CROPS Must Bring Food To The Table

Transcript

He's a security engineer at Brave and co-author of EIP-6963 for multi-wallet interoperability. Among many contributions, he has worked at Brave Wallet on-chain privacy and key management standard. Today, he will talk about loose leaders software and how it has become a key driver um in the propagation of open source software. Please welcome Kyle Den Ortok. >> Hello, everybody.

Uh yes. Um so, I'm sure many of you guys have heard about the concept of crops. Uh so, what I want to talk about is how do we balance it with the sustainability of actually funding it. Um so, first, who am I? Uh I'm a security and privacy engineer. I work at Brave browser. Um I've also done open source development basically my entire career. Uh previously, I spent um 5 years working on digital ID and and open standards. Um so, in that time, a lot of what we were building was kind of like base layer capabilities uh to be used.

Um originally, I got into the space in 2016. Uh and that was actually just through buying and trading cryptocurrencies. Um and it was at that time that I realized that like, "Hey, this party's not going to last. There's a lot of speculative capital that sits in here and there is a bull and bear markets that come with it." So, essentially, I wrote a uh blog post about this just recently. Um and that's kind of the where this uh talk comes from is splitting it into two parts. First, um how did Web 2 apply the concept of lost leaders software? And then, how do we apply it to Web 3?

There we go. Uh so, to start off like, I want to basically talk about like, how do we define what is a loss leader software? Um it's essentially derived from the economic concept of a loss leader, which is essentially a product that's designed to get somebody in the door. So, if you imagine like Thanksgiving sale or like a Black Friday sale where they've got the uh TV that's like $200 and you're like, how do they even make money on that? That is a loss leader. The whole design of it is get people in the door and then you can make money on other aspects of it. Um so, when you apply this concept to software, it's essentially how do you acquire users? So, if you can get users in the door, then you can somehow

convert some of them into a paying user. Um and so, what I want to talk about is why do we actually need these uh loss leader software? Um I'm sure many of you guys have realized that like we still haven't achieved the next billion users yet. Um there is a yin and yang balance that occurs here between it and in order to get that like next billion that we talked about for years, you first have to actually um bring the users on chain and then once you bring them on chain, that's when you can try and derive value from their different usages and stuff. Uh so, in order to do that, um we've done some parts very well. We've provided free open source software. It's

loved. You know, there's there's wallets, Ethereum, Bitcoin, all of these different sorts of things. They're different like economic values that are serving real world purposes. So, you know, you go to Argentina, you go to Turkey, you go to Nigeria, you go to Vietnam, there's people who are actually utilizing these technologies today for real purposes. So, what we're doing is working, but at the same time, I've heard numerous people who are working and going, "Hey, I'm not I'm leaving the space. I'm not able to afford it. There's companies that are shutting down.

There's trade-offs that are occurring because we didn't actually have a sustainable funding model. Um a lot of our operations have behind the scene operated on kind of an altruistic business model of like, "Hey, we're going to provide some grants or we're going to do some VC funding and hopefully we'll figure out the business model." But in the same way that Web 2 found that this didn't work for us OpenSSL uh Heartbleed I I'd like to remind people of um we need to kind of level up how we're approaching the concept of of the business models and distributing the actual crops that we want. It's all good that we want to be able to uh do the higher principles, but we need to do it in a sustainable way.

So uh one thing I'd like to point out is that Web 2 was completely built on this concept. Um so the entire web as we see it today is largely funded by ads. Um but how did those ads get there? Well, first it started by Google producing a search engine that everybody wanted to use and then they added ads into their system and like after year one they generated $70 million of revenue back in the early 2000s. Um they continued uh basically cycling through that model and and showing more ads to more users as they attracted them with the search engine. But eventually they reached a point where they actually got stuck. There was kind of a plateau

effect and so what they started to look at is how do we gain distribution of the search engine? When they looked at it in that sort of way, they started going, "Hey, there's value in us doing a partnership with Mozilla." Um and then also they looked at building their own browser themselves. So uh Mozilla is a free software. Anybody can download it. Nobody's paying for Mozilla in the same way that Chrome is a free software as well. Um and then this same concept continued on to Android as well. Android's not an operating system that we pay for. It's something that comes free with the uh devices that you're utilizing. And so essentially what's happening is this is creating this very large funnel that funnels people into their ads network. So, Android operating system has something

like 3.5, 3.6 billion different people that are utilizing it, and the defaults that are set is that you're going to utilize Google Chrome. And then Google Chrome defaults to Google Search, and then Google Search turns on ads. So, you've also got Mozilla defaulting to Google Search, and that turns on ads. And same thing with Apple. So, like essentially what's happening is this one key product is funding the entire web, and it doesn't just stop there. It also goes to like news sites. So, anybody remember BuzzFeed and how they had a ton of ads on them? They were completely funded by ad. You ever been annoyed by YouTube?

Yeah, that's all the ad revenue that they're generating as well. So, like one key thing that we do right can essentially become an economic flywheel that just kick-starts the entire engine for everybody throughout the ecosystem. Um some other things that I highlighted within the blog post as well that I think is worth considering is like there is an open core strategy, too. This is worked like auxiliary to the actual web itself, but like you look at Linux. Nobody's actually paying for Linux, but there's tons of developers who work on Linux full-time. Often times they're doing it because the companies that they work on derive value from Linux actually being capable. So, Microsoft pays engineers. There's entire

companies like Red Hat that are built on the the service models that are provided by Linux. There's conferences that occur, and there's like training modules and all of these different things that occur because of this open source software that's that's been like created. And so, essentially, like you can see these same sorts of like concepts applied in many different ways. Similarly, you can look at MetaMask. I think this is kind of the first touchpoint that I realized of how it applies to us is that swaps effectively within built directly within a wallet are funding an entire team of wallet engineers. So, if you look at MetaMask, they've generated something like $300 in revenue just based upon swap uh value alone. So, like that's a lot of money.

Um and that's just their swaps. That's not the bridging. That's not the new um uh perps trading that they have. But like you look at every sort of like thing that we're doing today, this is I think where we're kind of like fundamentally missing it is that our user and our primary user is a trader. A trader is going to essentially churn themselves out of the network unless they're making money, and most people lose money when they're trading cryptocurrencies. So, eventually you've got a life cycle of it. You've burned through all your users. What do you do next? Who's your sustainable user that you keep around?

So, that's essentially what I wanted to to cover here is that like we don't have this like aligned incentives. Our our primary like customer or user within the space is somebody who's um just designed to extract value or um bring value into the ecosystem and then have it extracted from them by others. Uh that's not a sustainable revenue model in the same way that like a bank operates. Like most people don't walk into a bank, sign up for a bank account, and then lose all their money, and 6 months later leave their bank.

Um but for some reason that's what we do with their wallets. Um but there is some some other aspects that come into play with this. So, um the growth was dependent on factors uh outside of the founder's control. Um so, essentially what happens is that like the distribution that occurs from this uh from the wallets and people showing up, um often times the incentives aren't aligned with uh those sorts of like goals that occur. So, um yeah, essentially where I'm going with this is the idea that like if a wallet is designed to make it easier for somebody to trade, but then they're uh being incentivized by VCs to like hype

the token that's being launched or a protocol is being hype incentivized to hype a token that they're launched, then that's actually reincentivizing the wrong problems that we shouldn't be thinking about in the first place. Um similarly so, I think everybody's experiences this. We've seen this right now, too. Public good fragility is a real thing. Um infrastructure starves when markets dip. So, you look at like what's happening with the EF right now. A lot of it comes down to the fact that Ethereum value is dropping. There isn't a sustainable model for the Ethereum Foundation. They've accepted that.

That's all right, but who funds the projects that happen after they're gone? That's the question that we're faced with now. So, this is the second part. Um one of the things that I've realized as I've worked on wallets for quite a while is that wallets act as the distribution hub. We are kind of the central point in the same way that a search engine is the distribution hub to the web 2. Um and so with that, um the wallet acts as a way for you to be able to access and and utilize the different protocols that occur. So, you want privacy, um it's much easier if you have it built directly into your wallet rather than having to show up at a website um and try and figure out, "Hey, has it worked?" uh "Oh, wait, I need a different wallet. Wait, my money is on a

different uh network. to go swap. Wait, there's different sort of trading things." Like that was an actual experience I just had when I wanted to go donate $100 to Roman Storms fund. I did a UX just like analysis on myself. It took me 34 minutes to onboard myself and actually domination and I had to go through eight different sites, generate two different onboard on-ramping events and I had to do multiple transactions for it. Like we've essentially built a model where all of the protocols exist in from the web's perspective and from a browser's perspective in untrusted land, but an extension is considered trusted.

Um, and so this is kind of the way that I've been thinking about it is the wallet is the loss leader as it exists today. It's a piece of free software that we can utilize. It's already extracting some level of basis points and I do think we should get better about what we're doing in order to be more competitive, but I'll cover a little bit more about that in a few slides. But essentially the wallet becomes that distribution engine that you can then build on top of and generate more volume through.

So how do we balance it? And like what does this look like? I don't know show hands here who's actually built a protocol and try to build a snap a metamask snap. Has anybody done this? One two. I'd venture to guess you guys didn't have a good time with it. Um, I I've spoken with many people about this as somebody who's looked at it and understands the underlying technology of what they're trying to do. I think they've actually built a model that aligned with the web paradigm much better than we realize. It's just not been done in the right way. So one of the things that we're actually looking at within brave wallet is how can we reutilize the snaps in order to improve

the security and make it easier for protocols to be able to access this. Um, so the idea here is that like as a wallet we don't want to have to go and build an integration for every single protocol in order for them to get distribution, but what we can do is we can build a defined set of interfaces so that any wallet protocol can plug in and just generate the transactions on behalf directly inside of the wallet. And so that's essentially what a snap is doing. Um there are security trade-offs that that come into play. So as soon as you build in the extension model of any sort of paradigm, this happens within a browser as well.

Essentially what happens is the extension becomes trusted in a way that could be manipulated. So if you download a malicious extension, that can become a problem. But what we're doing is we're actually solving the adoption struggle for the protocol developers. No longer are they having to come to wallets and do a BD sale where they're doing a rev share split each time in order to try and get the protocol adoption and and grow the liquidity. Instead, we're just saying, "Hey, go and build it, show it works, and if it works, we'll continue to market it to our users and continue to grow it on your behalf." Um And so this is the thing about it is like there was a talk or a blog post a

few years ago called like the fat wallet thesis. And essentially what I've realized is that's already here. The protocols are already coming to us. They're coming and asking us to add their protocols to get the distribution and then to align it with their goals. Essentially we're doing the rev share splits. That's how the wallet industry is being funded. And so this model is actually working in such a way that like we should look to figure out how we should re-align the architecture of Web3 to make this work a little bit easier. It doesn't mean we need to get rid of window.ethereum. It just means we need to refine how snaps are working in order to make that a little bit better and to make it easier

for these protocols to come into place. Uh so the nice thing about this too is that like no longer do you show up at 14 different sites in order to complete the actual one journey that you're wanting to do, it all just gets abstracted into the wallet. All of a sudden you're able to have one snap talk to another snap, and your user journey just becomes hey, I want to send money from to this address, and I want to utilize these tokens here, and I can figure out what their token is and what they accept. Uh maybe they publish it on an ENS um record of like what their preferred token is. And I just say send money to ENS, and under the hood my wallet's automatically doing a swap, a bridge, it's running it through privacy

protocol and uh as a user, I don't really need to care about that because all I cared about was the intent itself. Um similarly, so it makes it easier from a wallet perspective to simplify what we're displaying to the users. We don't need to take some hex data and try and figure out how to parse some data that's uh for some new protocol that we've never even seen before, uh and display it to the user what might happen with that. Uh this is like half of the problem that we have as wallets, and and one of the things that drainer malware is actually exploiting is the fact that many times a wallet doesn't understand the actual transaction transaction data it's trying to display to the user because it's a new protocol

or because they're doing something esoteric like uh like basically burying call data inside of a multi-call, um you know, 40 calls deep, and then all of a sudden the wallet is expecting to parse this in a very different way. Um so, the nice thing about these snaps is it also shifts the direction of kind of the security that we get. Um and so, we get essentially a minimalist architecture from this, and we realign with with how the web works in in that direction. Again, we're going back to the the site itself is considered an untrusted process. As a as a browser, we literally give it its own separate process. It gets its own separate memory. We don't let that memory talk to somebody else's

memory, and we have a separate browser process. Same things with extension is it's treated in a higher fashion because the user is is downloading it. That's essentially what we're trying to do um in order to achieve that. And because of this, we can actually probably reduce the supply chain because you don't need all of the UI capabilities. They're already built directly into the wallet itself. This is just a part of like the MetaMask design that they've come up with. But, there are still things that need to be improved, and this is why I asked that. Um snap extension stores are still going to be part of a problem.

Um you know, the uh wallet needs to create a snap extension store, and it needs to be actively marketed. Um this helps the protocols in some ways, but it hurts them in others. If there's, for example, some reason it needs to be taken down because it's not certain if they're malicious or not, these are the sorts of like dirty trade-offs that we have to figure out still. Um but ultimately, where I'm trying to go with this is that like essentially, if we rework these business models and we make it easier to get more users on board, we can actually generate more transaction volume. And the larger amount of transaction volume that we can generate, the more revenue we can generate. And that means we can have more engineers that can actually afford

uh to pay a salary and and be able to continue do it. What we do with those profits is the bigger question. I think this is what the EF is is trying to signal to all of us is to say, the the goal is don't make profit. Uh instead, it's focus on the crops. And there is an aspect that's useful to that, but there is a uh life cycle that will eventually die because of it. Um and so, this is part of the reason that I point out the Friedman doctrine is uh essentially, you look at the economics of today and this idea of like monopolies, a A of it comes from the concept of what he is promoting and I say it's time for us to challenge these things when we're building out this new web.

So, if you want to read through any of these blog posts, the first one is mine. There's other people who are talking about these ideas, too, and challenging my theories or presenting new ones. I highly recommend people take a look at these as well because ultimately we're not going to be here in 10 years time if we can't find a sustainable way to to do this or we're going to end up back in the Google model of don't be evil turns into oh well, we'll just collect all your data.