Original author | The Decentralised.co
Compiled by Nan Zhi (@Assassin_Malvo)
In the nineteenth episode of The Decentralised.co’s Podcast “Building on First Principles”, hosts Joel and Saurabh discussed with Pacman, the founder of Blur and Blast, the origins of Blur and Blast, Pacman’s judgment and analysis methods on the market and products, and Pacman’s judgment on the crypto market cycle.
Odaily Planet Daily will compile the key parts of this article. In order to control the length and ensure the reading experience, half of the content has been deleted. Interested readers can listen to the full content in the original podcast.
Viewpoint Overview
The problem with OpenSea and other NFT markets is that they position their customers as consumers, and they really think that NFT is a product that you like to buy, hold and buy. Pacman believes that a large part of users should be defined as traders. (Odaily Planet Daily Note: Just like what Pump is doing now, the community culture characteristics of Meme are removed, and users only focus on trading tokens with various different names.)
The problem with designing an incentive system is that it is difficult to avoid being manipulated (referring to large-scale Sybil attacks). In conventional retroactive airdrops, users will do a bunch of random operations, which Pacman believes is a huge waste.
Liquidity cannot be forged like the number of addresses, and regardless of whether Farmer exists, it provides liquidity for the entire ecosystem, so Blur only incentivizes one thing, which is liquidity.
The significance of the points program is that the project party can directly guide users (avoiding users' random operations) and accurately track users' performance.
Blast still targets the gap in the market: passive sound of underlying assets and gas income of developers.
The big cycle of the crypto market has not yet begun and is still waiting for a fundamental innovation.
Interview Highlights
How did you get started trading NFTs and plan to start Blur?
I started really getting into NFTs in 2021 after selling my first startup to namesheet and minted a blitmap as my first NFT which I ended up selling for about 25 ETH. So after that, I got really into NFTs.
I started trading NFTs regularly. I wrote scripts to analyze metadata and I really fell in love with it. As I was doing that, I noticed that the infrastructure for trading NFTs was very suboptimal, everything was slow, it kept crashing, and I needed to use dozens of different tools. So as a trader, I really knew I could create something better, so I co-founded Blur with my co-founder.
What do you think was wrong with OpenSea, the market leader at the time?
Interviewer: When you were thinking about building Blur, OpenSea had over 80% of the market share, what were the plans to replace it?
Pacman: As a trader, the experience with OpenSea was terrible, it was too slow, too painful. You talked to a few other people and a few other traders, and almost everyone had the same experience. From a user perspective, it was clear that something better was needed.
As a trader, I usually want to see a lot of information, like rarity data, sales data, listings and bids. I need a lot of information for traders. If you trade tokens, you may have experienced this interface. If you use Coinbase, you have this real-time trading experience where everything is updated instantly, it's very information dense, and you use all of this information in your trades.
This is not the experience on OpenSea at all, it's more like a marketplace, like eBay or Amazon, where they (users) are shopping. The mistake they (developers) each made is that they really think of NFTs as a product that you like to buy, hold, and buy. So they really treat it as a shopping experience, and I know as a trader that I like to buy NFTs, but there is also a large portion of users who buy and sell these NFTs who are actually more like traders, so we built for this group of people when almost no one else was.
I think it's no secret at all, a lot of the NFT markets, even the people at OpenSea, they're capable people. They have very deep insights into the data team, so they know where their volume is coming from and what the numbers are. I think a lot of it is just cultural, like Apple, at least under Steve Jobs, was very different culturally than Microsoft under Bill Gates. So a lot of times it's not that there's any unique information, it's just that the information is viewed differently. So I know if you look at OpenSea's behavior, and their public comments and things like that. They really just ignore the trader perspective. I think they think that's irrelevant, or they take it for granted that their primary customer base is shoppers. And I look at it completely differently because I'm more of a trader.
How is Blur’s underlying incentive design considered?
What’s interesting about Blur is that it was basically planned from day 0. We knew we wanted to build a marketplace for professional traders. We wanted to use a clear points program to incentivize behaviors that we thought would drive market growth, and this was all conceptualized before we built or started writing a single line of code.
Other projects airdrops are basically retroactive airdrops, every protocol that releases a token, basically never mentions the token, they just hint at it, and then users do a bunch of random actions in the hope of getting the airdrop. Then the airdrop is at some point in the future, sometimes it's a few months, sometimes it's a few years. Imagine you're a startup, you're acquiring users, and then after you acquire them, you have to pay them. This retroactive airdrop model makes no economic sense to me.
So let's be clear, if users are going to be incentivized through airdrops, they're going to put a certain amount of effort into getting the airdrops. Let's focus that effort on things that we know are valuable, rather than having users guess. Because having users guess is just a distraction, let's make it very explicit so they can accurately track their performance. From first principles, it really makes sense to make an explicit airdrop rule.
The problem with incentive systems is that they are hard to design, and it's especially hard to design an incentive system that can't be manipulated. A lot of the most common incentive systems are when projects do non-linear rewards, and all it does is incentivize users to use 10, 000 addresses and farm on 10, 000 addresses, and it doesn't actually achieve the goals you want it to achieve. On the one hand it does sometimes benefit the project, and their user numbers are really increased in a completely artificial way. But I don't think it's worth it.
The Blur incentive system incentivizes only one thing, and that is liquidity. The problem with liquidity is that you can't fake it. You can start 10, 000 addresses pretending to be 10, 000 people, but you can't directly start $10 million deposited into the Blur Bid pool. So even if a Farmer-type participant joins, they can only participate in a way that we have pre-defined as being good for the market, which is liquidity. The existence of people like Farmers is not a net negative for us because the incentive system is designed with these limitations in mind.
The origin of Blast
When we were building Blur, my co-founder and I started looking into various Layer 2s because we wanted to play and learn on L2. As we dug deeper into L2 and talked to different L2 teams and their founders, we found that there was nothing radically new or counter to the existing market structure, which is why we felt the need to build Blast, we could build something more capital efficient.
We realized it was possible to create an L2 with native interest, where the asset wouldn't default to 0% interest like a traditional L2. Instead, you could give it the default interest rate that exists on Ethereum and stablecoins, and then have SDAI with about 7% interest like MakerDAO's DAI.
What do you think are the key criteria for L2?
We had two realizations that there were two dimensions and primitive concepts that could be improved that were not optimized on the existing L2. The most important thing as a developer is what primitive concepts these infrastructures can drive. The first one is native yield. Take Blur, which historically has about $100 million in TVL over its lifetime, but the TVL in these pools has not generated any yield. If these TVLs could generate yield, they would be a very good source of income, incentive or reward. This is pure market inefficiency.
We realized one day that this market inefficiency was also prevalent on all the other L2s. For example, there were billions of TVL on these different L2s, but these TVLs were not generating yield, so the market was completely inefficient in this regard. We realized that we could build a network where ETH itself would automatically generate yield. If ETH itself generated yield by default, this would actually change the behavior of the entire system because now both users and applications are earning yield by default.
The second original concept is more oriented towards incentives. As a developer, you have two choices: you can publish on an existing L2, or you can create your own L2. Creating your own L2 faces the challenge of attracting liquidity, but the benefit is that you can control the gas fee income, which means better value capture for developers.
What we want to achieve is to create a network where the gas fee revenue can actually be returned to the developer. So as a developer, if I build on Blast, not only can I get the native yield, I can also own the gas fee revenue that I generate as a DApp.
Views on the current crypto market cycle
Pacman: There's a common perception right now that this cycle hasn't really started yet because it's just been driven by ETFs. I think there's a fair amount of credibility to that. I don't think there's anything that's anywhere near the level of traction that we're seeing in 2021.
Interviewer: But how do you define it? Like Pump.fun is making millions of dollars every day. We see friend.tech bringing in about $50 million for developers. How do we define traction in these markets? Is it millions of users on-chain, or is it just revenue? I'm also curious why you think the 2021 cycle is bigger.
Pacman: If you search for Ethereum and Solana on Google Trends and compare them, you'll notice that this cycle peaked around March, but it still pales in comparison to the 2021 cycle. Google scored Ethereum searches at 100 in May 2021, and at the peak of this cycle, Ethereum searches were only 26. So that's a 4x difference just from Google search trends. That in itself shows that this cycle is much smaller. But if you're a participant in this space, a lot of times you can tell by feel, and frankly, you can feel that there are fewer people around, less mainstream attention, and overall, less energy, which is not to say that there are no income opportunities, but this cycle is definitely nowhere near the levels that were previously.
If you look at each cycle, usually the previous cycles were driven by some fundamental market change. For example, when Ethereum first launched, we had ICOs, which made it easier for people to launch tokens and raise funds more easily around the world. This greatly reduced the friction of issuing tokenized assets. In 2020, we had the launch of Uniswap, which really launched DeFi. Now we can finally create markets for tokens on the chain and trade these tokens on the chain. Then we had the DeFi summer, now you can lend and borrow, and all those yield farm opportunities, which are new things. Even in 2021, we have a new asset class, which is NFTs, which tokenize pictures and put them on the chain. These are all new things.
This cycle is basically more of the same thing, but faster and cheaper. You could classify it as something new, but in reality it's not a fundamental innovation, but an improvement in the dimensions of speed and price. So I think another reason why this cycle feels so different is that it's not driven by this fundamental market change. I think when we see this kind of fundamental market change, that's when we see a really significant cycle emerge.
