Among the factors that threaten Bitcoin’s decentralization, cost is the smallest one

Chapter 0 Introduction

People always say that if the block size becomes bigger, it will affect the decentralization of Bitcoin. In fact, they don’t understand at all.

Chapter 1: The Specific Manifestations of Bitcoin Decentralization

What does decentralization mean?

1. Bitcoin mining is decentralized.

That is, the ownership of mining machines and mining pools is owned by many people, preferably distributed in different geographical administrative regions. The more the mining machine computing power is dispersed among more people and in more geographical locations, the more decentralized it is.

2. Bitcoin full nodes are decentralized.

The more nodes there are and the more dispersed their geographical locations are, the more decentralized it is.

3. Bitcoin protocol development is decentralized.

The more developers there are from more geographical administrative blocks, the more decentralized the development is.

These three are complementary to each other. Bitcoin is a public ledger. Mining is to record and verify the authenticity of transactions. Nodes are to store ledgers and verify the authenticity of transactions. Development is to improve and optimize the accounting method.

Chapter 2 Factors Affecting Mining Decentralization

1. Mining threshold. The higher the entry threshold, the more likely it is to lead to centralization.

The threshold of mining is mainly caused by mutual competition, which is reflected in the increase of the computing power of the whole network. The higher the computing power, the higher the cost of purchasing a mining machine with the same output.

The impact of block size on the decentralization of mining is that, in theory, the larger the block, the higher the storage and bandwidth costs, and the higher the orphan block rate, which in theory will push up the threshold of mining and give large mining pools more advantages. But in fact, the opposite is true.

First, we need to rule out the fact that larger blocks will lead to an increase in the orphan block rate. This can be solved through technology, such as thin blocks, speed forwarding blocks, and parallel block verification. In addition, the mining pool can choose how large the blocks are to be packed. The mining pool will not pack large blocks that will increase the orphan block rate. No one wants to be mean to their own money.

The larger blocks require larger hard disks and more expensive bandwidth, which is completely deceptive. The storage cost of hard disks and bandwidth may be the least noticeable cost of opening a mining pool. A 2T hard disk sells for less than 600 yuan on JD.com, and it can be used for 5 years for blocks within 8M. As for bandwidth, don't even think about it. There is no need to upgrade the bandwidth for 20M blocks.

The current trend of Bitcoin computing power is getting higher and higher, which means that the threshold for entering mining will become higher and higher. So is mining inevitably moving towards centralization?

The answer is no. This brings us to the second factor that affects mining: commercial profit.

2. Mining profit. The higher the mining profit, the easier it is to decentralize.

Capital from all over the world will not miss any promising business opportunities. For Bitcoin, if mining is considered promising, there will be a lot of people waiting to enter the market with money.

We don’t need to worry about the centralization of mining due to larger blocks, nor do we need to worry about the centralization due to high entry barriers. We only need to worry about whether Bitcoin mining will become a business model with no future, which is the root cause of centralization.

Small blocks will undoubtedly make mining a dead end business. Mining revenue comes from block rewards and Bitcoin transaction fees.

75% of the block rewards have been mined. Of the total 21 million bitcoins, only less than 25% are left for everyone to mine. So if you want to keep the bitcoin business model going, you can’t rely on block rewards. You can only rely on transaction fees.

Small blocks, 1M blocks, will undoubtedly make the business of earning fees a completely hopeless business. This will make Bitcoin transfers a very bad service, and users' transactions will not be confirmed in a timely manner. No one would think that a poor accounting service is a promising business model. Users are not stupid, and Bitcoin is not without competitors. No one would invest in a business model like yours that makes money by raising the transaction fee. I have never heard of it.

Larger blocks can accommodate more transactions and provide more stable transfer services. However, the marginal cost of miners in packaging transactions is close to zero, and packaging one more transaction does not increase the cost. Therefore, relying on the number of transactions to win is the best strategy.

At the same time, because it can provide better transfer services, users will be attracted, thus driving up the price of Bitcoin. In this way, the profit of mining calculated in legal currency will be higher. This business will have a more promising future. Mining can also be more decentralized.

Chapter 3 Factors Affecting Node Decentralization

1. The cost of operating a full node. The higher the cost, the more it affects decentralization.

But in fact, increasing the block size does not significantly increase the operating costs of a node. The main cost of running a node is nothing more than a computer, even a computer like a Raspberry Pi, especially a large hard drive, an Internet connection, and a small amount of electricity.

Increasing the block size will at most affect the cost of the hard disk, but the cost of computers, bandwidth and electricity will not increase at all. And now hard disks are widely used. A 2TB hard disk costs less than 600 yuan and can be used for more than 5 years.

Those who say that larger blocks will increase the cost of running nodes, thus causing the failure of node decentralization are all fools. They are hooligans who talk about the efficacy of drugs without considering the dosage.

2. People who accept Bitcoin. The more people who accept Bitcoin, the more potential people there will be to operate nodes, which can promote the decentralization of nodes.

If the block size is 1M, it is obviously unlikely to promote Bitcoin directly. Who would want to use something that blocks even the transfer of accounts? Under the 1M block size, Bitcoin can be promoted in an intermittent manner. A large number of users use offchain wallets, future lightning networks and side chains, etc. However, these users have no possibility of running a complete node. Therefore, the 1M block size is relatively more likely to lead to node centralization.

But how do you turn potential node operators into people who actually run a node?

That’s what’s needed to turn Bitcoin into something cool.

Let me give you a simple example. Lamborghini sports cars are cool and expensive, but Lamborghinis are sold all over the world. However, QQ cars are cheap but not cool, so they are not sold all over the world. Lamborghinis are not decentralised because of their high costs, but QQ is a failed product that leads to serious centralisation.

The same is true for Bitcoin. It needs a larger user base and more people talking about Bitcoin so that more people will be proud to run a full node. This will make the node more decentralized. This requires larger blocks.

3. The benefits of operating a full node. If running a full node is beneficial to one’s own business interests, it is beneficial to the decentralization of the full node.

This requires that the group of people using Bitcoin is large enough and the business models are diversified, so that Bitcoin attracts the attention of enough merchants. Obviously, this also requires larger blocks.

Bitcoin merchants, such as wallet operators, exchanges, and large companies that accept Bitcoin, have an incentive to run full node wallets as long as Bitcoin has enough users, even if the cost is high, in order to provide a better payment experience for Bitcoin users and prevent potential double spending. On the contrary, they have no incentive to run a digital currency wallet used by a niche user, such as Litecoin.

The most serious period of Bitcoin centralization was when Satoshi Nakamoto first launched Bitcoin. The computing power of the entire network was 100% centralized, and the coins were 100% centralized. Because at that time, no one cared about Bitcoin except Satoshi Nakamoto. Even at that time, the storage cost and bandwidth cost of running a complete Bitcoin wallet were lower than running QQ.

Therefore, in order to make the nodes fully decentralized, instead of worrying about the storage costs brought by larger blocks, we should focus more on how to make more people like Bitcoin, make Bitcoin a cool thing, and make its commercialization richer.

Chapter 4 Factors Affecting Decentralized Development

1. The difficulty of development. The more difficult it is, the fewer potential developers there are, and the more centralized the development tends to be.

In order to achieve more extensive development and decentralization, Bitcoin must be protocolized and documented. This will help more people understand Bitcoin accurately and clearly.

Questions such as whether XT is an altcoin or not, and whether Segregated Witness is an altcoin or not, are all because Bitcoin has no protocol and each developer is messing around according to their own preferences.

The core protocol of Bitcoin is kept as simple as possible, and complex functions that can be completed by commercial companies should not be included in the core protocol of Bitcoin as much as possible.

Things like RBF functions should not be included in the Bitcoin core wallet. Let commercial companies do it themselves.

For example, the Segregated Witness soft fork defines two more Bitcoin transaction formats and uses programming tricks to deceive nodes that have not upgraded. This has complicated the program and added unnecessary difficulty to subsequent development. These complexities are the main reasons for the centralization of development.

Anyone can understand that the block size can be directly expanded and hard forked without changing the program logic. However, tricks like Segregated Witness, which are clever but lose the wisdom, will only lead to Bitcoin relying on Core in the future and serious centralization of development.

2. The benefits of development. If development is profitable, development tends to be decentralized.

The reason is simple. It is really not possible to maintain decentralization purely by relying on ideals and sentiments. Where do so many Satoshi Nakamotos come from?

The premise for making development profitable is that Bitcoin is more commercialized. This is not possible without large blocks.

3. Make development cool. It’s a mental reward. Just like a hobbyist running a full node.

But the goal is to make Bitcoin a more widely used currency, and to make Bitcoin itself a cool thing. This requires bigger blocks.

Chapter 5 Summary

What can make mining more decentralized is to make mining a very good business.

What makes full nodes more decentralized is what makes Bitcoin a cool thing, something that is widely used.

What will make development more decentralized is to make Bitcoin evolve in a clear and simple direction, rather than making it more complicated.

The impact of the cost of expanding the block on decentralization is completely negligible.

Chapter 6 Conclusion

There is so much money in this world that no one can’t afford a hard drive.