A brief summary and reflections after the blockchain closed-door meeting

Last week, the world's first blockchain summit "Blockchain - New Economic Blueprint" was successfully held at the Hyatt Shanghai Bund. The concept and application of blockchain technology attracted the attention of 300 guest representatives from institutions and enterprises including the Financial Research Institute of the Central Bank, Shanghai Stock Exchange, Deloitte, UBS, Ant Financial, IBM, and Huawei.

Mr.肖风, Chairman of the Summit and founder of Wanxiang Blockchain Lab, said:

“Blockchain technology heralds the evolution of the Internet from an information Internet that transmits information to a value Internet that transfers value.”

In the financial world, if you haven’t heard of blockchain technology, you are really out of date. Many financial institutions around the world, including Goldman Sachs, Nasdaq, JPMorgan Chase, HSBC, Citibank, Bank of New York Mellon, Barclays, UBS, Royal Bank of Scotland, and Morgan Stanley, have cooperated with blockchain companies to study the application of blockchain technology in the financial market.

The essence of blockchain technology is a distributed (decentralized) database. It is not forgeable or tamperable. Moreover, within its transaction range, the efficiency and low cost of blockchain cross-border transactions are unmatched by traditional finance, which is why it can attract major financial institutions. The principle of blockchain technology is closely related to a mathematical problem - The Byzantine Generals Problem. Regarding this problem, I recommend you to read the brain-burning masterpiece "In-depth Discussion of the Byzantine Generals Problem", which is a great reading before bed! Bitcoin, as the first great application of blockchain technology, has attracted widespread attention in the financial world. What about blockchain applications other than Bitcoin (currency)? How will they perform?

The first book in Wanxiang Blockchain Lab's series, "Blueprint and Introduction to the New Blockchain Economy", gives everyone a glimpse of many blockchain applications. The author of the book also proposed the three major development stage concepts of blockchain 1.0 (digital currency), blockchain 2.0 (smart contracts), and blockchain 3.0 (distributed artificial intelligence and organization).

This book depicts a blueprint for blockchain applications, so it does not cover how to implement these applications in detail, nor does it provide solutions to many current problems with blockchain technology. In my opinion, the book provides more guidance on the direction of blockchain experiments. As for whether it can succeed, that is another matter.

The second day of the conference was a closed-door meeting. Fortunately, I was able to participate as a spectator. The closed-door meeting was divided into three venues, and I participated in the "Consensus Algorithm and Scalability of Blockchain Technology" meeting. This article is a summary of this meeting and some of my superficial views. If there are any errors, please correct me.

1. What are the current problems with blockchain?

1.1 Consensus Mechanism Issues

Take the Bitcoin blockchain as an example. It is responsible for maintaining the overall operation and security of the blockchain through the Proof of Work mechanism. Miners compete for the right to record the Bitcoin blockchain through random hash operations to prevent fraudulent transactions and avoid "double payments". This process consumes electricity to complete. It is estimated that the Bitcoin mining network consumes 7 million yuan of electricity resources every day. This has also attracted a lot of controversy. Opponents believe that the Bitcoin mining process is a waste of electricity resources and is unnecessary. This has also led to the birth of other blockchain consensus mechanisms such as Prove of Stake.

1.2 Issues with transaction confirmation time

Taking the Bitcoin blockchain as an example, the current average confirmation time for a Bitcoin transaction is about 10 minutes. If there are 6 confirmations, it takes about an hour to wait. Of course, compared with the 2-3 days confirmation time of credit cards, Bitcoin has made great progress, but if it is just like this, I think many ordinary users will still not be satisfied. Regarding this, the founder of Bitcoin, Satoshi Nakamoto, has his explanation. It is not that the confirmation time of blockchain transactions takes that long. This is not just a technical issue, but a trade-off.

1.3 Dealing with transaction frequency issues

In addition, the Bitcoin network currently has a limit of 7 transactions per second (7tps). In comparison, Paypal's total transaction volume in the third quarter of 2013 was 729 million, an average of 93.75 transactions per second.

Information on the official website of VISA, the world's largest payment card company, shows that VisaNet processed 47,000 transactions per second in a 2013 test.

Alipay, which we are all familiar with, achieved a peak transaction processing speed of 47,500 transactions per second during the "Double Eleven" period in 2014.

Compared to the major payment networks, the Bitcoin blockchain is more like a newborn baby in terms of transaction processing frequency. Of course, this was also a deliberate design by Satoshi Nakamoto in the early days. The Bitcoin block size was limited to 1MB to avoid malicious behavior by rogue miners and adverse effects on people (i.e. the problem of too large blocks). The reason why the Bitcoin blockchain payment network has grown to a value of billions of dollars today is that it is decentralized, which is exactly the peer-to-peer electronic cash system mentioned by Satoshi Nakamoto in the white paper.

1.4 Blockchain size issue

The size of the complete data of the Bitcoin blockchain has currently reached 44.43 GB. If users use the Bitcoin core client to synchronize data, it may take three days and three nights to complete the synchronization. In addition, the amount of blockchain data is still increasing, which brings a great threshold to the operation of the Bitcoin core client. Ordinary users are not suitable to use the Bitcoin core client. All these have caused the number of running Bitcoin full nodes to decrease instead of increase. Currently, the number of full nodes in the Bitcoin network has been reduced to 5186.

2. What are the corresponding solutions?

The main participants in this closed-door meeting were Ethereum founder Vitalik Buterin, Cryptonomex (BitShares 2.0 operating company) chairman Stan Larimer, Factom chief technology officer Jack Lu and other guests, so the meeting discussed more about the blockchain solutions of their respective crypto 2.0 projects (called altcoins in the Bitcoin community).

2.1 The debate between POW and POS

In the digital currency community, the debate over the merits of POW and POS has never stopped. POS supporters attack POW mining for wasting too many resources and being unsustainable, while POW supporters believe that the token distribution of the POS mechanism is unfair and will lead to the problem of the rich getting richer.

So what is the performance of the two mechanisms? Here, I will simply do some calculations and talk about the current performance of POW and POS.

Assume that A holds 10,000 yuan of Bitcoin and B holds 10,000 yuan of xx POS coins. (The following data is hypothetical and has no reference value)

A needs a mining machine to mine, so he uses the 10,000 yuan of bitcoins to buy a XX bitcoin mining machine, which can produce 100 yuan worth of bitcoins per day. Assuming that the computing power of the entire network and the currency price remain unchanged, 10,000 yuan worth of bitcoins will be produced after 100 days, and the electricity consumed is about 8,000 yuan. At this time, the price of the mining machine has dropped to 5,150 yuan, so A’s actual net profit is (10,000 yuan of bitcoins - 10,000 yuan of mining machine fee - 8,000 yuan of electricity fee + 5,150 yuan of mining machine value) = - 2,850 yuan.

According to the POS rule that B can earn 0.5% of the total number of coins held every month (the part less than 30 days is invalid), then after 100 days, the coins in B's hands will become about 10150XX coins, and the actual profit is about 150XX coins.
C holds 100,000 yuan worth of Bitcoin, and D holds 100,000 yuan worth of xx POS coins.

Similarly, C used the 100,000 yuan of bitcoins to purchase 10 XX bitcoin mining machines, which can produce 1,000 yuan worth of bitcoins per day. Assuming that the computing power of the entire network and the currency price remain unchanged, 100,000 yuan of bitcoins will be produced after 100 days, and the electricity consumed will be about 40,000 yuan (C has cheap electricity). At this time, the price of the mining machine has dropped to 5,150 yuan, so C’s actual net profit is (100,000 yuan of bitcoins - 100,000 yuan of mining machine fees - 40,000 yuan of electricity fees + 51,500 yuan of mining machine value) = 11,500 yuan.

According to the POS rule that D can earn 0.5% of the total number of coins held every month (the part less than 30 days is invalid), then after 100 days, the coins in D's hand will become about 101,500 XX coins, and the actual profit is about 1,500 XX coins.

In comparison, miner A lost 2,850 yuan, miner C earned 11,500 yuan, and miner B and miner D earned 150XX coins and 1500XX coins respectively.

The rule of POS currency is that since there is no cost issue, the problem of the rich getting richer is always valid (taking POS currency as the reference unit).

The POW rules are more complicated because of the added costs. If the miners do not have enough resources, they may end up losing money, but if the miners have cheap electricity resources, the problem of the rich getting richer will also arise.

In addition, the higher the interest of POS coins, the more unfair the coin will be. Generally, POS coins will also add some time rules, saying that doing so increases the constraints of time costs. If new blocks (that is, interest) cannot be generated within the specified time, this will add a mechanism that discourages consumption. What kind of thing is this?

The most controversial aspect of the POS mechanism is the initial distribution issue. For example, in the classic NXT (Futurecoin) case, there are many stages in its initial distribution:

Initial IPO (1 NXT = 2 Satoshi)

Week 1 (1 NXT = 100 Satoshis)

Week 2 (1 NXT = 400-2000 Satoshi)

Those who received the coins in the initial IPO stage must have made a fortune. Supporters would say that they took a greater risk, so the rewards should be greater. However, with the rapid distribution in such a short period of time, I can't help but ask, is this really a good thing?

What is certain is that the POW mechanism will be more easily accepted in the early stages, while POS has a clear disadvantage in the initial distribution. Simply put, POW is more suitable for early development, while POS is more suitable for later maintenance.

It is worth mentioning that Ethereum currently also uses the PoW mechanism. According to Vitalik, when the Ethereum blockchain reaches a certain block, its blockchain consensus mechanism will be PoW切换至PoS through硬分叉to save the electricity required for mining and reduce the interval time between blocks.

2.2 POS variant DPOS

There is a phenomenon in Bitcoin mining, that is, several large mining pools control most of the computing power of the entire network. For example, the total computing power of AntPool, FishPool and GuoPool in China has reached 225P, accounting for about 50% of the computing power of the entire network. In this regard, the founders of several crypto 2.0 projects, including Ethereum and BitShares, believe that Bitcoin mining has already had a centralized problem. Stan Larimer talked about BitShares' DPoS mechanism, which is called the share authorization proof mechanism (also known as the trustee mechanism) in Chinese. Its principle is to let everyone who holds BitShares vote, thereby generating 101 representatives, which can be understood as 101 super nodes or mining pools, and the rights of these 101 super nodes are completely equal. From a certain perspective, DPOS is a bit like the American parliamentary system. If representatives fail to perform their duties (when it is their turn, they fail to generate blocks), they will be removed from the list, and the network will select new super nodes to replace them.

However, there are also some problems under this model. Here is a question I raised: Suppose I am a loser and hold some BitShares, but I don’t know these candidates very well and don’t know what opinions they want to express. So I voted for a representative at random and he helped me make a choice, but in the end, the opinion he gave did not actually match my opinion. How should I solve this problem?

Or if I am a super rich person who holds one-third of the total BitShares, then the 101 candidates need to have a good relationship with me in order to get my vote. In the end, their decision is likely to be influenced by me.

The voting method of Bitcoin is to vote by simple computing power, which is relatively simple and direct. In fact, the so-called centralization of mining pools is just a false proposition, and DPOS is just a competition for resources.

So, what characteristics does this mechanism bring to BitShares?

It is reported that BitShares welcomed its 2.0 version on October 13. Ken CodeStan, the agent of BitShares 2.0, revealed that its test network recently completed a transaction rate of 10,682 transactions per second, and its ultimate goal is to reach 100,000 tps. In addition, according to the BitShares white paper document, the average transaction confirmation time of its network is only 1 second.

How about it? Doesn’t it look very impressive? At least the transaction rate has been improved, but what about decentralization? The development of BitShares has experienced many twists and turns, and it has been branded as a pie in the sky in the currency circle. Whether it can work or not, I think you, as a user, have the most say.

2.3 Sidechains and Lightning Networks and other Layer 2 protocols

The most direct way to expand the Bitcoin blockchain is to expand the block size. However, due to various factors, the community has been unable to reach a consensus, resulting in the current Bitcoin block size being limited to 1MB. The following figure is from "The Heart of the Bitcoin Scaling Battle" and roughly estimates the Bitcoin block size required to handle the corresponding number of users under different transaction requirements.

While Bitcoin block expansion is not making progress, another direction of technology is constantly being developed, which is what we call the second-layer technology. Sidechain and Lightning Network are the best among them.

Adam Back, the inventor of the HashCash proof-of-work function, is the originator of the peg-in sidechain technology, which will enable the transfer of Bitcoin and other blockchain assets between multiple blockchains (cross-chain transactions). This enables users to use new and innovative cryptocurrency systems with their existing assets. Since the sidechain is an independent system, technical and economic innovation will not be hindered by other factors. Although there is the ability to transfer in both directions between the Bitcoin system and the peg-in sidechain, they are isolated: even if the encryption in the sidechain is cracked (or maliciously designed), all damage is limited to the sidechain itself.

The Lightning Network is a concept proposed by Joseph Poon and Thaddeus Dryja. What does it achieve? In short, the Lightning Network proposes a mechanism for generating channels and payment networks. It allows small Bitcoin payments to take place on channels, taking those high-frequency, low-value transactions away from the Bitcoin blockchain. Some friends may say that this is just an off-chain solution, but what it wants to achieve is precisely decentralized off-chain. For details on how to achieve it, you can read its white paper (page 59, read with caution).

It is worth mentioning that the side chain project and the lightning network project are currently being developed by Blockstream, which has received $21 million in angel round financing from 40 investors including LinkedIn co-founder and Airbnb board member Reid Hoffman, Khosla Ventures, Canadian seed fund Real Ventures, and others.

Of course, the development of sidechains and lightning networks has also attracted a lot of criticism. For example, Mike Hearn described them as "complexity kills the cat", believing that the complexity brought about by their implementation will endanger the decentralized nature of Bitcoin.

2.4. SPV wallet, SPV node

SPV is the abbreviation of "Simplified Payment Verification". This concept was briefly mentioned in Satoshi Nakamoto's paper. He pointed out that it is not necessary to run a full node to verify the payment. Users only need to save all the block headers. SPV wallets can also be called "light wallets" or "thin wallets", which means that there is no need to download the entire blockchain data. When this technology is implemented, it also makes it possible to expand the size of Bitcoin blocks. For a detailed explanation, it is recommended to read the article "SPV, SPV Nodes and SPV Wallets".

2.5 About Factom and other applications built on the Bitcoin blockchain

The Factom network is a peer-to-peer network built by many full nodes. It is a universal data layer built on the Bitcoin blockchain. Through this data layer, users can easily create an independent virtual blockchain for their own data. Every 10 minutes, the Factom system will perform a merkle tree hash on all the data and upload the hash value to the Bitcoin blockchain.

The Factom team is also working with the Honduran government to design a land ownership record system for them. For more information, read the Factom white paper translated by Harvey Laolang: Factom Notary: A Ledger Maintained by a Consensus Algorithm.

Other consensus mechanisms, including Ripple’s latest Interledger protocol, will not be described here one by one due to space constraints.

3. Summary

Since the release of Bitcoin in 2009, blockchain technology has gradually grown into a disruptive force that is subverting the world. It originated from Bitcoin, but is not limited to Bitcoin. But in the eyes of some people, blockchain will have no other applications except Bitcoin. The concept of success is difficult to define. It is always good to try, and maybe one day you will use it. So, I think it is a good thing that there are teams developing blockchain applications. Of course, as the concept of blockchain becomes more and more popular, there will be more and more scams under the name of this concept. How to identify those fake or cheating blockchain projects has become a top priority.

Text/Sautuoxi

Source: Babbitt Information