Litecoin founder holds a hot AMA: plans to force Segregated Witness

Chapter 0 Introduction

Litecoin founder Charlie_Lee (Li Qiwei) did an AMA event on 8btc today. After reading the entire post, I was really shocked: Wow, so informative.

Chapter 1 LTC miners reveal 2% donation agreement

The most popular content of this AMA was not Li’s answer, but what Jiang Zhuoer of Litecoin mining pool LTC1BTC revealed during a conversation with netizen wanshun. Jiang’s original words were:

A miner wants to maintain the status quo and not donate 2% to development. He does not want to shut down or fork.

I just want to ask you whether it is “forced” that minority miners are forced to donate 2% to development.

The above two sentences show that Litecoin developers want mining pools to donate 2% of mining revenue to development, and have ways to force miners who disagree to donate.

Since no further information was disclosed, I contacted Jiang Zhuoer directly on WeChat to confirm the matter.

After confirmation, it is true. Specifically, LTC developers negotiated with mining pools to let miners donate 2% of mining income to the LTC Foundation. And the mining income is automatically sent to the foundation address using code, and a soft fork is used to implant the code. Therefore, miners who do not agree to donate will be forced to donate in the face of a soft fork, because after most computing power executes the soft fork, the small computing power that does not execute it will be isolated.

This plan only remained in the negotiation process between the developers and the mining pool, and the two sides failed to reach an agreement, so the matter was shelved.

I personally think that it is reasonable for miners to donate to developers to a certain extent, because developers only stay at the level of ideals and hobbies, and may lack motivation, which is not conducive to the development of this digital currency. However, if donations are built into the protocol, it will violate the principle of decentralization. If the protocol stipulates that donations should be made to a designated development foundation, then it will be difficult to achieve decentralized development.

Zcash uses this mining method to automatically donate 10% of its proceeds to six developers. In fact, this is a highly centralized digital currency.

Chapter 2: Lee confirms that miners who refuse to upgrade during the Segregated Witness soft fork can steal Segregated Witness transactions

When asked if the 75% hashrate threshold was too low, leading to insufficient security, Lee responded as follows:

75% is not low for a soft fork. If the remaining 25% do not support SegWit at activation, as long as the blocks they mine do not steal SegWit transactions, then they are on the right chain. If the blocks they mine do steal SegWit transactions, then they are mining on the wrong chain. If economic nodes such as exchanges and merchants are running new nodes, they will not accept the wrong chain. These miners who are mining on the wrong chain will soon realize that the coins they mined cannot be sold and will return to the right chain. This is why there is a two-week interval between SegWit lock-in and SegWit activation, we need to allow some time for miners to realize that SegWit is about to be activated and make sure they are mining on the right chain. Finally, SegWit will be slow to install at first, so if you are in doubt, wait a few weeks until all miners are mining on the right chain before starting to send SegWit transactions.

The above information is hard to understand, so I will repeat the worst case scenario in Li's answer:

1. 75% of the hashrate executes the SW soft fork, so SW is locked and then activated.

2. 25% of the computing power does not execute the SW soft fork. However, the software is a soft fork, so the 25% computing power will not mine a chain alone, but will continue to accept the blocks mined by the 75% computing power. However, the 75% computing power will not accept the blocks mined by the 25%, and the blocks mined by the 25% will be isolated and become orphan blocks. This 25% computing power has been mining in vain and has been completely isolated.

3. At this time, 25% of the computing power must either upgrade to support SW or initiate a hard fork and independently mine a new chain.

4. The process of "stealing isolated witness transactions" mentioned by Li occurs in the following situation: After SW is activated, for example, at block height 10000. 25% of the computing power that does not support SW mined 10 blocks, but they were all isolated; while 75% of the computing power that supports SW mined 30 blocks, and they contained SW transactions. At this time, the 25% computing power found that they had mined in vain, so they initiated a hard fork at block height 10030 and separated. After separation, the coins in the 30 SW transaction blocks just mined by 75% of the computing power can be "stolen".

5. Li divided the chain mining of this 25% computing power into two situations: one is "obeying 75% computing power"; the other is "disobeying 75% computing power and forking". The former is called "correct" and the latter is called "wrong".

6. The "correct" way is to keep mining isolated blocks for nothing. The "wrong" way is to form a chain after independence.

7. Li believes that the "wrong" chain will face: "If economic nodes such as exchanges and merchants run new nodes, they will not accept the wrong chain."

8. This assumes that exchanges, merchants and other economic nodes are running new nodes. As long as there are nodes that are not running new nodes, the coins on this "wrong" chain will still have a market.

9. Obviously, the probability of assumption 7 being true is very difficult. The greater probability is that there will still be economic nodes supporting the "wrong" chain.

10. The way to prevent the “wrong” chain is to “allow a two-week gap between Segwit lock-in and Segwit activation to make miners aware that Segwit is going to be activated and to make sure they are mining on the right chain.” And for users: “You can wait a few weeks until all miners are mining on the right chain before you start sending Segwit transactions.”

The above analysis leads to an obvious conclusion. After the Segregated Witness soft fork, if some computing power (for example, 25% computing power) falls back to the non-SW version, they can steal coins. Of course, the coins stolen are on a hard fork chain that is different from the SW version chain.

Chapter 3 Lee plans to force Segregated Witness to be implemented on LTC

When asked, “What if the mining industry forms an alliance and insists on not activating SW?” Li’s answer was very strong:

If all economic nodes support SegWit, and miners form a coalition for Bitcoin political reasons and insist on not activating it, then we may have to take some extreme measures. But changing PoW is not a joke, and we will not do it unless we have to. I don't think we will go to that step. We don't want to consider this extreme situation, but prefer to work with miners to make Litecoin better.

Lee made it very clear that he would implement Segregated Witness whether the miners agreed or not. Of course, he added a prerequisite that "all economic nodes support Segregated Witness". I don't know why miners cannot be considered "economic nodes".

Chapter 4: Lee believes miners have no right to vote on whether to adopt SegWit

When asked if miners have a vote on whether to activate the soft SW, Lee’s answer was very strong:

There is a misunderstanding that we do not ask miners to vote on whether to activate Segwit; in fact, we only ask mining pools to let us know when they are ready to support Segwit. If 75% of miners are ready, then we can start the implementation. Users can choose whether to use Segwit. It is up to the users. If most miners are unwilling to support Segwit, then this will indeed prevent the activation of the soft fork Segwit, but if users really want Segwit, Segwit will definitely be activated.

Later, Li gave an example, saying that the status of a miner is equivalent to that of a guard hired by a jewelry store. The guard has no right to interfere in the operation of the store. If the guard does not cooperate with the store owner's arrangements, the owner will replace the guard.

This answer is very serious. Li directly denied the power and status of miners, the executors of proof of work, in the digital currency ecosystem.

Bitmain CEO Jihan Wu once wrote an article titled "Bitmain CEO Jihan Wu: Analysis of the Governance Structure of the Bitcoin Ecosystem" analyzing the participants in the Bitcoin ecosystem:

Participants in the Bitcoin ecosystem can be roughly divided into three categories: developers, miners, and users.

In the Bitcoin ecosystem, the order of control power from small to large is: developers, miners, and users.

Wu's article analyzes in great detail the power structure and status of the three participants in the Bitcoin ecosystem. Miners are not "guards" who are hired, have no voting rights, and can be fired at will.

First of all, let's not talk about power, let's talk about ability. Miners have the ability to eliminate POW proof-of-work digital currencies such as Bitcoin and Litecoin, although it is difficult and requires 51% of the computing power. You can't say that participants with this ability have no power, and it is impossible for them to be fired at will. Just like a country's army, the army belongs to the country, and the army cannot interfere in politics, right? This is all reasonable. But you can't fire the army at will! And the fact in the real world is that the army has a great influence on national politics. This does not mean that China's political system is bad. Even if American soldiers are so particular about not interfering in politics, can American people and the president fire the army at will?

Secondly, when a blockchain forks, it is the miners who decide which chain is legitimate. This is what Satoshi Nakamoto wrote in his white paper:

“The majority decision is expressed as the longest chain, because the longest chain contains the most work,” and thus “nodes…use the longest proof-of-work chain as proof of transactions that occurred while the node was offline.”

The vast majority of users do not distinguish the technical characteristics of digital currency. They simply recognize that the longest and most secure chain is a very good standard.

Finally, mining is the currency issuance policy of Bitcoin and Litecoin. This is not just a "guard" job, it is a producer's job. Mining is at least one of the most important parts of the economic ecology of POW digital currency.

Coincidentally, today, Gavin Andrese, one of the most important developers of Bitcoin and the successor of Satoshi Nakamoto, published a technical definition of Bitcoin.

"Bitcoin" is a not-previously-spent ledger of validly signed transactions included in the blockchain starting with the genesis block (hash 000000000019d6689c085ae165831e934ff763ae46a2a6c172b3f1b60a8ce26f), with a total supply of 21 million coins and a cumulative double-SHA256-proof-of-work mechanism [1].

Gavin also reiterated the importance of Bitcoin's longest chain to its definition.

Therefore, I personally disagree with Lee’s statement in this AMA that miners are “guards” who are only supposed to obey orders . Although I also believe that in digital currencies such as Litecoin and Bitcoin, users are the group that gives currency value, I do not think that miners are only supposed to be on standby when notified. It does not mean that users and developers can “arbitrarily” fire miners.

Chapter 5 Conclusion

What an exciting AMA. For more details, please read the original article.

AMA original text: http://8btc.com/thread-46125-2-1.html