Subscribe to Powerlaw to receive new posts directly to your inbox.
Subscribe to Powerlaw to receive new posts directly to your inbox.
Nine out of ten people can be almost anything they want and the only limiting factor for them is time and the price they are willing to pay. - xh3b4sd
Humans have to agree on a lot of different things in life, every single day. Consensus describes what a group agrees on. The way in which we come to consensus is defined by a protocol, a set of rules that allow for complex behaviour to emerge. Nation states in the western world thrive for a protocol called democracy, which exists in many different flavours. Blockchain networks mostly utilize the two most prominent consensus mechanisms called Proof of Work and Proof of Stake. Both of those protocols work in fundamentally different ways. And when people talk about them, they often confuse the core of those protocols with a lot of different mechanisms that are blockchain specific at a moment of time. What I want to say with that, is that a child for instance is very different at a moment in time when we say that child is 3 years old, compared to that same child at a moment in time when we say that child is 9 years old. The child is the same, and it is human every time we look at it, but its composition and ability may be vastly different across space and time. Looking at blockchain networks, they too evolve on a temporal arch, representing different values and expressing different behaviours at any given moment in time. Now, blockchain networks are supposed to be reliable and never change in some very important ways. We expect from a blockchain that nobody may change its history, because that very history may say that I own 10 tokens of something valuable. And anyone changing that history may take those 10 tokens away from me. The way in which blockchain networks guarantee that nobody can change this history is called consensus. This week, let us look at one of those two consensus mechanisms below.
Proof of Work as a consensus mechanism provides everyone participating in the network with a puzzle to solve. The purpose of this puzzle is to proof to everyone that you did something very difficult. The nature of this puzzle is therefore mathematical. Solving this mathematical puzzle in the context of PoW implies to process a lot of digital calculations in order to find the required answer. And since we wander the digital realm, we need powerful computers to do this work for us. As we can imagine, powerful computers require a lot of electricity to operate. So on a global scale the fundamental challenge for participants in PoW blockchains becomes to be the best at, at least one of two things. Either network participants have infinite electricity to their disposal, or computer chips able to solve an infinite amount of mathematical puzzles instantly. The kicker is that neither of those extremes can be realized today. In any event, there might be somebody out there who mastered the best technology at the time, providing that player with the fastest computer chips on earth. In a similar vein, there might be somebody out there who can afford to spend the most amount of energy on all of those PoW puzzles. Those dimensions would provide fundamental advantages to any one player. In consequence, every enabling force may end up being a vector for centralization. If you are the best at something, you dominate the game. Blockchains are supposed to remove that aspect of domination by effectively neutralizing it. That is where any consensus mechanism defines the rules to level the playing field. In an interconnected, capitalistic world, PoW tries to neutralize centralization vectors by utilizing commoditized resources. Those resources are electricity and GPUs. On average, high performance GPUs and cheap electricity is somewhat evenly distributed across various pockets on the surface of our planet. And that means that PoW blockchain networks like Bitcoin are relatively decentralized. Up to this point we only touched the consensus mechanism itself. We did not discuss any given moment in time of Bitcoins evolution. We did not even mention cryptocurrencies. We left out token issuance, difficulty adjustments and security budgets all together.
The great equalizers for Proof of Work consensus mechanisms are our ability to make fast GPUs and cheap electricity widely available. Though, those two things alone make no blockchain network. The incentive to participate in any endeavour are inputs and outputs, economically speaking. Humans only do something if their perceived output is at least equal to, or greater than the perceived input that they themselves sacrificed to begin with. That anticipated delta between the input provided and the output received is the reward that motivates any individual to expend any amount of energy. The incentive that we talk about here in the context of blockchain networks are token rewards that network participants may generate for themselves by acting honestly on behalf of the network that those participants are serving. That is then the mechanism by which BTC is mined when solving the mathematical puzzles that we began our story with above. Given the ability to mine BTC, there must be an underlying capacity or limit, to which extend those cryptocurrencies may be mined. And when we talk about Bitcoin's token issuance schedule, then we automatically talk about Bitcoin's security budget. Because Bitcoin is designed to spend ever more resources on mining ever fewer coins. A network's security budget describes how sustainable said network is at operating, and how good it is at securing itself over time. With higher operating costs and fewer rewards to mine, Bitcoin as a network always needed the value of BTC to go up in order to sustain its own system. One of Bitcoins challenges in the future will be to sustain its own system while fewer and fewer BTC rewards may be mined, all the while higher and higher operational costs must be covered. Economically speaking, the cost basis to produce 1 BTC steadily increases, and will reach a level of 100,000 USD in the near future. And once the amount of money that someone can get for a single BTC goes lower than the amount of money that it takes in order for any BTC to be produced, then Bitcoin as a Proof of Work network shrinks in its ability to operate and secure its own value proposition. All of that is to say that long term, Bitcoin's security budget is unsustainable, and it is almost certainly guaranteed that Bitcoin as it exists today, will not exist in ten years time from now. The irony of Bitcoin's security budget is its unique selling point, which is to never change. 21 million BTC is all there will ever be, and given this constant, at some point in the future Bitcoin will first stop to grow, and then start to diminish, because eventually there will be no value left for anyone to extract by operating the system. And at that point Bitcoin becomes a liability to operate, an unsustainable energy leak. Technologically it is possible to make Bitcoin fit for the future, but the irony pointed out above is that this means change, and fear, and uncertainty, and doubt. And nobody can tell how any of this will ever be worked out. Ask 5 people today how Bitcoin is going to address its security budget in the future, and you will get 6 answers. Proof of Work itself is a beautifully elegant mechanism, because it forces humans to battle with nature by a way of pure energy and math. But Bitcoin today, at a moment in time, is destined to fail, or change beyond recognition. One way or another, Bitcoin is going to be no more. And no amount of hand waving is going to change that.
The decentralization aspects of a blockchain network are incredibly important, and by consequence incredibly complex. Decentralization is a very technical term that expressed in the real world can mean a lot of different things. On a purely technical level, decentralization means "in many different places at the same time", as opposed to the opposite centralization "in a single place". So decentralization is the property of something happening concurrently across a broader spectrum across space and time. The relevant question for us to ask now is on which dimensions this property should be expressed in order to actually qualify as decentralized in the context of blockchain networks. And the answer is unfortunately that there is no single right answer. In fact there are multiple dimensions on which decentralization must be expressed simultaneously, but the degree to which every single dimension should express the property of decentralization may very well vary dramatically. The dimensions that interest us the most in this context are for instance geographic distribution, because we want our blockchains to be validated from many different places all around the world. Then there is the economic barrier to entry, which dictates who might be able to participate in validating a blockchain network. And that barrier must optimally be as low as possible, because the lower the barrier to entry, the more different people may be able to help validate the network in order to keep it honest for everyone involved. There is also the ownership distribution of the associated network token, which is relevant because a blockchain is of no use, if some single person or entity controls virtually all of its coins. And to make the 4 horsemen complete, we need to consider the structural architecture that allows a blockchain network to evolve and adapt across space and time. That structural architecture is of deeply cultural nature. And so, any social layer rooted in complacency or impotence may very well cause its governed systems to perish eventually.
In the spirit of all of the above, our number for the week turns then out to be 5%, because that is the estimated amount of BTC owned by all Satoshi wallets. We can now go on to believe that this is a good thing because those coins will never move again, removing about 1 million BTC from circulating supply, and number go up. But, we can also believe that those coins will be the first to move once quantum computers manage to break modern encryption schemes, because that alternative version of our universe represents a real threat within the next couple of years. And then number go down. All roads lead to either death or change for Bitcoin in the near term future, and nation states governed by the most fallible, considering to put BTC into some kind of national reserve, do by my estimation not quite grasp the forces that they are meddling with. Until next time, let's try to get the big things right.
earnestly trying to find informed takes on POS vs POW, regardless of conclusion...what are the tradeoffs? https://x.com/genxdegenerate/status/1882302278767620161
good on you to seek informed takes greg from x is all maxi talk though if someone says its common sense, everyone agrees, OG crypto people know - i would run far away from them
guy was impenetrable when pressed for detail...i would love to be informed enough to argue either side, yaknow? i had assumed POS was good, but couldn't really defend it aside from the energy consumption piece
I saw your cast and wrote about PoW and PoS the past two weeks. It is probably impossible to cover each and every issue, so my writings are not exhaustive by any measure I assume. In any event, this one is about Proof of Work. https://powerlaw.systems/memo-w04-jan-2025
tysm! 420 $degen
✅ 420 tipped ∙ 512 remaining 420 / 932 (45%) 🟩🟩🟩🟩🟩⬜⬜⬜⬜⬜
And last night dropped about Proof of Stake. https://powerlaw.systems/memo-w05-feb-2025
@justindrake did his own writings on the topic again, in case you missed it. https://x.com/drakefjustin/status/1887108667675124174
imo PoW leads to more centralization due to economies of scale: big warehouses full of asic miners controlled by one entity. no one can mine on their own at home like the good ol days anymore, small timers join mining pools and already the top 5 pools control >60% btc hash rate. PoS runs on regular hardware = more accessible, and though eth also has staking pools there are way way more solo validators, best guess around 1:10 or 1:20+ in favour of solo validators vs solo miners
helpful. do you think the economic incentives are setup correctly for solo eth stakers, as-is? i hear that folks find higher, low-risk returns elsewhere, but i don't know very much
hm that is a tricky one, 3-5% yield denominated in ETH is amazing compared to any trad investments, maybe not so amazing compared to DeFi ops.. 32 eth is a big barrier to entry but i think reducing that is on the roadmap? DeFi yields are bigger but come with more risk, pros and cons on both sides. validator incentives could be improved for sure, luckily eth is all about upgrades and improvements as opposed to ossified stubborn btc