Dialogue DODO Radar Bear: DeFi has too much illogicality, we should create permissionless products and realize their value

This article was published in the Orange Paper, original title: "Dialogue with DODO: Is There Too Much Contradiction in DeFi? Because You Chose the Wrong Logic Template," Interviewee: Radar Bear, Co-founder of DODO.

One night last month, DODO finally listed on Binance, and its market cap shot up to $800 million. Many people were celebrating, and everyone was excited. I was also surprised; although there are many wealth stories in the crypto world, it is still rare to go from zero to $800 million in just six months.

The next day, I jokingly asked DODO's founder, Radar Bear, what it felt like to build an $800 million public company in just six months.

Radar Bear said he hadn’t had time to think about it yet; they had been hit hard by a DDOS attack the night before and had to fight it off all night.

Me: ?

Radar Bear is a sturdy young man from Northeast China, with a baby face. He speaks with great energy and always adds the condition "I think" when making conclusions.

We met in a WeChat group in early 2019, and I thought this programmer was different; he spoke plainly and was very sharp. As we got to know each other better, I learned that he studied physics at Peking University, dropped out of his direct PhD program, and came to work in blockchain, with strong beliefs.

I still remember the toughest time in March 2019 when we held an offline gathering. Radar Bear was there, and he asked me how long I planned to stick it out. I paused and said I would hold on until I ran out of money. He laughed and said he felt the same way, sticking it out until he couldn’t anymore.

Fast forward to the summer of 2020, during the DeFi bubble, the atmosphere shifted from extremely cold to extremely hot, and everything accelerated. Radar Bear and DODO shot up like a rocket: the idea came in April, they found their co-founder in May, started construction in June, launched the first version in August, and raised $5 million in September, with investments from Binance and Coinbase.

When we met again, it was already October in Shanghai. That night was a Binance event at the Shanghai Revolutionary History Museum. The two of us, country bumpkins from Beijing, marveled at Shanghai's prosperity while discussing DODO's mechanism design. Radar Bear was still the same honest and sharp-tongued programmer, the only difference being he wore a nice coat.

We chatted several more times, addressing many of my questions. I wanted to document these conversations because it’s likely no one else will do so: those who know don’t speak, those who speak don’t write. I understand all of this. To preserve his Northeast humor, I intentionally left many sentences in a colloquial form. Here are his candid words:

In a Suppressed Global Financial Atmosphere, a Free Market is Even More Necessary

Orange Paper: You previously asked me a question: "Does blockchain really solve problems, or does it just create wealth?" Why did you ask this question?

Radar Bear: When blockchain creates wealth, people believe in it; when it doesn’t create wealth, they don’t. So, do you think it really solves problems?

By normal logic, it should be that if blockchain solves problems, people believe in it; if it doesn’t solve problems, they don’t. Currently, blockchain has created wealth. However, we are internally hoping it can solve problems.

Existing trading and lending are just going in circles and do not connect with the real industry, so they are not truly solving problems; they only address issues within our system.

I believe the greatest significance of trading is to realize social division of labor, thereby bringing about more social production. With trading, a company can focus solely on producing zippers, and its employees can afford to eat and wear clothes. Without trading, they must produce both bread and clothing simultaneously.

The current issue is that we are trading digital currencies, which have no relation to daily necessities. Perhaps it brings you joy, but aside from potential wealth creation, does it change your life? Of course, if you trade cryptocurrencies and make money, then use that money to buy various things, it indeed indirectly changes your life.

In my view, when does blockchain count as solving problems? When it starts to touch the real industry. For example, if a factory raises funds by issuing tokens and then produces many clothes to meet consumer demand, and those consumers can use the factory's tokens to purchase clothes at a discount, I believe this closed-loop behavior solves problems.

The reason this model doesn’t exist now is that technology may only be a small part of the reason; more importantly, it’s political. Issuing tokens for financing is very close to issuing stocks. If the threshold for issuing stocks is so low, the market would be too free. Is freedom necessarily good? People in our industry might think so, but in reality, the disastrous consequences of a completely free market are not few. This is the conflict between ideals and reality, logic and experience.

Teacher Luo Xiang said that the life of law is experience, not logic. I think finance is the same. The life of finance is experience, not logic. Logically, you can assume everyone is rational, but from experience, it doesn’t work and can lead to disastrous consequences. Therefore, many people say that blockchain can only be believed logically, but from experience, it must be strongly regulated.

Orange Paper: The scale of finance is indeed hard to grasp. Which do you think is better, regulation or no regulation?

Radar Bear: I think it’s hard to say. The downside of regulation is that it blocks some truly excellent small teams from financing.

Currently, financial regulation is tightening globally. If you want to go public for financing, you must pay an extremely high cost. In this suppressed financial atmosphere, a free market is even more necessary. That’s why we insist on doing permissionless DeFi, even though I have complex feelings about Ponzi schemes, because I believe there should be such a free market. It may ultimately not work, but it cannot not exist.

We want to create an exchange with the lowest threshold, issue assets, and provide liquidity for others to buy your assets. Why say let a hundred million people issue tokens? Because issuing tokens on DODO is incredibly easy. You see, initially, listing tokens required centralized exchanges for custody, paying listing fees, and finding someone to help you with market making. Later, with Uniswap, listing fees were no longer needed, but you had to provide liquidity, and you lost the freedom of market making. Now we have released the freedom of market making as well.

In the new version V2, you can go from "wanting to issue tokens" to "having a market" all at once, without spending a penny, and it’s permissionless. This is what I mean by "let a hundred million people issue tokens." Uniswap cannot do this because ordinary people who issue tokens don’t have the funds to provide liquidity; only specialized project parties, either with money or who can engage in dual pool mining and community efforts, can have enough liquidity for others to buy their tokens.

I believe the vitality of DeFi comes from freedom; any design that is not free goes against the original intention of DeFi.

The Love-Hate Relationship with Ponzi Schemes

Orange Paper: From 2019 until now, how has your view on DeFi changed?

Radar Bear: There hasn’t been much change; the current DeFi is pretty much what I imagined back then. I thought smart contracts were incredibly useful; as long as they don’t violate market rules, you can create whatever you want, but complex products take longer to develop. Another thing is permissionless, turning the seemingly lofty financial system into something everyone can participate in. I think this is a very great initiative.

Speaking of changes, I recall something. In the past, I was very opposed to Ponzi schemes. But later I found that what truly made people recognize the goodness of DeFi was precisely Ponzi schemes. Without the allure of Ponzi schemes, it would be hard to attract people to try this new thing called DeFi. Moreover, many times, Ponzi schemes are just a small part of a complete product; many users are drawn in by Ponzi schemes and then slowly begin to understand other aspects. So now I have a neutral attitude toward Ponzi schemes.

Looking back, there are quite a few Ponzi schemes in the world; I guess it’s a way to find psychological comfort (laughs). Stocks, bonds, social security—these games all have elements of Ponzi.

My feelings about Ponzi schemes are complex. On one hand, they are indeed useful; on the other hand, they don’t quite align with my values and have no relation to the real economy, and sometimes even harm the real economy.

Sometimes I comfort myself by thinking that Ponzi schemes are like variety shows; people come to play. Does playing have an impact on the real economy? No. Does that mean playing has no meaning? Impossible. For many people, playing is the meaning of life. Some people find Ponzi schemes particularly interesting, much more enjoyable than a big meal. From this perspective, isn’t Ponzi also meeting people’s real needs? These thoughts confuse me greatly.

Orange Paper: I can relate; I also strongly oppose gambling. How should gambling be defined? Is it gambling as long as there is speculation?

Radar Bear: Anything with randomness can be called gambling. From a certain perspective, what we are engaged in is the gambling industry. Interestingly, after so many years of evolution, only 21 games have survived in the gambling industry, with well-known ones like "Zha Jinhua" and roulette. Why? Because they are simple and fun.

I feel the crypto world is also one of these 21, or rather, the financial system represented by the secondary market is one of them. Buy, then watch it rise and fall—simple, stimulating, and fun.

But you can’t deny that the financial system is very helpful to the real industry. The Soviet Union wanted to replace the speculative financial market with a centralized planned economy system, with a huge department calculating how to allocate resources. This is very idealistic and admirable, but this system doesn’t work. Although speculative activities in the financial system may annoy some people, the resource allocation facilitated by this speculation is more effective than a planned economy.

Many Projects’ Original Business Logic Doesn’t Work, Yet It Mysteriously Meets Another Kind of Demand

Orange Paper: How do you think Uniswap became so successful?

Radar Bear: The dual pool mining played a significant role; essentially, it is a fancy Ponzi scheme. Although I don’t like it, I have to admit that all fancy Ponzi schemes are very successful designs because they can attract many people to use them. When Uniswap first came out, not only domestically but also internationally, it was not well received. First, asking retail investors to provide liquidity is an unbelievable act; how could anyone take on the risk? On the other hand, you say you are permissionless for listing tokens, but after listing, there’s no money for liquidity, so there’s no liquidity. If you list a token but can’t trade it, what’s the use?

So people concluded that Uniswap was useless, and indeed, at that time, it was truly useless. But when did it become useful? When people developed the design of dual pool mining. Note that this design was developed based on the Uniswap infrastructure, which is also why I think permissionless is a good thing. Because the founder of Uniswap himself didn’t foresee the dual pool mining, but others developed it, and this model suddenly became popular.

Although dual pool mining is a Ponzi scheme, it effectively solved the liquidity problem for long-tail tokens. Any long-tail token can create a liquidity mining program, providing price support. Many long-tail tokens initially had no utility, but dual pool mining provided a very simple utility: you can use my token to play Ponzi (laughs).

So in my view, Uniswap has been freeloading from the beginning. Its product is indeed quite good, but does it deserve the massive scale and status it has now? I feel it didn’t come from its product alone, but from the countless people who developed on it.

This is also the core demand of our new product. I give you a toy that is much more flexible than Uniswap; if you can explore various models on your own, that’s the best.

Orange Paper: Was there no competition for Uniswap at that time?

Radar Bear: Because there was no one competing with it in that field (laughs). Let’s put it this way: AMM without dual pool mining loses 95% of its use cases. With dual pool mining, it becomes particularly attractive, and everyone rushes to do it; this is the very real situation.

So it’s very possible that you create something permissionless, and for a long time, everyone thinks it has no value, or it may never have value, but this is definitely a correct path; you just keep making such tools. Of course, it’s not just about making tools blindly; you need to identify the market pain points first, then create a tool that maximally satisfies market demand with freedom, helping some players find their preferred models.

Orange Paper: This is quite magical; it’s very different from internet entrepreneurship. Internet entrepreneurship generally emphasizes demand-driven approaches, seeing an unmet demand and then creating a product to satisfy it. What demand did Uniswap initially aim to meet?

Radar Bear: Its initial purpose was to meet the demand for DEX, but everyone thought this solution didn’t work. You see, it now has good liquidity for long-tail tokens and many users, all thanks to dual pool mining. After completing the cold start, it gained massive traffic, and now a large number of bots use it as infrastructure; this is a positive cycle. It has high traffic, and good liquidity, and on-chain bots connect to it, leading to a snowball effect.

Orange Paper: How long after Uniswap's launch did dual pool mining come about? A year?

Radar Bear: We knew about Uniswap in early 2018, and I even wrote an internal analysis report on the Uniswap model, pointing out that this model didn’t work at all (laughs) because providing liquidity as an LP incurs significant losses. I couldn’t understand why someone would provide liquidity for long-tail tokens; even with a 0.3% fee, could they earn it back? Especially since long-tail tokens have low trading volumes.

Orange Paper: In the crypto field, is it impossible to rely entirely on logic for entrepreneurship? The Uniswap example is quite illogical.

Radar Bear: In this field, your logic needs to undergo some transformation. Your goal should be to create the most permissionless thing to realize its value, rather than imitating centralized exchanges.

When you apply the business logic of centralized exchanges to the DeFi world, you will find many things that seem illogical, but in fact, the reason for the problems lies in your choice of the wrong logic template.

Orange Paper: Besides Uniswap, what else seems unreasonable but is actually reasonable?

Radar Bear: MakerDAO. I first learned about MakerDAO at the end of 2017, when the first version of Dai hadn’t launched yet. After reading their white paper, I concluded that I couldn’t understand why anyone would collateralize ETH to borrow, and still pay interest. They answered that these people might have ETH they didn’t want to sell, so they collateralized some to borrow Dai for daily expenses. I asked why they didn’t just leverage? Why hold the spot and pay high interest to borrow money?

At that time, I thought, my goodness, this makes no sense. Who would use such a stablecoin? Later, as Maker scaled up, I was always puzzled, not knowing who was borrowing; it felt like no matter how I thought about it, minting Dai was a losing deal.

Later we realized that some people just play with this; they don’t leverage on centralized exchanges. Some people only live on-chain and completely distrust the centralized world, and what they do in Maker is precisely leveraging; that is the first leverage tool they can get in the blockchain world.

Although it’s very crude and silly, requiring multiple rounds of borrowing to achieve a possible 1.5x leverage, it’s quite funny and a silly method, but in fact, it’s the only leverage product available to them on-chain, so many people use it.

In the early days, it was mostly Ethereum whales who naturally trusted only the Ethereum network, so they played here, and a few big players supported the entire Maker production system, minting the initial batch of large Dai.

If you ask whether this thing is logical, at first glance, as a stablecoin product, it doesn’t conform to business logic. But in fact, it is a permissionless leverage product, and its business logic works.

Radar Bear: Yes, many people in this industry do not play by the "rational person" rules; they have some of their own stubborn beliefs.

Then there’s Synthetix, which also went through a death spiral but survived, and then it turned out that after surviving, this system could actually work.

When Compound first started, we thought, who would exchange a highly liquid asset for another highly liquid asset? This doesn’t conform to the essence of the lending business. Later, we understood it was still about leveraging. At that time, Maker only supported ETH collateral, while Compound was the only one supporting multi-token collateral, so it also became popular.

Many projects’ original business logic may not work at all, but it somehow meets another kind of demand, leading to another way of use developed by people. This is not the application scenario it initially recommended, and it hasn’t provided any facilitation services, but its permissionless nature allows it to do this work.

So we believe that letting everyone create is a very worthwhile thing. It’s possible that your project initially doesn’t work, but precisely because you allow others to play with it in any way, it’s like the map editor in Warcraft; although Warcraft has declined, Dota in the Warcraft editor can still remain popular.

Ethereum is similar; if you look at the initial use cases written in the white paper, there were no ICOs and no dual pool mining, but who could have predicted that these two became killer applications (laughs)? This is why my feelings about Ponzi schemes are so complex; they work too well (laughs). They do not belong to any idealist, nor are they the original intention of engineering creators, but they are indeed the most effective things.

All the things you thought would definitely be useful ended up not being used by anyone. If you say you want to create something particularly concrete that will definitely be popular, I see the probability is quite low. But if you create a platform, a tool, that is the correct direction. To some extent, this represents the direction of advanced productive forces because it can unleash user creativity. The bullet comments on Bilibili are a positive example, while the negative example is the skits from the Spring Festival Gala.

We Are More Willing to Speak Our True Thoughts and Have Opportunities for Trial and Error

Orange Paper: You are only 27 years old now, and even among DeFi entrepreneurs, you are very young.

Radar Bear: Yes, Dai Dai might be the youngest. We sorted it out before; CZ started Binance at 40, Robert started Compound at 46, and Sergej from 1inch seems to have started at 30 during a hackathon. Dai Dai was particularly anxious before, staying up late every day. After checking the ages of these people, he felt much more relaxed (laughs).

The advantage of youth is that there are not too many concerns; we don’t have wives and children. We are more willing to speak our true thoughts and have opportunities for trial and error.

Orange Paper: Yes, I think trial and error is quite important, especially in the DeFi world. You just mentioned a lot of logic, but I feel that this logic is hard to predict.

Radar Bear: Yes, when it comes down to whether something works, no one knows. There were many models that everyone thought were very effective back then, but in the end, they didn’t pan out. Several projects that were particularly popular back then, like Augur, everyone was very excited about, and then there’s EOS, right? It’s really hard to predict.

This industry doesn’t fear your failure; it fears that you won’t persist. You see those projects that aren’t doing well; without exception, the founding teams have given up. As long as you persist, there will still be support.

I have speculated a lot about why many domestic projects don’t create new things themselves; perhaps they lack the confidence to issue a contract they wrote entirely themselves, fearing problems. This indeed carries a lot of psychological pressure. If something goes wrong, it’s a huge blow, so most projects choose to fork already market-validated code first.

This is also quite frustrating; doing existing things is too boring and meaningless, so we can only bear this risk.

In fact, abroad, this kind of risk-taking behavior is given relatively high recognition. They believe that as long as you release something new, regardless of whether it’s good or not, regardless of how many users it has, they admire you.

Our foreign fan base is growing quite fast; we now have 60,000 Twitter followers.

Orange Paper: Do you have any KPIs?

Radar Bear: Our KPI is DAU. Let’s put it this way: trading volume and locked value can be easily inflated; you can amplify mining rewards by 10 times, and immediately the locked value will increase. If you do trading mining, people will come to arbitrage, boosting your trading volume. The only thing that’s hard to inflate is DAU, which is the most valuable resource.

Speaking of the current DeFi rankings, I really don’t think it’s necessary to take it too seriously. For example, a transaction on 1inch might be counted in the trading volumes of 1inch, Sushi, and Uniswap simultaneously, so some transactions are inflated and counted multiple times.

Currently, no website has made a genuine judgment on the trading volume of DEXs because people assume that everything on-chain is real. All websites are listening to events, thinking that events are immutable data on the chain, but who says only transactions can trigger events? In fact, you can trigger an event without a transaction and directly record $100,000 on a market site.

Therefore, I don’t care much about these metrics; what I care about most is DAU—how many real users you have. This metric is very hard to inflate; it can be inflated, but it’s very expensive. You create many new addresses, then send some ETH to each address, letting them interact with you; all of this costs real gas fees, and you can see which addresses are being inflated, as these users don’t interact with other addresses.

What we hope most is that these real players in the market all use us; this is more important than anything else.