a16z Podcast - a16z Podcast: Cryptonetworks and Decentralization -- Building Blocks

Episode Date: August 1, 2018

with Chris Dixon (@cdixon), Ali Yahya (@ali01), and Devon Zuegel (@devonzuegel) “Show me the incentive and I'll show you the outcomes.” At the end of the day, observes a16z crypto general partner Chris Dixon, Satoshi's whitepaper [the original bitcoin paper outlining a peer-to-peer decentralized network and blockchain sans centralized third parties] is nine pages of incentives. It's the kind of incentive design that you can use to build many other things on the internet (which itself is driven by very simple core protocols and could even upgrade itself as a result). But only with the right incentives (and alignment of those incentives among different entities), of course. Which is where cryptonetworks come in -- especially since they don't rely on hardware buildout (as with earlier generations of internet deployment), but rather on software (which is essentially just logic, the kind of building block you can use to build countless other things). The breadth of possibilities is endless. This means that platforms and networks (and operating systems, for that matter) can spend less time, energy, money, and frankly, suffering due to fighting -- thanks to distorted business models that lead them to extract value from users and compete among complements (vs. substitutes/better alternatives). The internet-native business models baked into crypto, however, could lead to greater competition and better options for users. But what are the missing building blocks, that can help make such networks more iterated games vs. one-off prisoner's dilemmas? And what will it take for these networks to truly reach web-scale, as it's still just the beginning? Because decentralization is the means to an end -- not the end in and itself -- observes a16z crypto partner Ali Yahya, so what do we need to build next to get there? In this episode of the a16z Podcast (guest hosted by freelance software engineer and writer Devon Zuegel), Dixon and Yahya share their thoughts on where we've been, and where we're going with the internet. Please note that the a16z crypto fund is a separate legal entity managed by CNK Capital Management, L.L.C. (“CNK”), a registered investor advisor with the Securities and Exchange Commission. a16z crypto is legally independent and operationally separate from the Andreessen Horowitz family of fund and AH Capital Management, L.L.C. (“AHCM”).  In any case, the content provided here is for informational purposes only, and does NOT constitute an offer or solicitation to purchase any investment solution or a recommendation to buy or sell a security; nor it is to be taken as legal, business, investment, or tax advice. In fact, none of the information in this or other content on a16zcrypto.com should be relied on in any manner as advice. You should consult your own advisers as to legal, business, tax and other related matters concerning any investment. Furthermore, the content is not directed to any investor or potential investor, and may not be used or relied upon in evaluating the merits of any investment and must not be taken as a basis for any investment decision. No investment in any fund advised by CNK or AHCM may be made prior to receipt of definitive offering documentation and due diligence materials. Finally, views expressed are those of the individual a16z crypto personnel quoted therein and are not the views of CNK, AHCM, or their respective affiliates.  Please see https://a16zcrypto.com/disclosures/ and https://a16zcrypto.com/disclaimers/ for further information.

Hi everyone, welcome to the A6 and Z podcast. I'm Sonal. Today's episode is on crypto networks and decentralization as a means to an end and on what building blocks are needed for us to get to web scale as it's still just the beginning. The discussion is moderated by guest host Devin Zougal in conversation with Chris Dixon and Aliya, both of A6 and Z crypto. Speaking of, please note that A6 and Z crypto is an independent fund managed by C&K Capital Management, a registered investment advisor or the SEC, it's separate from the Andrus and Horwitz family of funds. The content here is for information only and should not be taken as legal business tax or investment advice. It does not constitute an offer or solicitation to purchase any investment or a recommendation of buyer

seller and in fact the content is not directed to any investor or potential investor. It may not be used to evaluate or make any investment. CNK is not seeking investors. An investment in any fund advised by C&K may not be made without first getting final offering docs and diligence materials anyway. For more details, please also see A6NZ Crypto.com slash disclosures and A6NZ Crypto.com slash disclaimers. Finally, you can find this episode as a video as well with links, graphs, and more on A6NZ's YouTube channel. Hi, I'm Devin. I am a software engineer and writer based out of San Francisco. Today I'm talking to Chris and Ali from A16Z Crypto about potential crypto business models and the building blocks that we need to put in place to take

advantage of that full potential. So for starters, can you guys tell me a bit about why you're excited about crypto and where you see the potential is going? You know, it's a very, this, the cryptocurrency kind of movement is a very interesting thing because it can be looked at through multiple lenses. So, for example, from one lens, it's, it's a movement of software developers who are, you know, kind of analogous to the open source software movement that started 20, 20 plus years ago of, you know, it's a movement to people that are kind of collaboratively creating communities, and building projects together and building networks. You know, you can look at it as a lot of the kind of press does

for more of a financial aspect of like, you know, the prices of Bitcoin and just so like as a, you know, store of value and a new kind of alternative financials. You know, being kind of, being technology focused, we kind of, we like to look at it, or one way we like to look at it as a, as you can think of a crypto network like Ethereum or Bitcoin or one of the many other new,

innovative networks out there as a type of computer really. And so whenever you have a new computer, if you kind of look at the history of computing, you had different kind of new computers introduced along the way so you started off with like mainframe computers back in the 40s and 50s and then you had

in the 60s, I guess, and then you had the PC really kind of start in the 70s and rise to prominence in the 80s. You know, you obviously had the internet in the 90s you had smartphones in the late 2000s and each time you had a new computer you had new kind of things you could, capabilities these, it unlocked new applications.

So let's just take the smartphone as an example. That was the first time we had a computer where you would actually have it on your body with you all the time as you go around, right? It has a built-in high-quality camera. My iPhone does or your Android phone does, right? It has GPS. It has like all these other interesting things, which you didn't have in computers before. And that then unlocked, if you think about like some of the most popular applications

on smartphones, it's like, for example, Uber and Lyft, right? take advantage of the fact that you have the phone with you all the time and you have GPS and the driver has GPS or you know Snapchat and Instagram take advantage of the camera that you always have with you and you can do interesting things with it right so you had a new kind of computer and had new capabilities and that unlocked new applications that kind of uniquely exploited those capabilities right so if you think of if you think of a blockchain computer like a theorem as a new computer through this lens the way you think of it is what what are the new capabilities of this computer right and I think one of the things that kind of throws people off in the space, when they think about it is they're used to thinking about new computers just as sort of being better in terms of performance and size and other things. Crypto networks like Ethereum are actually

kind of worse than some of those dimensions. So for example, in the performance side, it's not as performant as, let's say, AWS or Google Cloud Services or something if you wanted to write a video game or something, it's not as performant. But it does, it is, it has had this other very, very powerful new property, which is around trust.

And so the idea that through cryptographic and game theoretic mechanisms, a blockchain computer can provide, you know, provable guarantees that, for example, if it says, if I, if you know, if you have a currency like Ethereum and if that computer says that I, you know, by owning a private key, own a piece of that currency, I can trust that that is the case, right? And so I know. And so it allows you to do things like financial things. So if you think about the basics of any kind of financial operation, whether it be something simple like owning something or having property rights or owning a piece of currency or owning a digital good or something like this, it requires trust. It requires trust that you do indeed own it, right?

And so there's different dimensions in which sort of a blockchain computer enables new trust. One is with users. So users can feel like they own something, whether it be like a piece of property or a piece of currency. it also creates new trust with developers. So software developers, you know, when they're building on a network like Facebook or Twitter, you know, there's a long history of a whole bunch of software developers who tried to do that. So, you know, Zinga is a very prominent company that built games on top of Facebook. There were all these, you know, Tweedy and all these, like, kind of Twitter clients that built on top of Twitter.

There, you know, they had to trust the centralized organization, i.e., Facebook and Twitter, those developers did. And, you know, and what has happened with many of those large networks is that over time, their business interests change. So Facebook's did and, you know, and Twitter's did. And that ended up having, you know, negative effects for those developers, right? Contrasted to the first year of the Internet when developers were building on an open network, you know, if you think about the web standards of HTTP and email, SMP, and all these other kind of open protocols that were kind of governed by digital.

centralized protocols and non-profits and things like this, back then developers could build on it and know they trusted it, right? And so blockchain computers reintroduced this idea that developers can build things and trust them, which by the way has kind of a cascading effect. It means not just the developer can trust them, but an entrepreneur can trust them who builds a company around the developer. The investor behind the developer can trust as well and know that if you invest in this thing that you'll have sort of guaranteed kind of developer rights, right?

the same reason. If you look at economic studies, like one of the top predictors as to whether a country receives outside investment is whether they have strong property rights. And so a similar idea here, that if you have a network that has strong kind of rights for the users and developers, people are more likely to invest in it. And they're more likely to trust it and build on top of it. And the users are more likely to adopt it. And so I think what's very interesting now is we're in a, you know, we're 10 years into cryptocurrency in the sense of Bitcoin was in, you know, the paper first came out in 2008, but really the design space really dramatically expanded, especially sort of the rise of Ethereum around 2015, and since then many more of smart people

entered the space and have been having always creative ideas. And we're in this very interesting period now where people are saying, what can you do? You have this new kind of primitive, this concept of trust that you didn't have before with large-scale computer networks. And now that you have that, what are all the things you can create? And so, you know, you can create, for example, a pretty new idea that just sort of happened last year was the idea of crypto collectibles, crypto goods. And this came out of Ethereum, it's called ERC-721, is this sort of a new idea that something came up with,

which is you could have, instead of just having sort of digital currency, fungible goods, you could have non-fungible goods, and they can have graphics and all these other things, and you have things like CryptoKitties, and you're just at the very, very beginning of people sort of thinking about what you can do with this new concept. One of the most exciting features of this new building block is that collaboration with humanity

tends to be bottlenecked on the requirements for trust. And so if you have a, a technological building block that reduces the activation energy for collaboration to happen, then that can open the space of possibilities dramatically. And I think that there are a number of different things that people tend to look for. There are three great facilitators of collaboration. And they all revolve around trust. One of them is the requirement of trust in mechanisms of enforcement,

that there are guardrails around the collaboration that people believe will hold. and will prevent certain bad outcomes from happening. And there's also the existence of an incentive structure that enables the participants to believe that their counterparty or counterparties will not behave in a certain bad way because it's not in their interest to do so. And then there's also trust in social signals.

If there is some mechanism of reputation that exists and is in place, then people are able to rely on those to feel good about the way that our collaboration is likely to go. And I think everything that Chris is talking about around trust, like blockchain technology facilitates two of those very well so far. It provides a new mechanism of enforcement because now all you have to do is trust math. You trust the algorithms that support the blockchain

to feel good about the kinds of interactions that will occur and the kinds of outcomes that will be. possible. When you transact with someone on the blockchain, you don't have to rely on some external, some third party to believe that the transaction is going to go the way that it was sort of constructed from the start. And it also helps with trusting incentive structures. You're able to build sophisticated incentive structures that are simply just code that give you confidence that participants in any one particular interaction will behave any one way. The one that is still missing is that of social signals. There's still no real good mechanism for reputation or for

identity on the blockchain. And that's one of the things that many people are working on, and that's a building block that will be needed going forward. And we're still very much at the very beginning of this new technology. Even though it's been 10 years, it's still nascent to that. To add to that, one of the challenges in building these systems on either blockchain systems or systems on top of blockchains, is all of these systems today have to assume that any participant in the network could be adversarial, meaning you don't know who that person is.

So you just have to assume the worst. And therefore, you have to build the trust through the kind of game theoretic mechanisms and the cryptographic mechanisms that the system as a whole has these emergent trust properties, even though individual actors could be bad. If one were to have a better kind of system of reputation, or identity, you could relax that constraint,

know that participants have some history, you're playing kind of an iterative game, and it would enrich, I think it would widen the design space of possible applications you could build. What would these identity or reputation systems look like? Well, I think one of the hardest problems in the space is mapping cleanly, one public key or address on the blockchain to an actual human being.

So this is known as the civil problem, the problem that any one participant can create multiple addresses for themselves and pretend to be many different people. And so having a good way to map an address to a real person is a very difficult problem would solve a lot of these problems because then we could associate some metric of reputation with that particular identity. And so I think in the old web, in the web 2.0 and even web 1.0, this was solved via DNS where you have an authority certify that a particular domain name on the internet corresponds to a particular organization and to their IP address and their server. And that allows people who are interacting with their

websites to know that they're actually talking to the right people and that you know, they will sort of, their reputation is actually theirs. And so some equivalent in this new crypto world is necessary. And you can imagine many different solutions. You can imagine

people trying to build something like the certificate authority model for crypto where you have some authority that vouches for the mapping between a public key or an address and an actual human being. Or it could even be more decentralized than that, and it could be that people vouch for each other's identity on the blockchain. You can end up with a web of trust kind of system where the trust is emergent and is bottoms up as opposed to top down.

That's really interesting, and something that's newly possible with blockchains. What are other ways that blockchains enable identity? to be sort of treated in different ways from Web 1 and 2.0? Well, I think there's a related topic which is kind of control of user data or what people sometimes call data sovereignty. And so, you know,

as we see in the news, this is a very hot topic. You know, GDPR is a new European law, which is going to impose some more regulations on tech companies. And just this general kind of meme out there that, you know, you sort of had made these agreements with social networks and other kinds of search engines and other things,

users had sort of used these services expecting their data to be kept private. Of course, the business models of these companies is to use that data to target ads and things like that. And there's a sentiment out there that maybe that kind of has gone too far and that we need to kind of swing the pendulum back. And as technologists think more about architectures that allow users to have more control of their data. I think this will become more important over time, just number one, as obviously the Internet becomes, you know, it will continue to grow, will continue to work its way and more deeply in other way, part parts of the economy. You know, I think also, you know,

we have these issues like, you know, healthcare data, you know, who's going to hold the healthcare data, who's going to hold AI data. So, you know, one of the big. big sort of other megatrends in the world right now is that AI is getting better and better the algorithms are getting better and tech companies are investing you know tens of billions of dollars into this the you know the really valuable thing that people have that that need for these to create AI systems is they need algorithms and need computers but they also need data and data is really kind of the scarce resource and you you create data I'm creating AI data right now as I'm speaking and we all are right so throughout our lives we drive around we're creating AI data

And, like, that's what companies like Google and Tesla and things are trying to capture, is that data to get better and better in their algorithms. You know, and one question is, should that data just only reside at these large tech companies or should it be an open asset that other people can use? Should the users own them? Like, I don't know what the answer is, but it does seem like we shouldn't just sort of blindly assume that the old model for how we kind of gave data away for free to tech companies is to it should always be the right model going forward.

Well, there are a lot of business models on the web these days, But one of the most dominant ones is that users will do behavior on the web. They will input posts on Twitter, on Facebook, they'll click on ads. And they're getting the product for free, but it's not truly free in the sense that a lot of these services then use the information that they're getting from them. What are some new business models that crypto now makes possible? Yeah, I mean, it's a great question. I think one of the most interesting ones is sort of the quote, business model. model of, let's say, Ethereum or Bitcoin, and it's a brand new model, and let me explain it.

So the model of it, let's just take Ethereum, is they have a certain number of, a certain amount of ether, which is the kind of native currency to the network, and the way it works is sort of the company, it's really the foundation, non-profit foundation, the developers behind it, have a certain amount of ether, which they use to fund the operations of the company, right? So let's just say, like, the rule of thumb is 10 to 20% of the currency is kind of held back for funding the operations of the company, and it's kind of the business model of the company, right?

And the idea is if they go out and build a great network and it becomes really popular and successful, demand for that currency will go up, the value of their holdings will go up, and that's their, quote, business model. What's so nice about that model is it also lets the users participate in the upside of the network, right? So, you know, I was an early user of Twitter,

I don't know, whatever, more than 10 years ago, you know, what if all of the early users who kind of helped kind of, you know, galvanize the, you know, the early network before it really had crossed over and gotten kind of major mainstream network effects. What if those users could have participated in the upside of that network, right? You know, what if, you know, the developers who were building stuff on the side of it could have also, you know, precipitated and they could have held tokens as well? It would have been a great thing because everyone would have been aligned. The, you know,

the Twitter itself, let's say had it had this model, they would just want to see the network at use. They wouldn't need to say, hey, I need to take away. You know, if this, if Tweedy is getting more users and that makes the demand for my token go up, like that's great, right? I don't need to, why would I cut them off? It's the opposite. I want to help them, right? The users would say, hey, you know, I want to go tell my friend about this thing. I want to get them on there because I own part of it. I want my friend to own part of it, and we can all, like, kind rise together, right? And so it's a really just sort of elegant and kind of beautiful business model where it aligns all of the participants of the network, whether they be the core

development team, the users of the network, the developers, you know, other kinds of parties. So it's really, I think it's one of the great business model innovations of the last few decades. And I believe, you know, having worked on the internet for almost 20 years now, to me it's finally like the native business model of the internet. Like I think up until now we've been grafting really what are just like you know like you think about the ad model it's like you know it's an old model it's billboards and tv ads and they're intrusive and you know there's a whole bunch of issues with them and most people don't really like advertising but you know that's what we knew and that's what we sort of took and so in the early beginning I think when you look back on the multi-hundred

year history of the internet yeah at first it's sort of like whenever you have a new media you know you tend to sort of impose you know older ideas on to it you know and if you go back and watch the first movies that were made. They had, they were like plays and they filmed them and they didn't understand that you could do. It's a whole new grammar and a whole new kind of thing. And so I think, you know, we're kind of still pretty early in the development of the internet. And we've now discovered this great business model, which I think will be the dominant business model of the internet and is the true kind of native business model of networks. And then the corollary is that crypto tokens, coins, assets are the native kind of asset

of the internet. And as more and more networks are built and grow and become more important in our lives, those assets will become more important and more valuable. The Twitter example is fascinating to me, because like you were describing before, Twitter currently has to monetize directly through the platform. They have to put ads on there somehow. They have to have promoted tweets. And for them, having these third-party apps that may be nicer clients that do different things

for users that they want, It's just something that they have to maintain. And, like, sure, it does, like, on net increase the value of the network, which eventually comes back to them. But mostly, it's an externality. If it were a crypto network, they'd be able to capture it. That's right. And they would all be aligned, and they'd all be helping each other.

And it's just a really beautiful thing. I think one of the people underestimate it. If you've been in the tech business, you've seen this. But I like to say the most vicious battles in tech are not between what economists call substitutes. They're between compliments. And so what I mean by that? So a substitute is, you know, you're at a picnic and you're going to have you with a hot dog or a hamburger, right? A compliment is when you have a hot dog, you also want a hot dog bun.

It's a thing that goes along with it, right? In technology, you know, you have, so, you know, the equivalent of the hot dog and hot dog bun is, you know, is windows and, you know, quicken or something back in the old days. And like in the modern world, it would be Twitter and tweet deck or something, right? Those are compliments. They each make each other better, right? But the most, if you actually look at a history of technology, the really vicious battles are between, complements, not between... I mean, the substitutes have battles, too, but really

it's the compliments. Because what happens is there's this kind of natural S-curve development of platforms and networks, and so I'm including, by the way, operating systems, I think, they're kind of forms of networks, they're platforms of developer networks, so Windows, iOS, Android, Twitter, Facebook, I think Google Search, in a lot of ways, can be seen

as a platform because you build websites that get shown in the search results. And if you look at the history, it, like, you know, like Yelp is up, you know, going and testifying in front of Congress talking about how unfair it is if Google is listing their own restaurants ahead of Yelp, right? They're not go up in front of Congress complaining about the substitute, whatever, you know, the competitor to Yelp, right? You know, they're famous, their books written about all the stuff that Microsoft did

and huge antitrust case and everything else around all the things they did to fight with their compliments, right? You know, the Facebook Zingas and the tech world's famous and the, you know, the Twitter and its ecosystem is a famous. battle. And so we spend so much, it's really kind of a shame, I feel like, you know, someone who works in the industry of like how much time and energy and money and, you know, and just suffering is around like intra-network fighting, right? And that has to do with the fact that we have this, I feel like we've had this kind of like distorted business model

all along, which is like you've got to like control everything and own and everything. And it's kind of hoarding kind of mentality because it's just, it's forced on, I'm not criticizing these companies necessarily. It's just the way the logic of the business model and the way the capital markets work and all these other kind of forces that kind of force people to do that. But now we have a better way. Like now we have a way where these people can be aligned and there will still be competition, but it'll be competition between networks and much more like substitutes. And, you know, competition is a part of any kind of market system. You know, you want it, but you want it to be healthy, right? You want it to be like, for the

right reasons. You want people to be competing by creating better products and more useful things. as opposed to leveraging my existing incumbency to hurt the user experience to get more profits, right? So I think it will still be competition. I'm not sort of being naive about that, but I think it could lead to sort of more healthy competition and more intra-network kind of harmony. Incentive structures end up defining almost everything at the end of the day. That's what Charlie Munger famously said, show me the incentives and I'll show you the outcome. Yeah.

And crypto, I think, has created a whole new set of incentives. It's turned things like the Twitter situation from a one-off prisoner's dilemma into an iterated game. What are some other places where you guys see this dynamic playing out right now or on the horizon? Well, to some extent, I think that the technology enables broad experimentation with different incentive structures, simply because it's now a software. So you can now use code to express all sorts of different kinds of incentive structures that participants can engage with. And whilst Ethereum is one manifestation of this, I think there are a very large number of different experiments that are currently being run.

Yeah, I mean, as, as Lee was saying, one of the amazing things I think about software is that, you know, I think it's software is sort of misunderstood sometimes to be sort of a field of engineering that's sort of a relatively, you know, constrained set of decisions you can make when in fact, I mean, you know, as we know from like mathematics and philosophy and things, that if something can be expressed in logical form, it can be expressed in a term complete programming language. And so I think people really underestimate kind of, you know, the breadth of the possibilities of what you can build software. And one of the beautiful things about this world, the crypto world, is that, you know, the internet is basically, I mean, it'll always get better, but it's basically, the hardware side is basically built out, right? We've got broadband in a lot of places. I mean, we need more. And there's lots of places in the world. We're developing world, for example, we need to get more internet access and more smartphones out there and things. But it's, but it's pretty well built out. Like a lot of, you know, three billion, three and a half billion

people have smartphones and internet access and that'll continue to grow. And so right now it's mostly about software and you know and the beautiful thing with you mentioned incentives is that if you think you know the way the internet is architected right it's one of the genius one of the many genius things around the internet is the the idea that the the core protocols are relatively simple the brains sit at the nodes not in the edges right they sit at the on the computers the protocol itself like TCP IP and all these things are UP whatever they're all very relatively simple and that means the internet has the ability to kind of upgrade itself with the right incentives. So if you build, you know, if you look at, there's two ways to look at Bitcoin

mining, and you'll see some articles that criticize it for taking too much electricity and everything else, and that some of those critiques are valid. On the other hand, like, it's kind of amazing that a, you know, eight, nine page, whatever, the original white paper, nine page white paper, someone can write nine page white paper, and ten years later, there's, you know, there are thousands of data centers filled with computers running software, right? I mean, that just shows you the power of incentives, right? whole new economy on top of this way. That's right. And it all came from, it literally is, I think the best way to think of

the Satoshi's original paper is nine pages of incentives. I mean, there's computer science and game theory and other important things, but the incentives are kind of the key part. And those incentives created this whole new industry. And so now what if you take that and you kind of riff on it and like what else can you do when you design those incentives? What if instead of those miners just kind of performing idle computations, they were actually storing useful information? And what if you took those same incentives? and you had them do things that would provide a purpose that does caching and redundancy and all sorts of other things you might want from a storage network.

And there's a whole bond, and that's just the beginning. I mean, there's a thousand interesting ideas of riffing on how you can use this kind of incentive design to do other things. Yeah, because IPFS and that whole ecosystem of tools seems to be sort of the next step of things that need to be built out before we can build out the application layer. So you mentioned broadband. You know, if you don't have internet access, you're not going to be doing very much on the internet. Once that's fully deployed, and we're getting closer and closer, then there's sort of another level of infrastructure. Right now, we have Amazon AWS, but we also have other things for the Web 2.0. For crypto, it seems like IPFS is sort of the AWS of crypto, or at least the S3.

Can you dive into some of the other building blocks that need to be implemented before we can expand the potential? I mean, so yeah, I think there's also just on the compute layer, there's a whole bunch of things need to be done to make these computers more scalable. And so there's a whole bunch of work going on at different levels, both core blockchain work. So, for example, Ethereum is working on moving to what's called proof of stake, but also sharding and a bunch of other things that's sort of so-called layer one scaling, and then there's layer two scaling, which is like what they're called state channels and plasma chains and all sorts of other stuff. And then there's sort of side chain stuff, and there's like this, there's interesting work in, like, fundamental cryptography around these are these techniques called like Stark's, for example. that let you do more computation off-chain and only do verification on-chain, which gives you effective performance improvements.

Intel has this technology called S-GX, which also lets you do something similar, which you can do more computations off-chain. There's really interesting kind of side network design. So there's all sorts of stuff to sort of just generally scale and make these things web-scale. So by web-scale is a term we use in our business to mean something used by billions of people, right?

Right now, like, some of these networks are not at that scale. So you need the core kind of network to scale, and then you need other things that allows you to do kind of private, secure transactions, because most blockchains are, most of the data is, all the data is public. There's, as we were talking about before, like things around, like reputation and identity, around the user side. There's just a lot of basic tooling that, like, you're used to, I'm sure, as a software developer,

I think you would know this better than I would, but, you know, better IDs and debuggers and whatever, just like the whole thing, you know, libraries and, you know, verification, you know, security review, and like a whole bunch of other kind of stuff that you just want, you know, in a modern development environment that just doesn't exist yet because it's still relatively early. So it's really a long list of things. I mean, you can look at that as like, okay, we got a lot of work to do. You know, that's true.

On the other hand, it's also a great opportunity, you know, there's just so much more. It's sort of this, you know, we just discover this new continent and, yes, we don't have houses yet and we don't have roads yet, and you can look at it that as a negative or you can flip it around and say, you know what, this is like a great place to go and have a career and we're seeing more and more of that happen and that's one of the really

exciting things is that we're seeing just a real wave of talent of people coming out of colleges and tech companies and other places and saying you know what I'm sort of tired of living on this other continent it's all filled up and it's gotten kind of mature and mainstream and it's not that exciting anymore and I want to go I want to go get in the wild west go in the pioneer be a pioneer and that's how it feels so there's a lot of work to do but it's also like a lot of opportunity and there's a ton of smart people working on stuff. Plus, all this stuff had to be built out for the web to begin with. So it's not like it can't be done.

I think there's a lot of revisionism today. I would argue, having lived through the web to build out, like people sort of, I think, sometimes imagine that it kind of popped out, fully built or something. The experience of using the Internet in the 90s, it was literally like you literally had to wait for a single image to load and it would go like this and actually you would change your browser settings. Sometimes you would have like different rendering methods

because it was that slow. and video did not work. YouTube finally, like, it was 2005 when you finally had, like, enough broadband out there. The websites, it's always funny to look at those 90s websites things. You know, they had the animated jiffs and, like, all the other ridiculous stuff,

and they just, and they didn't have, like, responsive design, and they couldn't, you had to reload the page. You know, it was just, like, a fun thing to do for people that think the web came out, fully built, is to go watch movies for the 90s. So you watch movies from the 90s. It's, like, a good historical record of the Internet in people's daily lives.

lives in the 90s. And the answer is there was no internet, except every once in a while, you'll see somebody go and they'll, there were no smartphones, no other thing. It wasn't a thing. You didn't even talk about, like, LinkedIn and Facebook and Twitter and Tinder and whatever, and they were out at dinner, and they were like, who is in that movie? And they're like, I guess we'll never know. I guess we'll just never know who's in that movie. That's how it was, right? Like, that was what you did in the 90s. And no one thought you could just, well, let me take out my magic supercomputer, like, you know, and that's connected to all information in the world and type, you know, a few words and then know the answer to everything.

Like, it didn't happen. And what, when they do connect to the internet in the 90s, they would be this thing where they would go and it's like a ceremony. They would go in front of the computer and then of course the screen was like this thing that was like, you know, 20 inches deep and then there would be the sounds and like as your modem is connecting and then you've got mail and then you'd spend basically the only application

of email in the 90s for really that people use besides like, you know, developer types or nerds or whatever like us, was email. In fact, there was a famous McKinsey study and it was like late 90s or something like would you want broadband and no one said and they surveyed people and were like why would I pay and it basically but people said no and they they people had

included the broadband wouldn't get wide adoption because at the time broadband but just meant faster email and well yeah that's pretty good you know whatever I don't need to get faster so so the web you know and by the way like the other thing about the web is that you know the idea that today was a fait accompli and like it's obvious like Bill Gates wrote a book in 1995 called the Road Ahead, predicting what's going to happen. The future is 95. This is, you know, I think this is well after Yahoo and Netscape and things like this. And the word the web did not appear in the book, right?

They actually released an edited version the year after. They added it back in. He did, however, talk about the Information Super Highway. So there was two revisions of the world back then. People sort of, you know, some people just didn't believe in any of it. And then a lot of people didn't believe in any of it. And then some people believed in the Web and some people believed in the Information Super Highway. And Information Super Highway was going to be, like it was like Comcast.

and Disney and Microsoft, and they were, like, running broadband fibers into, like, a town in Florida. I forgot which town, and they were going to have... And their argument was, like, look, it's going to be this top-down centralized thing, and we're going to have... But it's going to be awesome because we're going to have high-speed video and day one, and we're going to have all these other things. The downside is you literally had to install, like, a $20,000, like, whatever it was, like, a SGI computer in your house or something. And then the web was this rag-tag militia of, like, hackers and developers who were, like, yeah, it kind of does suck, and, like, it takes forever to download and, like, even load an image

and nothing really works, but you can just, like, anyone in the world, can just take a computer and connect to the Internet. And then what happened was, you know, it inspired all of these developers who said, you know, this is cool. Like, this is, I could do something. Like, what if I created a website that did this? What if I created a website that did this? What if we did create it, you know, back then, like, you didn't have encryption on the Internet.

You didn't have a Nescape. People forget sometimes. They create a JavaScript. You have a programming language. They created JavaScript. They create SZL. They created cookies. You didn't have a way to keep state.

That's what cookies are. And so they had to create all. these things. And, you know, there's sort of two sides of the coin. There always are with these new technologies. One side of the coin is, wow, this thing is early and this is not much. The other side of the coin is, oh, my God, look, I can make a great career out of this because there's so much opportunity and so many new things to do. And so that, you know, to the web, I mean, really, I don't think the web is seriously usable by mainstream people until the mid-2000s. And you could even argue even then, like, it was still this thing

you went in. If you look at the statistics, like the time spent on the web was not that great until smartphones really until it became just like this thing that was always the whole concept of online offline right like the very fact that you had to like go and go online was such a hurdle that most people like I'm going to be online now I'm going to go surf the web it was like this thing you did that was kind of like TV but slower or something you know now it's this sort of this constant state that we're in for better or worse where it's just as you know it's finally fully really built out so that's a long way of saying it took a long time I'm more optimistic about the crypto stuff happening sooner because it's software-based, meaning the hardware's already built out.

And now we just need to, like, a few quick, you know, a few clever incentives and some nice software, and you suddenly have, like, you know, some rapid adoption. Earlier, you mentioned the difference between the information superhighway and what became the web. And we all know information superhighway is not really a term we use anymore. Can you dive a little bit into the difference between the growth curves for both of those things? There's two types of, two ways to take a tech product to market, right?

And the first way is there's sort of a one-step process where you have to build a centralized system and go directly to end users. And there's a second way to do it, which is to first to sort of build a platform for developers and then have those developers build things for end users. So an example I like to use, it's not a strict analogy, but is Wikipedia versus Enkarta. So Enkarta was a Microsoft-owned encyclopedia product that was sort of dominant in the late 90s. And so it was like you'd go there. And it was very nicely done. It would just like you look up like whatever, like a tiger. and you'd see a page for a tiger

and someone wrote that out and it would have a nice graphic and everything else. And then Wikipedia was started in 2001 and it was basically a bunch of blank pages, right? And but it had this ability for anyone to come and become a contributor and to have a discussion about it

and do all sorts of other things and it started to get some more and more users. But, you know, at the, it sort of, it had a different, it couldn't just be, it wasn't a one step go to market. And by that, I mean, it wasn't, you built Wikipedia,

like Jimmy Wales and his co-creators and then regular end users came to that website and got value out of it, right? Instead, he first had to sort of say, hey, let me go get these contributor communities excited about contributing to it. And then slowly over time, as they'll create some content, and then some users will come,

and that will sort of make the contributors happy that people read their content. And then other contributors will see that, and then it'll sort of get this flywheel going. And if you look at the today, you can look at the growth curves. And Wikipedia took a while. And it was actually very controversial, like 2004 and 5

because it had a decent amount of misinformation and just gaps and other kinds of things. But then it just like hockey sticked up once it got going. There was a famous nature article in 2007 that studied its accuracy against like Britannica and Carter and things and said it was actually as accurate. So that's kind of what hit that point.

It was getting banned on campuses and things. It was very controversial. Today it's kind of funny to realize that because it's so obviously the gold standard. for, you know, for a lot of this information. So, and then it just sort of hockey sticked up. Meanwhile, Encarta, you know, started off in a better position because on day one, they could have, you know, a really nice page for a whole bunch of content.

But over time, they just couldn't scale in the same way. And they shut down. I think it was 2009. Microsoft shut it down finally because it was just getting crushed, right? And so I would use that as an analogy where, you know, Karta is the equivalent of kind of a centralized system. So, like, the Information Super Highway was an attempt to do a centralized version where you would literally have a company or a set of companies go and lay all these cables and build

all these services and build content and do the whole thing end to end. And the advantages on day one, it would be really nice, just like Encarta was. But over time, it would never get that flywheel. And that's what the web won on it. So the web started off worse, but it got better at a faster rate by getting this flywheel of entrepreneurs and developers and users. And so, and that, you know, so the web did that as a great example, like Wikipedia versus Encarta is a great example. In the case of the information superhighway, There was no real way of scaling it as well. It wasn't one of the limiting factors that you can't really do that for the entire globe or for the entire.

Yeah, I mean, the thought the theory would be, you know, the way you'd, the theory they had was you would try to scale it like you would any centralized service was you just like get a whole bunch of people to sign up and they'd pay more money than the costs and all these other things. And you would just kind of fund all the content development and do all these other things. But, you know, they had that theory, but, you know, the practice, you know, we know how that history ended. Look, I don't think it's one-size-fits-all. I think there are cases where you want, you know, centralized services and other cases we want decentralize, and there's different pros and cons. I would argue that the pendulum has swung too far towards centralized today.

So we have, you know, essentially, you know, the vast majority of people's time they spend on the Internet now is spent at Google, Apple, using a service by Google, Apple, Facebook, Amazon, Netflix, et cetera. And so, you know, and I think that, I think that I believe we're at sort of peak centralization and we're going to see more and more interesting things happening and sort of this kind of you know it's like a lot of things in life like politics and you know fashion trends and other things like there's a natural kind of equilibrium

state and we've sort of really hit sort of centralization and now there's going to be more and more and that's motivated through cultural factors I mean you know people who want to work on new stuff and entrepreneurs and all sorts of other things like don't you know like the incentives I was talking about earlier of like not having the right incentives when you build on centralized systems It just naturally pushes a lot of entrepreneurial energy outward, and we're seeing that, and that's a lot of what we do, and our investment businesses sort of try to back those people. I think people talk a lot about decentralization as the thing that is the ultimate good, the thing that we should optimize for, but there is this question as to why.

Why is it that decentralization matters? What is it actually by you? And one of the frameworks that we've been toying around here at the firm is trying to break down what are the four primary things that actually matter for end users and that actually matter for developers? So the four are security, which is sort of the obvious one, because that's the kind of thing that ultimately inspires trust. There's scalability because in order to build anything that is web scale, as Chris was saying,

that billions of people can get on, you have to have some mechanism via which technology can scale, and that's not really the case yet. There's usability. how do we interface with end users and with developers? There has to be a usable experience that is seamless and that is at least comparable to existing solutions for perhaps the same things that they may be trying to do. Yeah, on the topic of usability,

something you've talked about, Chris, is the importance of stable coins. Can you unpack that a little bit? Yeah, so stable coins are very interesting concept, which we've made a few investments in, which is the idea of having a cryptocurrency, where the value of the currency is pegged to something like the dollar, for example. It can be pegged at different things, but the dollar is the obvious thing.

And the advantage of that is that if that can be properly done, is that it makes, you know, kind of the financial use cases for cryptocurrency much more mainstream accessible. You know, most people want to denominate things in their local currency. That's how they think. That's what they pay their taxes in. That's, you know, what they need. So, like, if you want to have, like, a mainstream lending platform, you know, or payment platform or other things like, I believe that a stable coin would be a critical

ingredient in that. And there's a bunch of really interesting attempts to do that, to kind of how do you, the question then is how do you create that kind of technically, how do you create that mechanism that keeps the coin stable, right? Because just in a normal open market, the reason that the prices of Bitcoin, Ethereum, et cetera, go up and down is that they just change with supply and demand. So how do you do that? And so there have been a lot of interesting work.

In my mind, this is one of the really interesting things that I think will likely start to become more prominent over the next year or two and could unlock a whole new wave of kind of applications and just broader mainstream adoption of cryptocurrency. Awesome. And then there is a process via which the protocols themselves can evolve, which is governance. Because given that we are in this decentralized context, there isn't really a good answer as to how you evolve forward. If it's the case that no one really controls the crypto network, how do you change things about it that don't quite work?

So there's an important, like, last pillar. And so for the four of them, we talked a little, Chris alluded to many different building blocks that are contained within each. For example, in the case of scalability, there are solutions like sharding and, like, plasma on the Ethereum, on the Ethereum blockchain. There's this Intel SGX, there's Starks, so we covered a lot of those. In the case of security, I mean, one thing that's interesting to note is that there are various different threat models and different applications require different levels of security. So certain blockchains, like say, like the Bitcoin meme is that Bitcoin is fully decentralized and is extremely secure, secure to the point in which it will be resilient even against attacks by nation states like China. But there are applications out there that may not need that level of security. And so there is room for other kinds of crypto networks to emerge

that don't necessarily strive for that level of security. So it's an axis via which developers can sort of explore different possibilities. In case of usability, one thing that I think is fairly underserved by the current business models, as we were talking about earlier, the token business model, which is sort of a tempting thing to do in the current climate, the current sort of hype ecosystem, is working on developer tools,

making the technology usable to developers in order for that flywheel to emerge. And similarly, building good interfaces for end users who don't care at all about decentralization to come on board and to begin engaging with the technology. And then finally, this is a question of governance, and I think there already is a mechanism of governance, which is that of open source.

And people organize themselves in a normal sort of human way. They decide which projects to work on. They decide whether or not they should continue to work on those projects or fork them. So basically the act of taking the code and creating a new instance of it and going on your separate way. But there are also ideas as to how you can streamline that process. How do you make it so that humans or a small group of people don't have outsized control over the evolution of the network? And so one idea is you can also build a protocol for the evolution of the protocol itself. And that would be like an on-chain governance kind of solution where people are able to vote.

They're able to express their opinion as to how the system should evolve. And the system would take that into account somehow in order to make a decision. And this depends heavily on building blocks that I think as we were talking about earlier, identity and reputation are a very important building block that helps make the games that people play with one another iterated as opposed to one-off Prisoner's Alema-style games. So that's maybe a broad outline of a framework for which we can think of the many different building blocks that are needed. Decentralization is a means to an end.

It's a means to each of these four things. It provides you with security. If done correctly, it can provide you with scalability of the underlying world computer. And it can help, I think, I mean, usability and governance are things that we have to build despite decentralization. The centralization makes it hard to build those things, but those are absolutely necessary for any of this to work. Well, thanks, guys. I think we'll wrap up here. Thanks for taking the time. Thank you. Thank you.