Science | Bitcoin governance

This article is based on my speech at the Chain-In conference:

Why do we care about the governance of Bitcoin?

Bitcoin governance is so important because it is not only the originator of cryptocurrency, but also the most liquid and most prestigious of all cryptocurrencies. In the words of Michael Goldstein: “The sound currency is the backbone of civilization. Bitcoin is such a powerful tool that can promote social collaboration.” If the governance model of Bitcoin is flawed, it will not be able to realize its full potential. If the governance model of Bitcoin is flawed, its stakeholders should work hard to fix these defects.

Whenever people talk about the governance of Bitcoin, the focus will always be on the question of "who is the ultimate decision maker." It is nothing more than the group of miners, nodes, and investors. The purpose and mechanism of governance are inconspicuous. Even out of touch with reality. The evaluation of past governance situations usually refers to which party “wins” or “fails” in decision-making, rather than whether the decision-making process is sufficient.

What is bitcoin governance?

Bitcoin governance refers to the process of agreeing and implementing a set of transactions and block verification rules. The so-called "execution" means that everyone uses these rules to verify that payments from transactions and blocks meet their subjective definition of "bitcoin." As long as two or more people have adopted the same set of verification rules, they have reached a social consensus on the definition of "bitcoin" from a subjective level.

What is the goal of Bitcoin governance?

What goals should Bitcoin Governance achieve? Everyone's views are different. What kind of results are achieved in governance?

  • Matt Corallo believes that trust-free is one of the most important features of Bitcoin. Matt's definition of trust-free is "in the process of using Bitcoin, users don't have to trust anything other than running open source software." Bitcoin loses a lot of advantages if it lacks the feature of trust-free.
  • Daniel Krawisz believes that maximizing the value of Bitcoin is the ideal outcome for governance. Daniel said that "the general rule for Bitcoin upgrades is to adopt only those upgrades that increase the value of Bitcoin."

As far as bitcoin governance is concerned, the above two views reflect the classic differences between morality and result theory. I prefer that Matt focus on the idea of ​​trust-freeness. Throughout the history of the currency, from the ancient Mint to the modern central bank, the result of trusting others to make money is, without exception, the abuse of this trust. If you give up the trust-freeness, Bitcoin can only reach the peak of price in some areas and cannot create new heights on a global scale. In addition, there is no evidence that the price of Bitcoin is related to the upgrade of the Bitcoin Agreement. The basic value of Bitcoin may be affected by the upgrade, but this cannot be effectively reflected in the market price, because Bitcoin has strong liquidity and volatility. If we can't see the impact of the upgrade on the value of Bitcoin, then the result-maker's approach loses its reliance (becoming inoperable).

We should first define the current Bitcoin governance process in order to evaluate the two goals of maintaining trustworthiness and enhancing value as mentioned above.

How does the current Bitcoin governance process work?


The purpose of the Bitcoin governance process is to maintain a set of validation rules. At an abstract level, this set of validation rules covers syntax, data structures, resource usage limits, integrity checks, time locking, reconciliation of memory pools and major branches, coinbase rewards and cost calculations, and block header validation. Improving these rules without making any sacrifices is not that easy.

Most of these rules come from Satoshi Nakamoto. Some rules have been added and improved to deal with failure and denial of service (DoS) vulnerabilities. In addition, some rules have been changed to implement new and innovative projects. For example, the Check Sequence Verify opcode has been added to implement new scripts.

the study  

Every rule change begins with research. For example, segregated witnesses stem from research that addresses transactional plasticity. Transaction remodeling is a very serious problem, and if it can't be solved, Lightning Network can't be deployed on Bitcoin. With the joint efforts of researchers outside the industry, the isolation witness technology was finally created.

Critics point out that the researchers want to study what, what users expect and what is good for network attributes, and occasionally there is a disconnect between the three. In addition, academic computer scientists like "scientific simulation" more than "engineering experiments." This is one of the reasons why the circle of researchers is so anxious.


After the researcher finds a solution to a problem, he or she will share his improvement proposal with other protocol developers. The main sharing channels are: email to the Bitcoin developer mailbox list; a formal white paper; and/or a Bitcoin Improvement Offer (BIP).


To implement a proposal on the node software, either by the proponent who gave the proposal itself or by other protocol developers interested in it. If the proposal is beyond the capabilities of the researcher or is not peer-reviewed, it will be put on hold until it is abandoned or modified.

While this may give you the impression that contributors to the development of the Bitcoin protocol can vote on a proposal, in fact researchers can bypass the developer and directly make the proposal public. In this case, if the researcher does not have sufficient visibility and credibility, it will be at a disadvantage.

There is another problem in the implementation phase. If a reference implementation is not recognized by the Bitcoin protocol developer or even other members of the Bitcoin community, the implementation maintainer will not incorporate it into the real software implementation. in. The maintainers of the reference implementation will consciously respond to changes in consensus rather than forcing a change in consensus. The C++ reference implementation hosted on inherits directly from Nakamoto's code base and is now mature and reliable, and has always been the most popular bitcoin node implementation.

To bypass the maintainer of the reference implementation and get rid of the impact of social consensus, simply copy the Bitcoin code base and post it according to the improvement proposal. The user-activated soft fork (UASF) of the BIP-148 is just like this.

Proposals related to modifying validation rules may result in soft forks or hard forks. Some proposals can only be achieved through hard forks. From the perspective of the fork front node, the soft fork implementation is forward compatible. If it is a soft fork, the nodes before the fork do not need to upgrade their software, and can continue to verify the consensus rules before the fork. However, these pre-fork nodes do not validate rules that have been changed by soft forks. From the perspective of the fork front node, the hard fork is not forward compatible. The network consisting of the forked nodes will not accept the pre-fork node.

What effect does the hard fork and soft fork have on the network and users? This has been controversial. Soft forks seem to be less risky than hard forks because the former does not require nodes to make explicit choices, but soft forks can also be considered mandatory, because if someone disagrees with soft forks, they must pass hard points. The way to fork reverses it.


Once an implementation is deployed to the node software, the user must be convinced to use the revised node software. Not all node users are equally important. For example, the "blockchain browser" is more important because many users rely on their nodes. In addition, the exchange can decide which currency of the validation rule set corresponds to the currency symbol. Speculative traders, major holders of currencies, and other exchanges will check the verification rules for these currencies.

Individual users may reveal which version of the node software they are using on social media, which may trigger a witch attack. The ultimate test of consensus rules is whether the payment received by your node software meets your subjective definition of bitcoin and whether the payment you send to the counterparty's node software conforms to its subjective definition of bitcoin.

There is an on-chain governance proposal called BIP-9 (with a timeout and delayed version bit) that is related to soft forks. The proposal determines whether or not to implement a soft fork by measuring the support rate of the miner. In other words, use the support rate of the miner to replace the support rate of the entire community to decide whether to pass a proposal. Sadly, as mining activities continue to become more concentrated, there is a conflict of interest between miners and users, and the results of such alternative measurements may not be accurate. The mechanism that allows miners to “vote” on the chain will continue a series of myths, such as: bitcoin is the miner's ruling, the miner decides the transaction and block validity, and so on. BIP-9 is only useful if we are aware of and accept this alternative measurement.

carried out

  Changes to the validation rules are performed by a decentralized peer-to-peer network consisting of fully verified nodes. The node uses the validation rules to independently verify that the payment received by the node operator is included in a valid bitcoin transaction and bitcoin block. Nodes will not broadcast transactions and blocks that violate the rules. In fact, if a node receives an invalid transaction and block, it will disconnect from the sender node and refuse communication. As StopAndDecrypt said, "Bitcoin is unbreakable in terms of verification." If everyone agrees that a block is invalid, the miners who dig out the block will not receive the coinbase reward and transaction fees.

The miner's job is to generate a proof of effort-adjusted workload based on a set of transaction sequences, including a release function proof (often referred to as a "timestamp"). On the one hand, miners consume hardware and electricity costs to provide computing power, and on the other hand, they receive income through coinbase rewards + transaction fees. A miner is like a mercenary. In the past, miners did not perform full rule verification during the provision of mining services. Due to the concentration of mining activities, it is impossible to believe that the mining union automatically and consciously performed the verification.

Does the current bitcoin governance model enhance trustworthiness?

I believe that the current bitcoin governance model prevents a decline in trust-free levels. Over the past five years, the number of transactions on the Bitcoin chain has increased dramatically and the momentum has not diminished. Last year, miners called for doubling block capacity. If the bitcoin governance model didn't stop them, then it would be a sacrifice of trustworthiness while increasing transaction throughput.

Does the current bitcoin governance model increase the value of Bitcoin?

I don't think there is a causal relationship between the two. The price of bitcoin has risen sharply compared to two years ago, but the process seems to be driven by the psychology of the trader, not the technical foundation. It is undeniable that, in terms of technology, the current bitcoin governance model has modified the consensus rules so that the lightning network can be deployed on the Bitcoin network. I have already experimented on an existing channel and paid for lightning: I think the Lightning Network has undoubtedly increased the value of Bitcoin.


Original link:

Author: Pierre Rochard

Translation & Proofreading: Min Min & A Jian

(This article is from the EthFans of Ethereum fans, and it is strictly forbidden to reprint without the permission of the author.