Introduction to Web3.0 Application Token Engineering

Original title: "Octopus Network Liu Yi｜Introduction to Web3.0 Application Token Engineering"

Original author: Yu Cheng, BeWater Community

On December 1, 2021, at the 10th BeWater Live, Octopus Network founder Liu Yi shared "Introduction to Web3.0 Application Token Engineering".

This in-depth and systematic sharing, the core content:

1. Overview of Token Engineering:

- Token Classification

- Main Contributors

- Design and Engineering

2. Analysis Model

- Two Thinking Models

- Network Effect Analysis

3. Token Design

- Token Distribution Design

- Basic Token Model

- Token Engineering Process - Token Design Pattern 1. Overview of Token Engineering 1. Background 1. Hello everyone, today I will talk about token engineering. This PPT is made by me for the 4th lesson of Octopus Network Accelerator. Let me talk about the background. Octopus Network Accelerator has a series of courses. The first lesson is the introduction of the basic concepts of web3.0, the second lesson is the product design of web3.0, the third lesson is the community building of web3.0, the fourth lesson is token engineering, the fifth lesson is about the compliance of encryption projects, and the sixth lesson is the financing of encryption projects. It can be said that every important progress in the blockchain industry is due to an important innovation in the token economic model, and every important innovation can create a new field. The most typical examples are Pow and Pos. In addition, we have seen that DeFi has been popular from last summer to this year, for more than a year, and the main reason is liquidity mining. Liquidity mining is a very important innovation in the token economy model, which triggered the DeFi summer. In addition, the current gamefi/play-to-earn is a new achievement of the token economy token project, which has driven a number of projects.

If you are now working on a web3.0 encrypted network. If you are not working on the field of encrypted asset trading, but on a web3.0 application, such as a music platform, a writing platform, or various decentralized applications, it is very likely that there is no proven and effective token model in your field. The best effort you can make is to understand the basic principles of token engineering, master some of the best practices that have been identified, some tools that have been proven to be effective, and then find other projects in this field, see everyone's results, and then try to innovate on this basis. As long as you have a small step of innovation that is proven to be feasible, it is promoting the progress of this industry, and I believe it will also bring a great first-mover advantage to your own project, and it will be very good in terms of investor recognition and market recognition. However, it is actually very difficult to make any unprecedented innovation in token engineering or economic models. We must have awe for the complexity of this field.

2. Token Classification

First, let’s briefly talk about the token classification. This time, we will not talk about non-fungible tokens and security tokens, but only the cylinder in the middle, and the network token at the bottom. The network we are talking about is actually a chain, either a public chain or an application chain. It is definitely decentralized. A truly permissionless blockchain requires someone to run nodes, which has expenses and costs. The basic model is to run the network by issuing additional network tokens, network-native tokens. At the upper level is the untility token, which is actually a more general concept. It is a reward for those who contribute to the system. These people are rewarded for certain specific behaviors. When many people are guided by incentives to participate in the system in a specific way, a specific purpose will be achieved.

3. Main Contributors

I would like to mention three people. The first one is Vitalik, who has been talking about token economics since 2015. He has made many contributions. The person in the middle is not very familiar to everyone. Shermin is a scholar. I recommend her book "Token Economy". I think it is the most systematic book on token economic model design. The person on the right is Trent McAvoy, the founder of Ocean Protocol. His most important contribution may be the introduction of token engineering, which pushed token design to the engineering level. We will talk about why this is important later.

4. Design and Engineering

We talk about design and engineering. Why do we call it token engineering now? It and design are two fields with a high degree of overlap, but compared to design, engineering emphasizes more on scientificity and reality, that is, design can design something based on any value orientation, but engineering emphasizes that the thing you design is a structure that can operate in the real world, and this structure must create some kind of utility or welfare for people.

If you want the structure you design to operate in the real world, it must be scientific and must go through a rigorous process. And it must be accepted by the real world, it should be legal, economically feasible, and actually acceptable to your target users at the moral level. In addition, a very important sense of responsibility in engineering is that as engineers, we must take responsibility for the consequences of our work in society. That is, we must do engineering with a responsible attitude.

II. Analysis Model

1. Two Thinking Models

The following are two models for the design of universal economy. The first one is the top row in the table, which is the token system to motivate individual behavior, so that these individual behaviors can form different contributions and ultimately achieve a common goal. The advantage of this model is that it is very versatile. It covers all kinds of potentials of what the token system or encrypted network can do. The token-related practices that many people are doing now are not limited to the Internet, or even limited to the economic field. It can be extended to the field of social governance.

But its disadvantage is that it is a very new thing from this perspective. From the perspective of economics, it can be considered as an extension of new institutional economics, and some practices can be done using papers and mechanism design methods. I have tried to learn it before, and I feel that this path is at least difficult, or rather roundabout. Because it is too abstract, and there is too much to learn, and there is a lack of actual cases. For example, when we talk about POW, many scholars are studying it after Bitcoin is up and running, and then they talk about it from the perspective of mechanism design, what kind of thing it is, and how it achieves equilibrium. But these are actually some retrospective studies. First there is the Bitcoin network, and then everyone analyzes why it is feasible, its advantages, why it can achieve equilibrium, etc. But can this method first get the goal, and then use the mechanism design method to get a system. At present, I don’t see such a way.

The second thinking model is what I proposed, which narrows the scope. Let's not think that tokens can do this or that, we can say that tokens can enable crypto networks. What is a crypto network? It is a platform owned by the community, so web 3.0 is the natural evolution of the web2.0 platform economy. The original core role of the platform economy is the Internet platform owned by the company.

What is web3.0? We evolve the core role into a decentralized platform owned by the community. The advantage of this thinking model is that it can get nutrition from the platform economy. Because the economic characteristics of web2.0 have been studied for 10 years, there are many research results, including talents. These knowledge and talents can basically be inherited in the web3.0 era, that is, how to analyze platforms, how to analyze bilateral markets and multilateral markets, how to analyze network effects, how to start network effects, etc. We will talk about it later, and these things can be inherited.

What is the problem with the second thinking model? Our original platform economy was studying how to build a moat for the platform. Web3.0 does not talk about moats. The idea of web3.0 is exactly the opposite. The more commission a platform economy gets, the greater the opportunity for web3.0. Because web3.0 is to return the created value to the participants as much as possible. For example, for taxi platforms, it is drivers and passengers, and for music platforms, it is musicians and music fans. We want to return the economic value added of the platform economy to these people, rather than leaving it on the platform itself to create profits for shareholders. But at the same time, we also hope to create defensibility, that is, to maintain the existence of these factors in the encrypted network, because the essence is still to hope that this network can grow bigger and bigger.

So Web3.0 and Web2.0 have different values and goals. But the actual methods used are very similar to web2.0. Is there any essential difference here? Is motivation important? I think it is still very important. If we want to succeed in web3.0, the values of the project founder are very important. From a practical point of view, if you want the crypto market to agree with your project, what can you rely on to make everyone agree with it?

I think there are two most powerful forces in this market. One is the speculative force. If you can let others see that your model is very powerful, a large amount of funds will flow in, your currency will rise, and your goal can be achieved. In fact, there is a stronger force. Many of us practitioners ignore how this market crypto market came about. The first group of participants in the establishment of the market may be some libertarians or anarchists. These people do not enter the market entirely for profit. And precisely because they are not purely for profit, these people are more able to obtain huge profits and experience many market fluctuations because they have faith. Those who can get BTC from 2011 or 2012 to now have increased their wealth by tens of thousands of times. After they have money, at least a considerable number of people say that I will support those things that I like.

If your project can resonate or resonate with some of these people in terms of values, it is very likely to succeed in the crypto market. We currently do not have a recognized valuation standard for crypto projects, and most people are guessing what others think. But some people do not guess, they are based on their own values: I think your project is good, I think your project is interesting, or I think your project promotes the direction of society that I want to see, so I will support you. After these people make their own choices, others will guess based on their logic, and further reach a consensus.

So this point is actually what I think many projects have overlooked, especially some people who have entered this circle from the Internet or technology field. They think this thing is too stupid. You are actually not good enough! I understand the Internet better. But these people are not crypto natives. He does not know what we, the people in the currency and chain circles, think, so usually these people's crypto startups will not succeed. Your understanding of the market, your understanding of the market participants, your understanding of their moral goals, and your ability to resonate with them are very important. But these things are not in the theory and practice of Platform economy, which are all about competition. They are all about how to create a moat, how to defeat competitors, how to maximize profits, or how to subsidize first and then take commissions, etc. The goals of Web2.0 are different from those of Web3.0, and we cannot copy them.

2. Network effect analysis

Network effect is the core of platform operation. We have now narrowed the scope of discussion to the token economy as a decentralized platform. If the platform has no network effect or the network effect is very weak, it is not worth doing.

So you have to analyze what kind of network you want to build, what the network effect is, whether it can create enough network density, create enough value, and then break through the critical scale and enter a stage of spontaneous development. And it will not encounter anti-network effects too early in the development process. Anti-network effects simply mean that the more users there are, the worse the service will be. Usually there are two situations. One is network congestion, which we have seen in Ethereum. Because of the technical architecture, the more people use it, the higher the cost will be. The other is network pollution, which means that the platform will become watered down after users come in.

The negative effects of many new users outweigh the positive effects. These are areas where the platform economy is very mature. Sometimes after you analyze it, there is no way to change the internal mechanism of the network effect, because it is determined by the market, or you have determined what market to enter. Sometimes its own network effect is solidified, but it is definitely better to know it than not to know it. Other practices can help you better build network effects and reach critical mass faster.

Three, Token Design

1. Token Distribution Design

Next, we will focus on tokens. There are two prerequisites for designing a Token distribution model. One is that participants must contribute to the operation and development of the network. The best case is that these contributions can be verified on the chain and can be quantified, and corresponding contribution rewards can be allocated in the protocol. A network may need multiple contributions to run. For example, the simplest Bitcoin requires people to run nodes and do software development, both of which are indispensable. However, software development contributions cannot be verified on the chain, that is, contribution cannot be proven, but POW can be proven. Therefore, the Bitcoin network can only reward contributions that can be verified on the chain within the protocol. Some contributions cannot be rewarded in the protocol.

What should we do then? We need to use governance based incentives. Generally speaking, if the protocol cannot be stabilized quickly at once, you should not quickly hand over governance rights to the community. Instead, you should maintain the efficiency of the core team's decision-making. The core team decides how much money to spend on marketing, or how much salary to pay themselves, which is actually governance based incentives.

However, governance based incentives have limitations. Daily incentives should not be governance based. If you need to first decide who to reward every day or every week through governance, this network will not work. Why? Because the efficiency of community governance is always lower than the efficiency of corporate decision-making. Many things should be done in a corporate way and solved with corporate management systems, rather than encrypting the network.

We say that the protocol is fully automated in terms of rewarding contributions, and it is open and transparent, so it has transaction cost advantages in this field, so it is established. If there is no transaction cost advantage relative to enterprises or platforms owned by enterprises, then there is no need to do web3.0.

Provable contribution design

So provable contribution design is a prerequisite. You have to think about how to verify the contribution that is most important for the daily operation of the network on the chain? The liquidity that DeFi needs most is on the chain itself, so the contributions of provide liquidity can be verified and quantified.

value capture design

Another prerequisite is value capture design. We say that tokens must have value. The most common indicator of value is the circulating market value, which is the market's measure of token value. However, the unit price of the token is also very important, because the use of all tokens is based on the willingness to hold the token. Therefore, if the growth rate of the value of the token is far behind the speed of issuance, all market participants will have no motivation to hold the token, and this model will definitely not work. Therefore, in the long run, the growth rate of the token value must exceed the speed of token issuance.

How does the value of the token grow? There are 4 ways. The first is that the cash flow or assets that the token can share are increasing. The second is to create purchases for the token. The third is to eliminate the token from circulation. The fourth is that the rights attached to the token are increasing. These four are not mutually exclusive and can be used in combination. To design a value capture mechanism, it is based on one or more of the four. Only after these two conditions are met can token distribution design be done.

Subsidizing strategy

The first step is called subsidizing strategy, which is a subsidy strategy. Why do we need subsidies? Because the crypto network is a platform economy with network effects. Generally speaking, if such an economy is a two-sided market, one side is usually more difficult to obtain. We call those service providers difficult to obtain sellers' markets, that is, as long as you find sellers, buyers will naturally follow. For example, the taxi-hailing mentioned earlier, such as Didi, subsidized the drivers at the beginning, and then as long as the drivers came up, the passengers would follow. For example, Dianping and Meituan Group Buying, in the early days, they all subsidized merchants, and there was naturally demand when there was supply.

Of course, there is also a buyer's market, that is, the group of people on the demand side are more difficult to obtain. For example, task-based websites or, it is more difficult to obtain the demand side. You need to find a strategy to find the demand side first. In the platform economy, it means that I provide subsidies to the seller or the buyer. In Web3.0, you reward the token to the person who provides the service or the person who consumes the service. This is the subsidy strategy. The network effect analysis determines which side to subsidize with the token.

contribution weighting

The second is contribution weighting. For example, maintaining a network requires different types of contributions. These contributions are all necessary for the operation of the network. How to determine the relative weights between them? This seems a bit difficult. For example, maintaining life requires both air, water, and food. Which one do you think is more important? It seems difficult. I found a relatively simple method: in the decentralized platform economy, it is usually possible to indentify the role of an owner, that is, who is the owner of the platform. Generally speaking, he should be the person who provides the service. For example, in a taxi chain, the owner should be the driver. If it is a music platform, the owner should be the musician. You can regard the incentives obtained by other roles except the owner as the cost necessary to maintain the network.

If you regard it as a cost, the lower the better. With this set of rules, it is easy to deal with. You can see that in order to attract these people to join the network in the market, for example, if he is a validator to maintain the network operation, how high is the cost of running the machine to run this node? How much is his professional knowledge worth? What is a reasonable profit margin? How much profit can he get from similar services? Just provide a competitive price. So in this way, that is, other than the owner, you can use the cost analysis method or the horizontal comparison method to determine the incentive level. Under the premise that the growth rate of the token value is lower than the growth rate of the circulation volume as mentioned above, the rest should be sent to the owner because they are the owner of the platform.

benefit from token price fluctuation

Then benefit from token price fluctuation means that the token economic model must be able to withstand the fluctuations of token prices. Price fluctuations are endogenous, and all cryptocurrencies have a very high corelation with the big market. The entire cryptocurrency market is fluctuating violently, and individual tokens fluctuate on this basis, so skyrocketing and plummeting are the norm. Your token economic model should at least not be harmed when the token price soars or plummets. That is to say, if the token price soars, it should not affect the cost of using the platform.

Another direction is that if the token price continues to fall, will it become a death spiral? That is, everyone rushes to sell, and finally the coin returns to zero. Then it is difficult to restart the network effect, which is something to consider.

Is it that anyone will buy when the price of the currency drops sharply? If your users want to use the necessary services, rather than just buying the currency for investment speculation. Even if the currency drops, the utility is still there for the user, and the price of the legal currency may not change, the user will still buy to use the currency.

For the service provider, the incentive he gets is not only the token issued by the network, but also other cash flows. Tokens can usually be designed as the working capital of the service provider. That is to say, holding more coins can gain a larger market share or service opportunities. When the price of the currency drops sharply, those who are in good operating condition, good financial condition, and want to expand their business will buy at the bottom. Those who are in poor financial condition and whose services are unpopular will naturally exit the market. Price fluctuations enable the platform economy to achieve the survival of the fittest, which is a good thing.

diachronic fairness

Another one is diachronic fairness, which is fairness between old and new users. This is very difficult. To achieve fairness, it is theoretically very easy, that is, the incentive of the additional tokens according to the current market price, that is, the marginal value, should be equal to the growth of the intrinsic value of the current network.

If this equation can be maintained all the time, it will be fair to both old and new users, but this is very difficult because there is no way to accurately calculate the value of the network.

Generally speaking, the token economic model will favor old users. That is, the first person to enter the network, because it needs to be started, so it is not biased towards him. If no one pays attention to you, the startup will be very slow and may even cause the network to fail. But if the network is lucky enough to get better, new users will feel unfair when they come in. Even old users monopolize the benefits, so that new users have no motivation to come in again. This has happened in many encrypted networks. That is to say, the long-term nature of the mechanism gives new blood the motivation to continue to come in.

token ditribution plan

After doing these things, you will get the token distribution plan. That is to say, the token plate is cut into large pieces. Which piece is for investors? The funds they provide are the main source of governance based incentive. Then what is the rate of increase? And which ones are airdropped or other ways to incentivize the community. This is the output of token distribution design.

2. Basic Token Model

We propose a basic model. There are three types of participants in the crypto network: Keepers, Providers and Consumers. The crypto network may have more than one type of keepers. Keepers allow the network to run normally according to the designed state and ensure that it is safe. Service providers provide services to service consumers through the network. There should be engagement between consumers. For example, consumers can use viral marketing, or make a contribution to make the service experience better or more valuable. These three types of participants get tokens according to their contributions.

Generally speaking, in order to complete his transaction, the consumer either takes out a portion of the value from the payment and puts it into the treasury, or completes the transaction by locking or burning a portion of the token. Generally speaking, the provider has to do staking, which means that because it holds tokens in the system, it can provide more services. The above is a basic model, but it is definitely not universally applicable.

3. Token Engineering Process

Token engineering process, the first step is to analyze the problem, what kind of platform do you want to build? What is the value of the platform? Who are its participants? What transactions occur between them? What are the limitations of the platform? With basic analysis, you can do network effect analysis. Then you have to classify the encrypted network. There are different categories, which probably explain the name. After classification, you have to find all the main projects in this category and study their white papers. Find research reports in this field. Know enough about this industry and this field, otherwise it will be difficult for you to convince investors, and it will also waste a shortcut, which is to find the best design so far. You look at all the existing designs, which can save time, because each project spends a lot of time to get its own model.

After having this foundation, you can start designing. First, you need to choose a basic model. This basic model can be the one you admire most or the one you think is the most suitable among similar projects. If you don’t have one, you can start with the general basic model I just introduced. Then do token distribution design, which is the process we talked about earlier. Then some specific problems can be solved by using the token design model. After solving these problems and getting a basic model, you need to do attack & defense. That is to say, find some knowledgeable friends to assume various attack methods. The most important thing is to contribute to the coins, but it does not really contribute value to the grid. You need to solve these problems by adjusting the model or adding new constraints.

Finally, it is verification. After having a preliminary model, you can look back and see if your goal can become bigger, more valuable or more realistic. It is even possible that you need to do a major reconstruction. Can you introduce a new participant to make the platform more valuable, or reduce a participant to simplify the model, or increase or reduce transaction types, etc. This is an iterative process, which is very similar to software engineering.

4. Token design pattern

Every engineering field has its own basic component library, and we hope that the field of token engineering also has it, that is, the token design pattern. There are many token design patterns that have been identified, but their maturity is definitely far behind other fields. And I personally doubt whether every designer of a crypto network should master all token design patterns. After all, most people will not have more than one opportunity to design a token economic model. Perhaps a better way is to seek help from outside.

External aid actually already exists, because there are some people, such as researchers from investment institutions. They have to look at a large number of projects, see a lot, think a lot, and help these projects after investing. They have enough motivation and resources to invest in studying token design patterns. Learning has a scale effect, and after the investment is converted into professional knowledge, it can help many projects. There is also an ecosystem like ours, where there will be many projects. We can also spend time to sort out these things, and then help many projects reduce their investment in this area and avoid detours. Today I have exceeded the time limit a lot, and this is what I want to talk about.

Original link

欢迎加入律动 BlockBeats 官方社群：

Telegram 订阅群：https://t.me/theblockbeats

Telegram 交流群：https://t.me/BlockBeats_App

Twitter 官方账号：https://twitter.com/BlockBeatsAsia