What's Wrong With Existing Blockchain Models? w/ Santiago Velez & Evan Cheng

0:00Your favorite neighborhood spot grows with Square. Indeed, my favorite neighborhood spot has quickly become Todd Snyder in Williamsburg. Todd Snyder is one of my favorite menswear shops and has supplied me with all the clothes I have needed this quite hot summer. Every business has different goals, but Square is the business platform that supports them all. From opening a new location, selling something new, or just expanding their reach. Indeed, I've seen it with Todd Snyder. In Square, also, you can get real-time insights, so don't wait for end-of-day reports. Go to square.com forward slash go forward slash realvision to learn more about how your business can grow with Square.

0:43That's S-Q-U-A-R-E dot com slash G-O slash R-E-A-L-B-I-S-I-O-N.

0:57hey everyone if you like this podcast go behind the paywall to get privileged access to the smartest minds in finance join the real vision community and learn how to become a better investor visit realvision.com slash rvpod and use the promo code podcast 10 that's podcast 10 to get 10 off our essential membership for the first year now to the top analysis of today's crypto markets. So we're going to do something different with this interview. This interview is with Evan Chang, who is the co-founder, came out of Facebook on the Libra project, and is the co-founder of SUI, which is one of these giant new layer ones.

1:37And I think it's a really interesting ecosystem. But the issue is here is I'm actually sit on the SUI Foundation here in the Cayman Islands, advising on where they allocate capital to develop their ecosystem. So I need to step aside and let Santiago Velez, who's much more knowledgeable in the complexities of new layer ones and protocols, to talk to Evan, and we can figure out ourselves what you think of it all as well. So again, I'm trying to be impartial and make sure that everybody's aware that I do have a vested interest in this. There's obviously no money change hands to do this or anything else.

2:14It's because they're a new large player. We've also had Aptos, who also came out of Facebook. They've been on the platform as well. It's all in the general manner that we believe that things should be done at Real Vision, which is everybody should have a voice. Okay, I hope you enjoy it. The world of crypto is an incredibly exciting journey that we're all going on together. We don't know where it's leading to, but we know it's going to be absolutely massive. Join me, Raoul Pal as I guide you on our adventure to discover just what this new world will look like. Welcome to Real Vision Adventures in Crypto.

2:49Obviously, I'm not Raoul, not as good looking, but I'll do my best to fill in today. You know, full disclosure, Raoul has some positions in the company we're going to discuss today in the ecosystem. So he invited me to speak objectively and explore this project for everyone watching. So I hope that we can all learn something together. This is the first time I'm doing a deep dive on this project. So thank you for joining us. I'd like to introduce today, Evan Cheng, who's the co-founder and CEO of Mistin. Welcome, Evan. It's my pleasure to be here. Excellent. So let's first start off with kind of your personal background.

3:28Tell us a little bit about you, how you got into this space, Web3, blockchain, crypto, and then we'll get into a little bit of a deep dive on the protocol itself. Yeah, so sure. I've been in the tech business for now, I'll say about 26 years. I did a bit of startups early in my career, then joined Apple for 10 years, where is, that's where I made my mark. I was the person who is in charge with a lot of very, very difficult, low-level, bare metal type of software. And I received a 2012 ACM Software System Award for my work, along with my colleague Chris Lanner for my work in LVN. Just to give you an idea that the award is also won by people who invented things like TCP, IP, Java, web browser.

4:20so my software has reached every billions literally many many billions of people it's in every cell phone every smart device android ios every watch smart watch out there so it's giving me a taste what you know the open source software impact can be and how powerful it is right give me the thirst to do something big again and after 10 years at apple i was recruited to facebook uh we're also building our organization but then in a couple years in, I sort of got the crypto bug. I think this is about 2017, maybe. I saw opportunity on both sides, right? One is, well, the opportunity for this technology to deliver massive impact through the world.

5:04And also my personal opportunity for delivering impact, because I saw some very, very critical mistakes in the design of, you know, Ethereum and others, right? And I think that's unsolvable. It doesn't even seem to be a recognition of how broken they are, specifically things around stability and more. And so I saw, hey, this would be a great opportunity for me to sort of do that again. Right? Opportunity to deliver impact to massive number of people. Again, I love doing things that's hard. So that gave me this idea. This is something I want to take on. Next. So So, but the tricky part is this is back in 2017, 18.

5:50And I look around, there's just not enough sort of the kind of talent I was looking for. I want to do some real serious R &D on this. I want the world-class experts in distributed systems, in program language form of verification, in cryptography, economics, and all that. and I wasn't finding those talent that's interesting in this space given how early it is right it's a bit also has a bit feel of a bit wild west to a lot of people so that wasn't possible so I stuck around until the Libra and DM project came about and then I joined Novi to basically run R &D you know there. So that sort of started my official journey into the blockchain crypto world.

6:44Fast forward a couple of years in 2021, it was very clear, Diong wasn't able to deliver on its promise because regulatory pressures and others. My co-founder and I left to start Mr. Lab. And since then we built Sui and launched just a month or so ago. That's fantastic. That's an incredible, very long view of a journey through all the different iterations of the internet. And of course, along the way, you noticed many of the problems for each phase. And one would argue that the problems that are occurring now in what we would call Web3, and we'll get into that definition as you see it, there are in some problems that you hope to solve.

7:32So first, can you maybe talk a little bit about some of the primitives that are used in each phase? You mentioned TCP IP in phase one, and we had social networks in phase two. What problems and then what primitives kind of arose to help solve those problems and move on this iteration, this journey? Perhaps you're talking about how the Internet has evolved, right? And I think it's very important to go back to the premises of the Internet. It's actually peer-to-peer, right? It's allowed any two individuals to be able to connect, right? It's about the ability for anybody to create a phishing marketplace, right?

8:08So, you know, TCPIP is very, very low level, right? Primitive, right? You have protocols on top of that for emails, for fire transfer and all that. It's all very peer-to-peer. And that's powerful because it allows anyone to create a phishing marketplace. So you are not constrained. You are equaling the playing field. Fast forward to Web 2. Well, it's not, right? You know, the web is extremely centralized, right? It's going back to, well, power law, right? If you have the network file, you have the distribution power, you get to control anything. You know, anything from content distribution, when you talk about Instagram, Twitter, YouTube, TikTok, you name it, right?

8:55Almost every single one of these platforms build on essentially the content is great by user. User-generated content, but most of the users never profit from them, directly at least. There's no transparency on how their data is being used for advertising. Then they actually can't take that connection out of the product, the platform. Think about it. Your social graph on Twitter, you can't really utilize it outside of Twitter. Same thing with all the other platforms, right? So it's becoming very centralized and there's a cost, right? People don't recognize the cost is being passed down to the consumers, right?

9:37And if you look at it more directly, say application distribution channel, they all take a cut, right? You know, Apple with App Store is dominant, right? They take a steady percent cut on just by everything in-apps purchase take 30 % or whatever percentage it is and also place a lot of restrictions on what you cannot do. So it's like every layer sort of squeeze the layer above, right? Using the distribution power to squeeze. Perhaps the most subtle is the harm it has on small and medium business. If you think about it, I bet there are studies out there perhaps the industry should do more about this, right?

10:22Small media business is a lifeline. It's kind of using social as a target customer acquisition channel. Well, so while great, the policy of app tracking has fundamentally, perhaps Apple and others will argue this is for the good, right? You need to evolve with the time to care more about privacy. But the flip side of that is, well, attribution is far worse than before. We saw this directly. Snapchat still hasn't recovered. Facebook has recovered some, but not fully. Target advertising far less effective, far more expensive. All these costs are passed down to consumers. It's also very restrictive on what you can do.

11:11so that's a fundamental issue here right when people talk about web street that's what's about right so can we go back to a world that's more peer-to-peer right and that's distribution that's monetization it's also about even for brands it's about reaching your audience directly have that direct relationship not through some middlemen that control access of the information so a lot things has gone wrong. And this is a technology that could shift the balance of power a bit more towards decentralization or more peer-to-peer. Hey, everyone, we're going to take a quick pause and hear a word from our partners.

11:49We'll be right back.

11:54So how does blockchain technology, and we'll get into your specific version of your vision, but how do you see that as solving the problem? Is it the incentive systems that are broken? Is it the access to a software stack or protocol? What is it that is really broken? How can we fix it with this tech? How does that work? Yeah, I mean, let me give an example to illustrate that, right? So people talk about using NFT as a way for content producer to reach the audience, right? You can sell directly. So you have basically this triangle, right? Content producer or product producer, the buyer, the consumer, and they have you a platform that help you sell.

12:39Well, right now, if the seller, any one of these platforms, Instagrams and whatnot, right? You're at their mercy how much, how they monetize your goods, right? And how much do they take? The idea is for blockchain to disintermediate, right? To eliminate that and replace with software, right? Because if software is done, it doesn't have feeling. It's not going to change its mind. It's like today I decided to change my mind. I'm going to get more greedy and then going to charge you more. And there's more transparency. There's more trust, right? So that's the whole concept of behind blockchain is disintermediation.

13:21And I can touch on that a bit more because it's nuanced, right? But I want to say this, right? And so far, it hasn't proven to be true. Looking at the NFT sales, what happens? Well, it used to be you pay royalties for a secondary sale because we all know that's a lifeline of artists. It's not the primary sale. It's a secondary sale. You want to take a cut. The marketplace, these are controlled by humans. They decided not to do that anymore, not to enforce royalty payment because it's good for them, to have higher volume of trading and look what happened, right? So the promise is there. The implementation is wrong.

14:03Therefore, your results where it's still the same, centralizing entity in the middle, sort of controls everything. They utilize their distribution power for their own benefits. Again, right, I'm not like, hey, they need to make a business of it. They need to make money. But it's about fairness. It's about transparency. It's about an opportunity to, you know, define what's right for you, right? For anybody producing content, other content being used and distributed. Yeah, it's interesting that the very network effects in what's called Web 2 that gave to the growth of social media, these power laws, they were very beneficial to help us connect with one another.

14:43Unfortunately, those same network effects created kind of a disproportionate bargaining power between the owner of the marketplace or the owner of the social network and the individual participants. So, you know, you get the issues of censorship or royalty splits or, you know, how it's monetized and who controls the data. So I agree with you that blockchain, this idea that you could disintermediate and still retain network effects. So you have public permissionless networks that could arise to facilitate the same kinds of connectivity, but without the compromises of having to give it to a single aggregator, all of your data and all of your monetization, et cetera.

15:24So you saw Web3 as exists now. What problem exists with Web3, right? Because there's a lot of people working on solutions for this. what problem did you see that you thought, hey, Mistum Labs can solve it better than everybody else? I mean, to be frank, right, I mean, a lot of the existing blockchain model just won't. You know, I'm not even talking about scalability performance here. I am talking about the data model as well as the programming model, right? In Ethereum, for example, EVN's most popular chain, right, tokenization of content or assets is skim deep, right? You basically represent ownership on chain where the actual contents lives elsewhere, right?

16:12So think about it for a second, right? The whole point is for smart contract to process assets for you, right? Not relying on human. Well, except you're not keeping anything on chain, right? So, you know, NFT is literally just URL, right? Point elsewhere. And it's off chain. The smart contract cannot do anything about it. Yes, there's ways around it, hack it. oracle and all that is highly inefficient right and and you can't easily change the asset right remember in the real world everything changes right today i have a baseball tomorrow somebody signed my baseball baseball player signed my baseball that's immediately worth a lot more that mutation the progression of throughout the life cycle for anything that cannot be represented easily right and cannot be represented on chain right you're losing all that promise as a result for this, to be frank, weird and completely wrong data model.

17:08And people are trying to hack around it, right, by all kinds of things that's just not to the point, right? And coming back to the other things, right, about the disintermediation, why is that so hard, right? Why is it so hard for software to enforce things like royalties, right? And this is the important part. when you really need to think about this concept, what makes an asset valuable? What makes an asset asset, right? It's all these attributes, all the information. If I have a character, it's a level one, it's a level 10, it has weapons, it's an elf, it's a dwarf, and all these things are made up an asset, in-game asset, right?

17:51And, you know, so these things are important, right? If you don't keep the information on chain, you know, they all look identical, right? And also the rules around it, right? we talk about secondary sale of our artwork, right? You as a producer should be able to define and say, I want to charge 5 % for every secondary transaction, enforce it to software, not to human. And why is that not possible and very, very difficult on Ethereum? You know, I know, again, there's an effort to hack around it with a lot of data participating and all that, right? But those are hacked, right? So your programming model is wrong.

18:29Your interface, say this is ERC, some, some, some thing, right? Everybody implement according to the interface, right? But interface is abstract. Implementation is done by individual teams that implement according to the interface. You don't know what they do. You can't enforce it. You don't know if they're good or bad or they have a poison implementation. They claim to serve royalty, but they may not, right? So you go fall back into social consensus. you expect all the humans to play by the book. That's just fundamentally wrong, right? Rules around asset transfer and creation, right, are parts of what an essence of an asset, right?

19:11So think about in things applying to securities, you know, require KYC and mail checks before it can be transferred. Well, how are you going to do that with this model, right? Are you assuming everybody's going to play by the book? you can't right has to be encoded embedded secondary sale royalty payment copyright verification these things are all essential so as has rules around them so this is the fundamental problem with the blockchain data model the programming model everything associated with it even ignoring all the challenge with you know how inefficient they are and how expensive they are right because if you can't do that you cannot fulfill the promise of the blockchain that is this intermediation replace human in the middle with software because if you cannot trust software these software they're managing potentially billions of dollars you need to be able to say ahead of time prove them are correct they have to be correct they have confidence otherwise you fall back to human, right?

20:23This is why a lot of tokenization effort looks like, well, we do primary insurance on chain, but then, oh, by the way, we actually have to do all the secondary processing at, you know, consortium chains or whatever, because that's the only way we can control and trust it, right? This is why banks use blockchain, but, you know, private setting, right? You have a fundamental mismatch about your intention and your technology. And And this is why I think everybody will agree there's very little concrete disruption of products that we all use, you know, from business or consumer side by blockchains today.

21:03Very, very few examples. Maybe a bit of payment with stablecoin. That's quite a great example because those things are simple, right? Stablecoin is stablecoin. You only need to keep balance. And it's a better instrument that's done. But anything more complex, right, where you actually need to enforce rules around it, you actually need to know what the asset look like, you need to compose assets, right, be able to track its lifecycle, the changes of asset over time. It's really not that effective, right? You know, like people being hacked around and trying to make it work. But why? Fix the fundamental infrastructure.

21:38Hey, everyone, we're going to take another quick break and hear a word from our partners. We'll be right back to the Real Vision Crypto Daily Briefing.

21:48Yeah, and it seems, you know, at the root, there is a problem, you know, the Ethereum virtual machine, which is kind of the machine that runs the environment on the Ethereum blockchain and Solidity, the programming language, you know, that's used in many L1s, not just Ethereum. But one of the fundamental problems as I see it is it's not object oriented, which, you know not to go too far into computer science but it has some fundamental limitations that it can only be used essentially as a very sophisticated and immutable registry right it's just a indexing database that's shared publicly and what you're saying is that you can have much more rich programming environments that not only can be shared publicly but can be amended and can live in real time in the same way that assets live in real time in changing ownership, changing content, changing everything that an object might be able to change if you're a developer, you know, to generically incorporate all of these other elements that right now you can't.

22:52So it seems like for a period of time, there was a mismatch between the rate of conceptual development, all these great ideas about how to fix and disintermediate third parties. but the layer at the base really isn't sophisticated enough to realize that promise. And so you have to come in and kind of backfit with these hacks. So can you tell us a little bit about how your lab, how it fixes that problem, what is unique about it, and how you think this will solve the problem? So if you look at Solidity, what are some of the things that are wrong with it? Dynamic behavior. We're re-entrancy bugs, right?

23:32You have delegation. You have all these things. Let's basically make it not trustworthy enough. You keep on finding bugs in them because, well, you just cannot analyze ahead of time to say this is provably correct. And that's a fundamental mismatch in how you design a language for a smart contract. You need to be able to trust it because these things are going to manage a lot of money for you. So that's a very, very fundamental problem. So even back in the days where Facebook, we, you know, recognize the need to have a much, much better, different smart contract language. No dynamic behavior. Actually have asset as a first class concept, right?

24:15Allow you to have a language to describe all that, you know, what an asset look like, not just a URL or balance, right? Something very, very basic. ownership models are captured and modeled correctly and use the runtime as much as possible to force these things right because another problem is stability is like if you forget some kind of ownership checks somewhere you'll probably be in trouble right so lots and lots of bugs we were seeing about that right hundreds of these kind of issues right get all these things right and don't place a burden on user on developers right because not everybody's a security engineer right and and finally build the tools, right?

24:55You know, have specification, has designer language along with the specification language allow you to formally prove, formally verify this offer is correct. It doesn't get better than mathematically prove something is correct, right? So all these layers of defense, right? It makes it much more expressive, much more powerful and correct and safe and all that sort of thing to do. Now, fast forward to suite, right? Once we recognize, right, S3 needs to model objects, assets as an object, right? So they're free flowing, right? And is SUI a blockchain or a programming language or both, a generic? SUI is a blockchain, right?

25:38So, you know, there's a larger aspect and there's a programming environment on top of it. I'll get to that in a bit, right? And that's a very, very sort of high level concept that it's not quite correct when it comes to S3. but you know it is a blockchain right but street is a very different object model i mean data model where everything is an object right it's a data store where everything is independent objects uh and then you have you have address that own you know where your your own object live rather than having the data trapped inside smart contracts right so but that model allow you to do something even more powerful right because now object is composable right you have language around that objects can be passed and stored and can be combined.

26:23A lot of things you can do with it. The ownership model is much more powerful. You can be owned by a single person or unowned, like something like a smart contract everybody can access. So we iterate on that to match the data model to really come up with a completely different sort of concept. And the blockchains also sort of think about things very differently. So for the blockchains, you think about tracking the history of assets being transferred. I send you 5 years and you send John 10 years. You track the movement of assets. But the assets generally don't change. You don't track the asset change themselves.

27:06You can think about three as sort of another layer on top. You also track the state change of all these assets. because again, you're describing assets as a rich type. The type, character with level, with race, with experience and strength, intelligence, these things change. So you also track the state changes on chain as well. So it's almost like it adds another dimension because objects are versions as a result. So all that gave us a much more powerful framework to solve the kind of problem we talked about. So this to me, this much more rich object oriented asset types, this could include all types of content like we talked about earlier, media content, NFTs, tokens themselves, but also within each object, the history of that object, right?