild West model, anyone is permitted to connect as a node, miner or user), you have to have PoW to filter out the bad actors.

In a way, Lightning Network (which rides on top of the blockchain) can be seen as kind of proof-of-stake model since you use your stake (Bitcoins) to open a channel. This permits a large number of real exchanges of value to occur, without any need to put them onto the blockchain until the channel is closed and, even then, the information about the individual transactions is discarded completely and only the final balance on the channel goes onto the blockchain. So, I think that Bitcoin is getting a bad rap in the press about how much energy it uses. LN is basically zero-energy since it's just a matter of sending a signed Lightning transaction through the LN network, which is basically zero-energy, no more energy than required for you to receive a single, all-text email. Each Lightning channel can support potentially many thousands of transactions before needing to close and settle on the blockchain, which means that the total "transaction bandwidth" of the blockchain is vastly larger than it appears from the number of raw transactions in the blocks themselves. Any of those transactions that are opening or closing a Lightning channel could represent hundreds, thousands, maybe tens of thousands of transactions, in a single transaction. All of that is possible through the proof-of-stake-like nature of LN channels.