The a16z Show - a16z Podcast: Mental Models for Understanding Crypto Tokens

Starting point is 00:00:00 The content here is for informational purposes only, should not be taken as legal business, tax, or investment advice, or be used to evaluate any investment or security and is not directed at any investors or potential investors in any A16Z fund. For more details, please see A16Z.com slash disclosures. Hi, everyone and welcome to the A16Z podcast. I'm Hannah, and today's episode continues our series on All Things Crypto, with Chris Dixon in conversation with Nick Tomeino, founder of early stage crypto fund one confirmation and formerly BD at Coinbase. In this conversation, they start off with a framework for understanding tokens to what gives them long-term value,

Starting point is 00:00:40 what's the role of stable coins in the ecosystem, as well as scaling issues, on-chain versus off-chain protocols, forks, and other trends in crypto. You can also check out more details about this episode, as well as past episodes in this series and a video covering the building blocks of crypto on our site at A16Z.com slash crypto. So people may have heard of tokens on coins. Token sometimes means like specifically ERC20, which means tokens built on Ethereum. Some people use the term more broadly. There's a little bit of confusion around the nomenclature.

Starting point is 00:01:10 But, you know, broadly tokens, coins. There are different categories of tokens, I guess. And maybe you can, so you have a kind of a framework here. Yeah. You can help walk us through that. There's a lot of different mental models people have. And this is just mine. But how I think about it is simply kind of two types.

Starting point is 00:01:24 So the first is what I call a usage token. And some people call it a medium of exchange token. Some people call that a utility token, but it's basically a token where ownership is required to use some digital service. And you can think of any blockchain as a digital service where you have kind of buyers and sellers. Bitcoin's the most widely known usage token where ownership is required to use this digital service. And in the case of Bitcoin, that digital service is a ledger that can be used by anyone to store and transfer value. The original usage token, by the way, were IP addresses and domain names, right? I mean, like the sort of scarce network resources of the Internet, which, you know, you needed an IP address to access the Internet and you needed domain name to set up a website.

Starting point is 00:02:06 Yep, that's right. And so if you look right now at kind of coin marketcom, the top 25 tokens by market value, I think 24 of the 25 are usage tokens. One of the interesting things is right now a lot of the ICOs that we're seeing are kind of masquerading as usage tokens. right. But how I think about whether a usage token can have long-term value is really, is the digital service that's being provided will demand for that increase in the future and are there kind of scarce, unique resources underlying that digital service? Yeah. Will there eventually be what I would call consumptive demand, not speculative demand, right? Meaning demand to actually use the service as opposed to simply to buy it and hoping it'll go up in price. Right. In the same way that, you know,

Starting point is 00:02:50 at the beginning of the internet, domain names, people bought them for speculative use cases, Eventually, you know, someone just bought pizza.com for $10 million, they bought it to use it, namely to sell pizzas, right? Yep. And so, or whatever their model is, affiliate marketing for pizzas or something. But eventually, the reason domain names eventually, or IP addresses, you need them to, you know, if you want to have servers on the internet or something like this. So eventually the consumptive demand replaced the speculative demand. And I think what you were saying about ICOs is a lot of sort of masquerading is that there are obviously people taking advantage of the, you know, just the money flowing in and sort of taking what is probably just a normal company that's not a protocol. and sort of calling it a protocol and making up a kind of use case as to how it's a usage token,

Starting point is 00:03:30 but it just is an excuse to raise money. Exactly. All right, so that's usage tokens. And then what's the other kind? So the other is what I call a work token. So it's basically a token where ownership gives the token holders the right to contribute some type of work to a decentralized organization. And so an example of this is something like the reputation token, which is native to augur, where you don't have to own rep to use the prediction market.

Starting point is 00:03:53 That's the design. This isn't a live product yet. But ownership gives you the right to do work in this decentralized organization. Why would you want to pay to do work? Well, you'd want to own a token, a stake that gives you the right to do work in the future. So basically, if you think that millions of dollars will be flowing through the auger network in a couple years, and you want a piece of that, right? So it'd be kind of like if Facebook were a crypto protocol, a work token, might be. might be the ability to vote on how the newsfeed algorithm works.

Starting point is 00:04:26 Yep, yeah. And you can see how, you know, at some point, if Facebook is helping to determine who win an election, like having some say in that algorithm could be valuable. If these crypto networks become very important, having some voice in it would seem to be valuable. Exactly, exactly. And I think it could manifest itself in many ways, right? So I think the most interesting work token will be when Ethereum switches from proof of work to proof of stake. And by the way, these aren't mutually exclusive. So you can have tokens that are both usage and work tokens.

Starting point is 00:04:57 And I think many tokens in the future will likely be that. And Ether will be when it switches a mistake. Exactly. You'll need Ether to pay gas to run smart contracts. And then you'll also, ownership will be required to do work in the network. And in that case, that work is validating transactions. And so when Ether switches, you know, people will want to own Ether. They may want to own Ether to earn Ether to earn the... the mining fees, if you will.

Starting point is 00:05:24 Yeah, I mean, that's the other reason you want to own work tokens, not just to vote on it, but in the case of like the proof of stake in Ethereum, you get paid for it. Exactly. Zerox is another kind of example of, you know, what they call a work token, where the work is just governance. So in some cases, you may just value, you think this is going to be important and you want to play an important role in it. But in other cases, like in the case of Auger and Ether, it'll actually give you

Starting point is 00:05:44 the right to earn fees in the network. And so I think, to me, I'm really excited about like completely new organizational structures, which Bitcoin is kind of the first. But, you know, on the work side, you can only participate if you, you know, have sophisticated hardware and a lot of computing power, right? So I think when we see proof of stake, we're going to start to see kind of more inclusive ways to create decentralized organizations that lots of people can participate in. There's another kind of tokens we'll talk about, which I would call like traditional asset security tokenization. You know, I guess my view on that is, And that is taking something that's, you know, a piece of real estate or something that has traditionally had value and creating a digital token that represents, you know, lets you trade that or something, right?

Starting point is 00:06:30 My own view is that's far less interesting and that's not even part of the crypto movement in a way. To me, it's not really a cryptocurrency concept. And it just may just be a kind of regulatory workaround or something. Yeah. I agree with that. I would say, I think, you know, it could be really meaningful. in the sense that we could see a lot of traditional assets tokenize, but from a technology perspective, I don't think it's that exciting.

Starting point is 00:06:58 But I do know that, you know, everyone around the world is seeing all this money that's poured into ICOs this year and thinking maybe we should tap into this kind of global liquidity pool and tokenize our real estate project or our security or things like that. And I do think it's likely that we do see particularly jurisdictions that want to be favorable to it, put regulation in place and attract kind of investors and entrepreneurs that are thinking about this. Do you think that the kind of tokenization has gone too far? I mean, the flip side would be, hey, maybe every project should have, you know, should have a token. And it's just kind of like a new business model and a new way to kind of, you know, align incentives around a new network or something. Yeah. I think it's gone too far. My sense was that the ICOs had slowed down quite a bit. but there were still 400 million in token sales in November alone. And so I think in this boom, there's certainly people that shouldn't be doing tokens that are.

Starting point is 00:07:59 So how that will play out over time, I'm not sure. But I would caution, you know, anyone who's looking at investing in tokens to think a lot about whether the token can have long-term value or not. or is this just an interesting project, but there's not a strong reason for the token to kind of persist over time and have significant value? Everything we're discussing is highly speculative.

Starting point is 00:08:24 One thing we should say here is that we do not recommend that people listening buy any of these tokens. And if you do buy them and, like, you do a lot of research and put a very, very small amount of money into them because it's super risky. And I think a lot of people, including me, think it's very likely we're in a boom part of a cycle. And while I think we're very, very excited

Starting point is 00:08:46 about the long-term technological prospects of what's going on in cryptocurrency, you know, I think it's very likely that the prices have gotten ahead of the actual progress. Certainly this year has been all about the token boom, right? There's been over $4 billion in token sales. What I'm really spending a lot of time thinking about is, like, what are the actual products that people are going to use over the next year or so, because I think a lot of these big projects that have raised a ton of money aren't likely going to ship anything for many years in some cases. But I think there are kind of a few different categories of products

Starting point is 00:09:20 that we're going to see actually start to drive real usage kind of in the next year. And we're already seeing that to some extent. So one example is like decentralized exchange. Just for the listeners, what is a decentralized exchange? So very simply, in a centralized world, which exists now, Now in digital assets, you're trusting a third party like Coinbase that allows you to trade, right? So you're storing your value on kind of a centralized service and then trading and then withdrawing. Decentralized exchange basically facilitates trustless exchange of assets.

Starting point is 00:09:57 So if you control your private keys, you can trade basically. And it's a more secure way and you don't have to give a third party. your funds, which is susceptible to hacking and things like that. So what are some examples there? So one kind of project that I think is quite interesting that's seeing traction is 0X. So 0X did a token sale earlier this year, and they're already seeing some projects kind of built on top

Starting point is 00:10:24 of what they call relays. Radar relay is a good example of that. And the architecture there is you use Ethereum as the settlement layer, but then you have sort of off-chain relays or order books that do the high throughput transactions, which is an emerging architecture we're seeing more and more of, which is sort of use Ethereum as less performance, but you can trust it and you can send money through smart contracts

Starting point is 00:10:48 and things like this. And so use it for what it's good for and then do everything else off chain. And zero X is sort of one of the more interesting examples of that architecture. And so an alternative to kind of the zero X approach is something like Ether Delta, which does everything on chain. So they do order book on chain.

Starting point is 00:11:04 And every time you change an order, for example, if you're using metamast, you have to pay gas and click a button. And it's kind of a poor user experience, right? Well, and expensive potentially. Design protocols that use both on and off chain is very tricky, right? Because you have to make sure that the off-chain stuff is also happening in a trustworthy way. Like on-chain, you get trustworthiness for free just based on the nature of the blockchain architecture.

Starting point is 00:11:27 But off-chain, you have to guarantee that as well, right? So you don't have like a rogue order book or something like this, right? And so that becomes a very tricky design problem, right? Yep. To be clear, it's still early. And I think there's still even in ZeroX a lot to be figured out. By the way, I am a personal investor in many of these projects being discussed. But I think what's interesting about ZeroX is they're solving a real problem

Starting point is 00:11:46 and they're seeing a real ecosystem kind of emerge on top. That's kind of characteristic of the projects that we'll be seeing work to get usage in the near term. They tend to be more crypto-focused because that's where the users are today of these crypto products, right? So as opposed to, you know, I think you and I both believe that someday people will have social networks and ride sharing and all sorts of things. using crypto protocols, but that might be farther away, just given the way this is evolving. That's right. Yeah. I think one mistake I see you see a great entrepreneur on paper that's solving some real-world problem with blockchain technology, and it sounds really good, but can they really execute? Is the technology stack really there at a point to solve an end-user

Starting point is 00:12:28 application? I'm most excited about projects that are solving problems for the existing crypto community rather than the broader landscape. What's missing, why can't you solve problems with the broader landscape today? Is it just the Ethereum scaling issues? Is it other services like, for example, identity services? Is it just that we're missing kind of core components? There's a lot of core components that are still missing. Take something like Auger, which is a decentralized prediction market, which is kind of an end-user

Starting point is 00:12:55 application. In order for something like Auger to become interesting to a mainstream, even if kind of mainstream users wanted to use this type of product, I think there's a lot of things missing. So one is a decentralized stable coin, right? So what is a stable coin? So a stable coin is a coin that has the volatility characteristics, similar to that of a dollar combined with the decentralization characteristics, similar to that of Bitcoin or Ethereum.

Starting point is 00:13:21 So taking kind of the best of the stability characteristics of fiat currencies and the decentralized characteristics of cryptocurrencies. The volatility of Bitcoin and Ether are such that they're moving. giving 5 or 10% in a given day, and no one that's actually using it as a medium of exchange for something or placing bets would actually want to use that, right? So a decentralized stable coin, in my view, is like a core piece of middleware that's missing in the ecosystem. I think another is identity. A lot of these services, you know, really rely at the application layer, rely on identity, and there hasn't been an identity-based protocol that's really

Starting point is 00:13:55 emerged and gained a lot of traction yet. So that's like if you build a decentralized ride-sharing service, how are you going to keep track of the rider and driver ratings as an example, right? And how you're going to avoid civil attacks and all these other kinds of things. I mean, one of the things I'm excited about in terms of kind of new projects is like continuous token mechanisms. So how do you drive usage of the product and give new users for some action new tokens? And something like identity, I think, is a piece that's really needed in most cases to do that in a way that can't be gamed by civil attack. So let's like a little bit about stable coins.

Starting point is 00:14:30 So we're investors in stable coin, base coin. First of all, I think there's two categories of stable coins. There's things where they try to peg the digital asset to a physical asset, right? And some of these are controversial. Yep, tether being the most well-known. And then there are what you might call kind of like crypto-native, the ones like make or die and base coin, where there's no tying into something. Instead, there's this kind of game theoretic mechanism that's been developed.

Starting point is 00:14:57 So maybe we can talk a little bit about how these, the maker-dye. die and base coin work. The actual mechanisms that they're using to create a stable coin are quite complex. At a high level, this idea of creating what some people call an algorithmic central bank, right? A central bank where you kind of peg some asset and based on market supply and demand, the supply of the token adjusts. And so in the case of Bitcoin, for example, Bitcoin has a very clear cut.

Starting point is 00:15:30 money supply, which is that 21 million Bitcoin will ever be created in the year 2140, and the rate at which new Bitcoin or mine decreases by half every four years. In the case of something like base coin, they're pegging, you know, the coin to some asset. You could say one base coin equals $1. And then based on the free-floating price, the supply of the coin adjusts. Right. So if the price goes up, more coins get printed. If the supply goes down, then a new type of token in their ecosystem called a bond gets printed and folks buy the bond with base coin. So that's kind of... Essentially, as you said, you can think of it as kind of a second reserve of currency that acts like a banker or something and buys and sells or, you know, it tries to adjust the price

Starting point is 00:16:19 to a target. And like with a lot of these economic things, it works to the extent people will believe it works. Because for the banking mechanism to work, there has to be real value there. Whether there's value there, it partly depends whether people believe there's value there. And so it has this

Starting point is 00:16:31 kind of reflexivity property until these things are running at scale for years. I guess we won't really know if they work. But it seems like a really important problem and one that's worth working on. One kind of fair question.

Starting point is 00:16:42 Some would argue, well, Bitcoin volatility has decreased every year and at some point it will get to a point where it's actually a good medium of exchange, which we'll see. Other economists would argue,

Starting point is 00:16:55 well, it's a fix-up supply. And so by definition, there will be more volatility fundamentally than a coin where stability is the actual goal. So we'll see. Yeah. Okay. So let's talk about forks. So for a big topic this year. Forking is a very powerful mechanism. And it's really kind of a new concept where if you don't like how a project's being run, you can copy the project and not just the code, but all the data from the code. There was a Bitcoin fork with Bitcoin Cash. There was a threatened fork with Segwit 2X that was called off. There was a famously Ethereum forked, creating Ethereum Classic and Ethereum.

Starting point is 00:17:30 The equivalent would be like, I don't like how Facebook is running their algorithm, creating Facebook 2. And it not only has all the code of Facebook, but it also has all the photos and everything else. Like, it's just the whole thing. It's just Facebook 2. One of the important things about it is it keeps the people governing the network in check, right? If they start to extract rent or do other things that are bad, the community can just fork them. In fact, that happened with, like, Monero came out of a fork, right?

Starting point is 00:17:53 So Monero is a very popular privacy coin that the community felt like the creators had taken too much of the initial token distribution to fork them. And now Monaro is far more popular and successful. But it also creates this really weird situation where you have these kind of alternative timeline history things and you have these weird coins. And like there's, for example, Bitcoin Gold, right? What's the story of Bitcoin Gold? I actually don't know the exact story other than it popped up a couple of weeks ago and it now has a total market value of, over $6 billion. And Bitcoin Cash had like a real community momentum behind it.

Starting point is 00:18:27 Yep. It was a set of sort of Bitcoin core community members who were unhappy with the governance and decided to create a new fork. Bitcoin Cash has like a $26 billion market cap now, which is more than the entire crypto space a year ago. Yep. But that at least has sort of this community and kind of use cases and Reddit forums and a whole bunch of other things, right?

Starting point is 00:18:49 Whereas you have these other things, which are just like mystery forks or something. You know, we hear so much about the ICO boom, but in terms of kind of like total value created this year, there's actually been way more value created by forks than ICOs. Yeah, right. Like the total ICO raise, what is, it's a couple billion. It's like four billion, four billion. Whereas Bitcoin gold, which is a fork that none of us even know where it came from or whether it's real is worth $7 billion. Yeah. It's a lot of crazy stuff, yeah.

Starting point is 00:19:20 Yeah. Yeah. My view, though, is that the developer team and community behind the project is the most important aspect of these projects. So when you have a fork, you basically have two different communities that have some disagreement about how the future should look, right? And in the case of Bitcoin versus Bitcoin Cash or B-Cash, you know, it's about the block size. Yeah, the block size, which is symbolic of a broader debate around being kind of digital gold or digital visa. Yeah, exactly. So, but an idea and a belief is one thing, but having like a team and a community that can execute on that belief is another thing. So what are some of the other things you're excited about in the crypto world?

Starting point is 00:19:59 So I think, you know, this idea of solving skillability problems, right, is very important and I think is going to manifest itself in many different kind of directions. What I call kind of middleware piece that's really needed for something really being used by millions of people is, you know, the Ethereum blockchain or, whatever blockchain you're using has to be able to process hundreds or thousands of transactions per second, right? And right now, Ethereum can process roughly 15, Bitcoin, roughly seven. And so there's a lot that needs to be done there. And so there's some interesting kind of approaches. Well, and Bitcoin's only going to get worse now that the Bitcoin, the scaling debate seems to be over. So the price of transactions will go up and the transactions per second, I think, will go down. But the hope of Bitcoin is that the so-called layer two solutions work.

Starting point is 00:20:49 the Ethereum development team on the other hand has a different approach, right, which is they're actually willing to kind of make pretty dramatic changes to the core, including sharding, right, which adds sort of parallelism. So instead of having 30,000 miners all running the same code,

Starting point is 00:21:02 they can run different code, right? That's right. It feels like there's a lot of interesting directions through which people are trying to improve the scaling, especially on the Ethereum side. For sure. And then the other thing that I would add

Starting point is 00:21:12 is kind of what I call horizontal kind of solution. So something like Cosmos, which is trying to create some framework, for new coins to be created and even port over kind of existing ledgers. They're working on a cool project right now called Etherment, which is basically taking the Etherland, porting that on top of the cosmos, what they call tendermint, consensus mechanism and their own pure-to-peer layer. And so if a developer is building something on Ethereum, but they want a thousand transactions per second or something like that, they could just port it over.

Starting point is 00:21:48 port it over and have the code work and also have the transaction history port as well. Both. But the transaction history from that point forward is different, right? I think of Cosmosis, I guess people are calling a pair of chains like Pocodot. And there's just sort of a set of proposals for doing this, which is sort of for interchained interoperability or something like this, right?

Starting point is 00:22:07 And you think of that as like another, I guess, vector through which we'll get scaling. It can be achieved. Yeah. It kind of goes back to the original side chain idea, right? Yep. back, you know, which was back from crypto v1 or crypto wave one, which was this idea that you'd have these side chains off Bitcoin, which you could basically burn coins on Bitcoin, take them over

Starting point is 00:22:26 to the side chain, do other things, try out new software and do other experiments and then bring them back, right? Yep, yep. So this is kind of like V2 version of side chains or something. Exactly. And then there's, you know, new chains that are just fundamentally more scalable, right? It's like DFINITY as an example, yeah. And the exciting thing as kind of...

Starting point is 00:22:46 an investor and someone interested in the space is that there's a lot of different projects kind of all happening at once, right? Where if you look back on Bitcoin, you know, a couple of years ago, it was really just side chains and then lightning came along. But it feels like there's a lot of stuff kind of all happening at once on top of Ethereum. I feel like the good news is the token model provided an economic model through which new projects could get funded and have sort of maybe a long-term business model as well to have the token appreciate and have the sponsoring entity hold some portion of those coins. The flip side of it is it almost worked too well, which is now the incentive being to go join the, you know, work on Ethereum, which in the old days would be like working

Starting point is 00:23:24 on Linux. The incentive is to create, you know, your own operating system, your own blockchain. Sometimes I worry that like this new incentive of having a token almost works too well and is creating kind of splintered development efforts. It's certainly possible. And yeah, some people are kind of hoping to see more like coin distribution for contributions and things like that. Which, like Frulson has this idea, right, this idea that you would have kind of through token inflation, you would have essentially like, you know,

Starting point is 00:23:53 bounties paid out for people that contribute helpful new code to a project. But I think right now, what is kind of interesting to me is that we are seeing all these different ideas being attempted, right? In my mind, in kind of crypto, the idea maze is kind of just accelerated, like 100x, right? We're seeing all of these different ideas happen at, once, I think the best kind of entrepreneurs are paying attention to what's going on in all of

Starting point is 00:24:21 them and seeing what's working, what's not, and taking kind of different pieces. And that was like Ethereum being the best example of that to date, right? Where Vitalik studied the ecosystem very carefully, knew exactly how Bitcoin worked, what worked, what didn't, knew how like the initial proof of stake systems like Purecoin worked, knew what was wrong with them, and kind of crafted together a lot of different pieces from like historical context and then built something. I've been involved as an investor in the space. We invested in Coinbase in 2013 and I've spent all a lot of 2014 and 15, 16 and 17 meeting with crypto companies. I probably see five what I would call like highly qualified teams a week now, whereas I might have seen five highly

Starting point is 00:25:02 qualified teams like in the year of 2014. It's just dramatic, right? To me it feels like the way the iPhone felt in 2009 through 11, where just like every smart person who knew how to program and design and do other things said, I'm going to build an iPhone app. Right? And of course, 99% of those didn't work,

Starting point is 00:25:20 but the ones that did work spectacularly well, right? The other interesting thing is now it's transitioned, it's not just these people that have sort of grown up in crypto. At least what I'm seeing is lots of teams leaving GAFA companies, Google, Facebook, etc., and getting into this and really good teams. people with real, you know, strong engineering and product backgrounds. And I think for me, at least it's the vast majority of it's centered around the Ethereum ecosystem.

Starting point is 00:25:47 Who knows which of these projects will work, but there's a lot of stuff going on. Yep, yep, agreed. What would be your predictions for the craziness to come over the next year or two? I think that we're going to see proof of stake really, you know, a lot of people are still skeptical on proof of stake and don't think it could be pulled off. But I think proof of stake is going to be a really important. kind of catalyst for the broader adoption and interest in blockchains. What specifically makes you think that proof of sake will spur broader interest? I think it solves a number of problems.

Starting point is 00:26:21 The one problem that a lot of people point to is kind of the electricity cost, right? So, you know, hundreds of millions of dollars are being spent external to the network of Bitcoin just to secure the network. This is Bitcoin mining, which, you know, you see all these articles about it. it's now passed. I think it's like the country of Ecuador or something in electricity usage, which is excessive. That's just too much. But more broadly than that, I think minors controlling a network has proven to be pretty problematic. And so I think kind of aligning stakeholders and miners and putting them as one will be important in terms of progress in making changes to protocols. just kind of making blockchains more cohesive, if you will.

Starting point is 00:27:10 It also, from a technical point of view, I think, allows for a lot more wider space of design of blockchain consensus mechanism designs that allow for scaling and other kinds of things. Things like DFINITY, you can't do as far as I understand it without proof of stake. That's right. And that has like all these nice features, like a couple second transaction finality time, which you, you know, which is a much shorter time than on existing popular blockchains. So your job is to invest in new protocols. What are some of the characteristics you look for for differentiating between, you know, attractive investment opportunities and unattractive ones? So I have four characteristics that I evaluate every project based on. The first is team.

Starting point is 00:27:48 The second is product. The third is community. And the fourth is token mechanics. So for me, someone with kind of deep historical context on crypto is important. Someone that has, you know, deep technical understanding of consensus mechanisms and crypto economics and things like that is very important. So let me push back on that a little bit. One thing I'm seeing is because the kind of movement is growing, by definition, you have

Starting point is 00:28:16 new people entering who don't have the same history in it, but do have, in some cases, very strong technical backgrounds and relevant fields like distributed systems and cryptography. One thing I debate is like, is that, you know, if someone is very strong in those areas but maybe hasn't, like you and I have, like, kind of lived through all the different, you know, events of the last four years, I mean, if you look at the Internet as a, analogy. A lot of the early teams that were working on it were kind of like hardcore internet believers, but then eventually

Starting point is 00:28:42 this sort of different, this new vintage kind of came in that was sort of more classic computer science and built a lot of the important things. No, it's a good point. I struggle with this. I think I'm just personally more biased towards kind of the deep crypto person as opposed

Starting point is 00:28:58 to the new kind of entrant. I think there will be a time in the ecosystem when the new enter and build some really important stuff. I think my view is that we're still in the infrastructure middleware build out. And so I think it's more likely that the deep kind of crypto person has that. What about and then so second was product. So how I think about product is like I'm a native crypto person myself. Like I was one of the early users of Coinbase. That's how I got involved. And I spend a lot of time in Redits and slacks and different forums, just, you know,

Starting point is 00:29:32 paying attention to new products being developed. So one takeaway from Coinbase, I would say, is that you need a great team, but you also need fantastic product timing, right? There's a lot of great teams, and you really need both. And so I spend time thinking about, like, what are the right products right now? Specifically, my experience with investing is the risk is often being too early, right, as opposed to too late. Like, you know, when you're deeply into a field, like, it's sometimes easy to go and kind of try to jump ahead

Starting point is 00:30:00 of where the reality is right now. And then third is community. In almost all cases, having a strong community is important to being successful. And so I take Ethereum as the best example where, like, Ethereum has, you know, great technology. But even more so, I would say Vitalik has done an incredible job kind of building a really healthy, vibrant developer community. And kind of leading in a way that's inclusive and open and things like that. I'd also say it's very technology focus. So if you just look at our Ethereum versus some of the other coins,

Starting point is 00:30:33 it's focused on the technology and the applications, not on just the price. And they made a concerted effort early to do that. In fact, they push price discussions over to ETH trader. Which is very interesting. And the community could be what. It could be GitHub, Twitter, Slack, groups, Reddit. They tend to be on sort of multiple networks.

Starting point is 00:30:55 I often look even at a founder's, Reddit activity. So is this a founder that's kind of been deep in the community and is very engaged and things like that? And then the fourth is token mechanics? Yeah. Does the token make sense? And do you have conviction that, you know, it's a token that could have long-term value? So that's one piece of it. And then the second is, I mean, I think maybe this is a little bit of a different perspective than some. But I actually think kind of the way that the token is priced and launched initially is pretty important to the long-term success to the project. So in token mechanics, it's one, do you think this token could have long-term value? And two, is it being launched in a way that is perceived fair and that can

Starting point is 00:31:38 attract a lot of people from all over the world over time? And so the mistakes people make are sort of unfair, you know, investors granted deep discounts. I've been hearing a lot more talk about things like air drops or other kinds of token sale mechanics that ensure broader distribution. Yeah. I feel like there's more and more things. thinking around that. Yep. And I hope that continues, right?

Starting point is 00:32:01 Like for AirDrops, I think Omisei Go is a good example of, you know, they gave everyone that owns Ether some ratio of their own token. I'm hopeful that more teams will be thinking about when they create a token, not optimizing for like the short-term cash grab, but more optimizing for product usage. And so thinking about how to create a token distribution where users do an action and get tokens. There's been so much attention on ICOs that, by the way, You don't need an ICO. Bitcoin never had an ICO. Ethereum had a relatively speaking small, $1.18 million.

Starting point is 00:32:34 Yep. And I don't think that was a key part of the network success. It's really kind of a secondary part of this. It's really about building networks and communities. Exactly. Yep. And I actually think if you focus too much on the ICO or the, you know, the token sale, it can actually really hurt your community.

Starting point is 00:32:50 And I think we've already seen that in a few cases. One thing I'm surprised on your list is governance. Do you think about that? My view is that in the early days of like a project, governance is less important. So I think governance once a project reaches a certain scale is very important. But I don't view like traditional startup governance is that broken. Something like Ethereum, which from a decision-making perspective is pretty centralized right now, that's actually been a good thing for Ethereum.

Starting point is 00:33:18 And it's allowed them to move fast and break things, right? Once it gets to a certain level, I think, you know, they'll recognize. is that shifting governance is important. But I think it's really hard to have, like, a really decentralized or empowering governance structure early on and really build something. There's some basic things like vesting as an example. Like a bunch of projects had no vesting for the key team members. And I think that then those people took their tokens and left, which in startup world, like vesting is just a fundamental concept. Okay.

Starting point is 00:33:48 So you're talking less about governance from a decision-making perspective from the protocol, but more how the team that's building is governed. Yes. Well, I think all of the above. It can be on-chain or off-chain governance, right? So you can imagine both, and people have experimented with both, but there can be governance in the sense of, like, how do you control the protocol, how do you upgrade the protocol,

Starting point is 00:34:07 do that on-chain or off-chain, right? And to the sense, it's off-chain, like, that means the human layer. And those organizations, how are they formed? Are they for-profits? Are they for-profits? Do they have vesting? There's the whole thing, right? Yep. Like, is it run well?

Starting point is 00:34:22 Yeah. Certainly important. And on the governance of the teams, I think there's certainly been a kind of learning curve this year. And at the beginning of the year, I think there was a lot of abuse of ICOs and, you know, founding teams getting liquidity right upon funding and things like that. And I think from that perspective, there's been kind of big changes already. And I think there will continue to be. Great.

The a16z Show - a16z Podcast: Mental Models for Understanding Crypto Tokens

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.