Editor's note: Starting from the birth of Bitcoin's white paper, the cryptic punks have given us another world possibility scenario of world possibilities, and WEB3 makes a complete assumption of the future world at the system level. The ideal is very full, the reality is very skinny, we still have to move forward, we want to know how the twists and turns will be achieved. Invited Liu Yi, founder of Cdot Network, to share the development path of WEB.
When discussing this topic with the Uncle Red Army, the original title is the technical path to WEB3. In the process of preparing for this sharing, I have considered a lot about what WEB3 can do, why do we pay attention to it, these businesses Or something logical in business. I think that at this stage, the logic of business may be more critical than the technical path, because if he can't finally land, then we don't think it is meaningful to discuss technology, so I changed this topic to become the gateway. The path of WEB3 , then let's talk about business logic first, then talk about technology path.
What is Web3?
- Bakkt Bitcoin contract rose by 257% in a single day, exceeding 1100 copies
- Risk of Compound being overlooked: insufficient liquidity and bank runs
- "Hengda Research Institute Blockchain Report": China's economy from high-speed growth to new growth points of high-quality development
- 蹭 Hot blockchain concept stocks: harvesting coins and harvesting stocks
- Former President of Zhejiang Commercial Bank Liu Xiaochun: Seven Conjectures of the Bank of China on Digital Currency
- Zuckerberg: Libra does not intend to replace sovereign currency and will expand US financial leadership
To talk about WEB3.0 will always encounter the first problem, that is, what is WEB3. At present, everyone's consensus is that the Internet is currently at a stage of WEB2. WEB3.0 is an idea and design for the next generation of the Internet. Since it is an idea and a design, everyone can put forward their own views, so it is very Not surprisingly, there are many different versions of WEB 3.0.
For example, as early as 15 years ago, Sir Berners-Lee, the inventor of WEB, he thought that the limitation of the Internet now is that HTML is for people to see, or for display, he has no metadata, so he can’t be machined. understanding. Then you should give meaning to these Internet data. With this combination of metadata and data, you can develop a smarter, more automated Internet application to achieve WEB3.0. At that time, he named the next-generation Internet Semantic web, which is the Semantic Web. But about five years ago, Sir Li’s perception of the next generation of the Internet changed. He launched the Internet Magna Carta campaign , calling on users and Internet practitioners to discuss the rules of the next generation Internet.
There are also many well-known Internet practitioners or researchers who have proposed this vision and design of this next-generation Internet. For example, Reed Hastings, the founder of Netfair, believes that everyone has 10MB of bandwidth, and the full-image Internet is WEB3. 0.
Eric Schmidt, Google's former CEO, believes that WEB 3.0 is made up of a number of fast-evolving small programs based on the basic services of web 2.0, also mixed, mashup. And it can be written once, run everywhere, and highly customizable, spread like a virus, this is Eric Schmidt's web3.0, he is a famous Internet analyst, called Mary Meeker of the Internet Queen, she was in 2010 Proposed web 3.0, which is an Internet composed of social networks, mobile devices and search .
Then we share the discussion of WEB3 this time, which originated from the concept that Gavin Wood proposed in 2014. So we can think of it as a Gavin Wood version of WEB3.0. Its concept has also undergone continuous evolution, and many people have contributed in the middle. This version of WEB3 can be roughly summarized as: WEB3.0 is a decentralized next-generation Internet architecture. WEB3's Internet is more free, fairer and safer than the current Internet. The vision of WEB3.0 is to make every Internet use. People can master their own digital identity, assets and data, and then master their own destiny in the Internet age.
Because about what is WEB3.0, or what is the better next-generation Internet, this is really too much, and we have no time to analyze what is the best WEB3.0, we will Gavin Wood is based on this version.
Will WEB3 be implemented?
Then after agreeing on the connotation of WEB3.0, the next question is whether WEB3.0 will be implemented? Then our IT Internet industry will produce many new concepts every year. Most of the concepts will be annihilated in history, and only a few will become reality to affect the society. To tell the truth, like WEB3.0, it sounds full of idealistic colors, and it is hard to believe that he can achieve it.
But don't rush to conclusions. Let's first analyze logically what conditions are needed to implement the new Internet architecture .
Then I made a thing similar to the decision tree in it. It is not a decision tree, but it should be viewed from the bottom up. It is to see what kind of conditions can finally achieve WEB3.0, then to achieve WEB3.0, It must be done first, that is, a large number of developers should provide new Internet applications based on this new architecture and new ideas.
Then there are a large number of users who are willing to use the WEB3.0 application. So why should developers develop and operate WEB 3.0 applications? He may be for the purpose of making money, or it may be out of love. But history has proven that actions based on feelings are difficult to form and difficult to sustain. On the other hand, what kind of Internet applications can attract a large number of users, nothing more than two kinds, either useful or interesting.
From the point of view of WEB3.0, he can't directly provide fun, so we can only consider it in this direction. In the economic sense, most of our Internet users can think of it as a betrayal. An individual whose income is earned by the labor force. So useful Internet applications can either help him make more money, or help him spend less, or help him raise his knowledge, that is, improve his ability to make money, or save time, which saves time and can be used for rest. It can also be used to make money.
Then we say that this WEB3.0 is a non-central P2P fully redundant architecture. Compared with the BS architecture of WEB2.0, it is much less efficient, so it is not suitable for developing productivity tools , and it It also has nothing to do with education, so I think that if WEB3.0 can make developers make money, it should be able to help users make more money and spend less. If you can do these three things, according to this logic WEB3.0 is possible.
The economic logic of WEB3
So can WEB 3.0 not bring benefits to developers, can it help users make more money or spend less? I think it is ok, even I think it is inevitable, that is to say, WEB3.0 analyzes from the economic point of view, it has certain inevitability.
So why do you say this? Because the core of human economic activity is trading , but information asymmetry can hinder the occurrence of transactions, or increase transaction costs. Since this information asymmetry is widespread, an intermediary has been created. The value of some intermediaries is to reduce information asymmetry. For example, real estate intermediaries and so on should be called information intermediaries. Some intermediaries use information asymmetry to decompose transactions that are difficult to directly occur, so that the transaction can be completed. For example, the role of banks in the loan and loan transactions.
For example, in the society, some people have idle funds, hoping to add value, some people need funds, but the direct borrowing can only happen between family members or very good friends, if it is said that large-scale need to borrow money. It is almost impossible to borrow from people who know. If you don’t know, you are less likely to trust him. The bank has played the role of this transaction intermediary. It absorbs deposits and gives the depositor It is an investor's expected return on interest. At the same time, the lender performs due diligence, risk control, etc., and then lends the money to the lender.
This is the bank's most basic business model. Of course, because the current commercial bank's bank that has undertaken part of the reserve fund, it assumes such a function of a super-currency currency, then it is more than just a simple transaction intermediary. The role, but this topic has nothing to do with the topic we are discussing today, then we can think, for example, in the financial industry, the big companies we are familiar with, whether it is this broker or so-called investment bank, or various securities, Insurance, etc., in fact, the core of his business is the intermediary.
We say that WEB2.0 Internet companies, especially the so-called platform-based Internet companies, its business model is also an intermediary. It is to use Internet technology to more efficiently and more extensively eliminate information asymmetry and reduce transaction costs, thus replacing traditional intermediaries or creating new intermediaries.
For example, Google is an advertising agency that implements the targeting of Internet advertising and advertising, or the precise matching of clicks. So Taobao is the information intermediary between buyers and sellers. Then Alipay is a payment intermediary . Then we will always use an example of a network car.
Before the Internet era, there were companies that called the car. However, because the efficiency of matching supply and demand is relatively low, and it does not form a large scale, Uber and Didi, which utilizes mobile Internet technology, can match drivers and passengers in a wide range, in real time and efficiently, thus developing into the Internet. The giant, then drip in this business of the network car, it is actually an information intermediary and transaction intermediary .
We say that Internet-based business has a network effect , which means that the more users, the better the service will be. Therefore, when the size of the Internet platform is large enough, a de facto monopoly will be formed. It is difficult to compete with similar new or small companies.
Or take this network car as an example. After this drop merged with the quick drop, it has almost all the net driver and passenger users at that time. The other taxis are difficult to develop because of the driver. It is very difficult to get rid of drip with passengers.
Didi does indeed bear an efficient role in this Internet intermediary. But Internet companies, which are themselves commercial enterprises, have a core mission of making money for shareholders. Therefore, it will certainly take advantage of this monopoly position and increase its intermediary fees as much as possible . For example, if he raises the charge for the car, or asks the user to add money, it can lower the driver's share to maximize his profit.
I think this is a kind of rent-seeking behavior . The meaning of rent-seeking is the behavior that contributes to the same in the production process, but uses the special status to expand the share of income distribution . Then compared with three years ago, Drip’s contribution has not changed during the entire transaction process. But because he reached the scale of this monopoly, he monopolized the market. Therefore, the drip, the subsidy network car has become a continuous increase in the transaction.
The way Internet companies monopolize the market is that he has a large number of data on service providers and service users . For example, to make a network driver, you must let the drip track your position in real time, then we have to use the drip to hit the car, we must enter their position from the software, and then where to start, where to go. But I don't think that mastering the data itself is the original sin of the WEB2.0 company, or that I don't think Internet companies should pay for the data. Because this is unrealistic, the data is generated in the process of using these Internet applications, the service provider or the user provides data, which is part of the transaction.
So it can be considered very fair. If the network driver or the taxi user is not willing to provide data, then you can not choose not to use this service, so this is what I think is a contract. It is very clear and very fair. Of course, this WEB2.0 company has the responsibility to protect the data and not abuse it.
I think the fundamental problem of WEB2.0 is that monopoly suppresses competition, and it makes efficiency impossible to further improve . For example, this drop, after reaching this monopoly scale, even if there is a newer and better network. Mechanism, or a more efficient software, there is no way to compete with Didi, and after Didi's internal company is big, this management efficiency is inefficient, people are overstaffed, but within a certain limit will not hurt him. The status of this monopoly, in this case, the business of the network car has become a business that has no way to continue to improve efficiency, no way to continue to evolve.
Then, for Internet companies, he expands his own income through rent-seeking, increases the cost of users, and reduces the benefits of actual service providers. This is also the problem of WEB2.0. Then WEB 3.0, which replaces Internet companies with a decentralized decentralized encryption protocol, makes the market always competitive, and it eliminates rent-seeking space.
Then we assume that there is a taxi chain, its experience and economic mechanism design are very good, and the market share is more than the drop, to achieve the market first. So can the taxi chain use his dominance to monopolize the market and rent-seeking?
This question is actually to be answered. It is very complicated. It is not as simple as it seems. Then he is actually related to the incentive mechanism and governance mechanism of the taxi chain. If the pass holder of the taxi can raise the price of the intermediary fee and get higher income, and they can modify the intermediary rate through governance, then they are likely to choose to raise the price, which is the search. rent.
In this sense, there is no essential difference between the shareholders of this Didi and the TokenHolder of the car chain, that is, the holder of the certificate that can have the right to govern.
However, this rent-seeking behavior of the taxi chain is limited by the fact that the blockchain can be forked. That is to say, when the taxi chain acts as an intermediary to extract too much revenue, there will be developers who have the power to fork. Car chain . Then the intermediary fee can be reduced on the new chain. Since the forked chain inherits all the data in the original chain, the driver and the user can easily migrate to the forked chain.
In addition to consistent rent-seeking, the forked feature makes the encryption protocol always in a competitive environment, because once a certain encryption protocol is stagnant, it can be forked. Then the fork chain can use more efficient trading rules or incentive mechanisms to win the service and users from the original chain.
Therefore, I think that the decentralized encryption protocol gradually replaces the Internet platform-based company, which is economically inevitable. Therefore, WEB3.0 entrepreneurs should look for those business areas that are monopolized by Internet giants and have very serious rent-seeking. From these areas to attack WEB2.0.
Then, in the decentralized encryption protocol, the user's ID, which is the key pair, is declared, does not need to be applied, cannot be banned, and cannot be disabled or spoofed. Direct control over assets and data access rights on the chain.
Without an entity he has the ability to control the identity, assets and data of the user. Because the encryption protocol is uncentered, all parties can only act according to the rules . So replacing the encryption protocol or replacing the Internet platform in some key areas means that the vision of WEB 3.0 has been realized.
The development direction of blockchain technology architecture
Let us talk about blockchain technology. The industry generally agrees that Bitcoin is 1.0 of the blockchain. Ethereum is 2.0 of the blockchain. So what is the next generation of blockchain, or what is the blockchain 3.0 has not yet reached agreement. Many people believe that blockchain 3.0 should solve the problem of Ethereum, truly making rich and decentralized The application can land. Or we hope that Blockchain 3.0 will be able to drive and implement WEB3.0.
But in this WEB3 technology stack, blockchain technology is just one of the components , in addition to many other components, such as decentralized data storage, decentralized messages, P2P network communications, and so on. So why do we pay so much attention to blockchain technology, or change the question is our goal for WEB3.0, is the focus on the technology of blockchain now overdone, or will we ignore the focus?
Then I think that blockchain technology is indeed the leader of the WEB3 technology stack, not in the technology itself, but in the economics. Because blockchain is a distributed ledger technology, it can distribute encrypted assets in a decentralized manner. Encrypted assets are scarce and supply is not elastic , which means that supply does not increase as demand grows. Moreover, the encrypted assets, that is, the exchanges are interchangeable, easy to identify authenticity, can be permanently stored, can be divided, can be programmed, can be quickly delivered in the world, and the transaction cost is very low, so it has become history The best speculation.
Due to the influx of speculative funds, the encryption of assets over the past decade has brought enormous wealth to asset issuers, the project team and early investors.
Therefore, because there are encrypted assets, the blockchain has solved the problem for WEB3. If the problem is solved by the logic of de-intermediation, then WEB3.0 can be realized.
And with the enrichment of decentralized applications, its technical architecture will become increasingly diversified. For decentralized data storage, decentralized messaging, P2P technology, other web3 requirements will continue to grow. It will also move forward along with Internet technology.
Random Capital, my investment institution, has done some research on the blockchain architecture, and we have explored many technical types, especially platform-type public chain projects, and summarized them into five directions, which we call generations. Discussion, layering, sharding, full stacking and cross-chaining.
Let's first conclude that in these five directions, our organization is most optimistic about the direction of cross-chain.
But in the other four directions, we can say that we are relatively optimistic. Of course, I should mention why I am relatively optimistic about the other four directions, but this discussion will be very complicated, because each technical direction has different choices, it is difficult to generalize, and I am relatively optimistic. It means that relative to the chain, I completely believe that in other directions, it is possible to produce important, even revolutionary, public chains.
So I can only talk about my views on these five directions. In this five directions, this direction is the earliest and the earliest . Going in this direction after EOS went online last year has already been achieved. Then why do we call him representative, because his ideas are like this.
They believe that the biggest problem with Ethernet is the lack of processing power. The reason is the fully redundant architecture of the blockchain. The more nodes there are, the larger the scope of its consensus, the more difficult it is to reach consensus, and the network burden of block propagation. The heavier.
So a possible direct test plan is to let all stakeholders vote, select a few representatives, and reach a consensus among the representatives of these few people. He represents this decision of all stakeholders, so We call this scheme representative. Then of course, this can reduce the security cost of the system, and the processing capacity is greatly improved.
The EOS annual issuance is only 1%. This evaporation rate is the lowest in all the POS chains I have seen, but it is economically feasible because there are only 21 super nodes and dozens of spare nodes to divide the money. of. And he agrees that as long as the 21 nodes reach a consensus, and these so-called super nodes have very high performance and network bandwidth requirements, so the capacity of its processing is increased by two compared with Ethereum. An order of magnitude.
For example, the current development of Ethereum is relatively fast. For example, a side chain of plasma faces such a serious problem. For example, the root of this trust is Layer1. But if this Layer 2 is doing evil overall, then the so-called mass exit is whether mass exit is feasible . This is something that is very doubtful. And it requires the user to monitor whether the side chain is evil. And challenge him after someone has a problem with the evidence.
It is impossible in the real process of implementation , because we can't ask every user, even with the help of a wallet, that he can't monitor the chain.
Moreover, the plasma side chain is also facing the problem of tracking the state of the cryptographic assets from one layer to the second layer. Since these assets need to be segmented into a form similar to this UTXO for tracking, it involves a severability and a state. A contradiction in the explosion.
So overall, I think that from the development of Layer2, as a fast payment channel is basically proved feasible, but as a fully programmable side face can be established, I think it is more and more doubtful.
Then the fragmentation is actually a high hope for everyone in the market, and it is also a technical direction with the most resources. Represented by Dfinity and Ethereum 2.0, I dare not go crazy about questioning Dfinity or Ethereum 2.0. I have seen a lot of technical design about sharding, but I don't think it seems to have seen one of the most fundamental problems of analyzing shards.
Then the question is the state fragment we are talking about , or the calculation of the fragmentation . Then, after implementing the fragmentation scheme, for example, on the chain such as Ethereum 2.0, what percentage of the cross-slice transaction will account for?
We know that this complexity of the sharding scheme makes cross-sharding transactions, it will be more than the entire transaction in the case of an intra-slice transaction, or its resources, for example, or it's this The processing speed is multiplied. If the proportion of cross-sliced transactions in the entire fragmented transaction reaches a certain level, then the fragmentation will lose its meaning.
How to perform sharding, how to cut the state or the calculation, so that cross-slicing transactions can be as low as a reasonable level? I don't see a very strict discussion about this at the moment . In addition, regarding the design of this specific fragmentation architecture, it faces many technical problems, and some of them have not been well solved. Therefore, I still have a lot of doubts about the direction of fragmentation, maybe Dfinity or Ether. Square 2.0 can better solve these problems that I am worried about.
But why do we rarely pay attention to Blockstack and rarely talk about it. And I did not see this relatively large-scale decentralized application development team using this blockstack to develop applications based on the full stack architecture.
The reason I think is still the reason for motivation, not for technical reasons . If we are an independent chain, if we send out a coin, he first solves the problem of team incentives and the incentives of this early investor, then even such a thing No one will use it in the future, but there will still be people doing this kind of thing.
But developing these applications on blockstack may be useful in the future, but who will do it? There may be people who are willing to try, or these small individual developers who are blockingstack wool are willing to try, then it is hard to imagine that they are difficult to develop high quality and long-term operation of a high quality WEB3 or Decentralized application.
Overall, I think that the full stack provides a good technical means, but instead it loses the incentives that bring the blockchain to us. If this is the case, even if the technology is good, this I am afraid that the direction will not develop.
The stable coins generated by this chain can flow through cross-chain hubs such as exchanges or other dapp applications. These dapp applications can also be based on other applications, such as based on this loan, based on bonds. Products, etc., superimposed this logic layer by layer , such an ecosystem is a scene we think of the future blockchain.
Q: What is the relationship between WEB3 and the blockchain? Can you think that the next stage of development of the blockchain is WEB3?
We believe that WEB3 is the architecture of the next generation Internet and a vision. The whole architecture of WEB3 is the next generation of Internet architecture that contains a variety of technologies. So the WEB3 foundation gives a picture called WEB3 tech stack. The blockchain technology is a technology in this tech stack.
The next stage of development of the blockchain is not WEB3, but 3.0 of what we call the blockchain . But as I shared before, we believe that blockchain technology is a leader in the entire technology stack of WEB3, and it is likely to use it as a breakthrough to drive the realization of WEB3.0.
Then I think that WEB3.0 does not extend new user requirements. Or it just uses another mode of intermediary, which is to replace the Internet platform with this decentralized encryption protocol. Then he should still be satisfied with the user's existing Internet needs.
If the design of this decentralization agreement makes this Internet enterprise that does not produce monopoly, there will be no such phenomenon that monopoly rent-seeking, so that the technology of Internet commercial intermediary is always in such an open competitive environment. This is what I think is the meaning of WEB3.0.
Q: Who is the chance that WEB3 will belong? What are the specific opportunities?
There are many differences between developing and running encryption protocols and developing an Internet business. This includes technical differences, mainly including different ways of thinking , that is, how to solve with a decentralized encryption protocol. A commercial intermediary problem, and the design of a reasonable incentive mechanism, so that people who contribute to the encryption economy can get reasonable incentives, and this mechanism can run stably for a long time.
This requires some new technology, and the new ability also includes a change in thinking mode, but I think it is in the past, for example, since the birth of Ethereum, people who entered the industry are themselves the most qualified to achieve a new generation. Decentralized Internet applications.
But it is true that the industry is generally pessimistic , especially the elderly who have entered the industry for more than three years, because they have been falsifying after a long period of trials. . As a person who entered the circle in 2013, it is also an old man who inevitably has some feeling of exhaustion. But from our team, we have seen this cross-chain technology, especially the hope of this kind of new-style decentralized application like the substance provided by the cosmos SDK. So we choose to continue to try, because we believe in blockchain technology, believe that the encryption protocol can change the economic relationship of social production, and believe that the vision of WEB3.0 can finally be realized.
As for the specific opportunities, the technology of the blockchain is basically built around the encryption of assets. Of course, this is a direction that can be explored continuously, and it may evolve into this asset. The tide of evidence, then after the asset is certified, it is equivalent to replacing the existing financial intermediary with this large number of so-called DeFi or open financial agreements.
We know that in Chinese listed companies, the profits of listed companies in the banking industry account for more than half of all listed companies. This phenomenon has lasted for ten years, which reflects the fact that the financial intermediaries appearing in the whole society are drawing blood . That is to say, the whole society is working for financial intermediaries. If we can, at least in part, turn this financial intermediary into a decentralized encryption protocol, then these participants will make a huge contribution to society and achieve these goals. The team will also receive huge economic returns.
Is it possible to trade off between safety and efficiency?
I always feel a little bit of pursuing absolute security and fairness. It is a bit like pursuing idealized democracy. If there is an ideal democracy, does it mean that efficiency will inevitably fall?
The reality of the world seems to be the case, is this a curse in the world of WEB3 ?
The so-called democracy, or the degree of decentralization and efficiency is contradictory in the field of blockchain, which means that it is impossible to triangle, because security must be guaranteed, so it can only be decentralized and this There is a trade-off between efficiency, then this is going back to cross-chain, which is why optimistic about the New Year, because cross-chain means that you can make a choice in the local , such as a specific purpose application chain, its Participants may be thousands of people on the order of hundreds of people. Then he may be 10 nodes or 20 nodes, and everyone thinks that this degree of decentralization is enough, then it will be very efficient.
So for a specific scenario, use a specific distributed ledger to solve, and then these books can be linked together through cross-chain technology to form a blockchain Internet, which is the future architecture of the blockchain that I think is the best. .
Q: As mentioned above, when the taxi chain does not meet the requirements of evolution, the user can choose to fork to continue the evolution. Currently, many POS protocols are actually embarking on the path of automatic upgrade and rejection of forks. What do you think?
This is a very serious problem. The three dimensions of decentralization mentioned by V God are at the level of architecture, at the level of governance and at the level of logic . The highest logical level is that En can maintain divisibility, which means that the chain can be divided into two, and then each runs separately. We have just talked about the significance of this WEB3.0 economics, we have already talked about The importance of this divisibility for WEB 3.0. I think that if a chain is indivisible, then it is not a blockchain.
Of course, the problem raised by the uncle, that is, this automatic upgrade, refuses to fork, it should be said to be an automatic upgrade, avoiding forks , for example, like Polkadot, this technology using WASM, it can be safely transmitted in the chain. This executable program.
So the possibility of forking caused by such upgrades, or for example, interruptions in the network, or server downtime, or the like, is reduced, even arguably eliminated, but still, This chain of Polkadot, if it is feasible to fork it, I have not seen a chain of technology that can prevent the fork.
Q: When do you think about the arrival of the WEB3 era?
I think it will take about 10-15 years.