Crypto First Search Engine or Lecurate Environment?

Live from Breakpoint 2021, Brendan Eich sits down with Anatoly Yakovenko to discuss integrating Solana into the Brave Browser, the huge potential for a decentralized search engine and NFTs as an entry point to the metaverse.

00:09 - Intro

00:54 - Integrating Solana in Brave

08:00 - Challenges with creating the browser

09:23 - How to scale crypto to the general public

11:57 - A Decentralized search engine

14:46 - NFTs as entry point to the metaverse

16:35 - Mobile vs. Desktop

18:00 - Languages and smart contract development

20:40 - How to grow crypto to mass adoption

22:44 - Global Peer-to-Peer environment in Crypto

DISCLAIMER

The information on this podcast is provided for educational, informational, and entertainment purposes only, without any express or implied warranty of any kind, including warranties of accuracy, completeness, or fitness for any particular purpose.

The information contained in or provided from or through this podcast is not intended to be and does not constitute financial advice, investment advice, trading advice, or any other advice.

The information on this podcast is general in nature and is not specific to you, the user or anyone else. You should not make any decision, financial, investment, trading or otherwise, based on any of the information presented on this podcast without undertaking independent due diligence and consultation with a professional broker or financial advisor.

Brendan (00:10):

Great conference.

Anatoly (00:11):

I know. Thank you. I'm really excited to be on stage here with you. You're-

Brendan (00:15):

Same, [crosstalk 00:00:17].

Anatoly (00:16):

... One of my heroes. As a programmer, JavaScript is a language that really revolutionized how we do application development, how we build. It's the foundation of the web. And I often think of web 3.0 really just being the web, just part of the bigger web.

Brendan (00:34):

Yeah, me too. That's how the web grows, by evolution. So we think the web 3.0 browser should be the gateway to a billion crypto users. And we are therefore integrating Solana into Brave soon as we can. And here's the cool thing, this is an evolutionary path. We're going to make it so any dapp that is Solana enabled, wherever other chains, EVM compatible or Ethereum, whatever it supports, if it supports Solana as well, we'll make it use Solana by default. So dapp builders who build for Solana as well as other chains. In Brave it's going to use Solana. And that's going to just help, I think, pull all the dapps on the Solana.

Anatoly (01:24):

Super exciting.

Brendan (01:25):

You like?

Anatoly (01:25):

Yeah, it's wonderful. Yeah.

Brendan (01:28):

Let's see what else. What do we like about Solana? We like NFT games, we like DeFi a lot. We want to make it easy for users to earn and get yield without having to be super expert or do a lot of complex operations. So we're going to work on building that probably in the first half of next year into the wallet so that you can just robo-earn, robo-yield. And we want NFT galleries and NFT transactions to be super slick. I was inspired by the Jules Urbach talk earlier today, and the demo earlier here with NFTs, there were several of them actually, it's all good.

Brendan (02:07):

We want as many NFT marketplaces integrated as we can, so that's on the agenda. And yeah, [Radium 00:02:13] is there, of course. Radium's still earning, yielding good. The thing that we do now with basic attention token tends to have to settle on Ethereum and it's going to cost you gas. And our valued settlement partners like Gemini, Uphold bitFlyer in Japan, but once we're on Solana, I suspect that BAT, which is already reflected through Wormhole, proxied through Wormhole, might just find it's better to settle on Solana. What do you think?

Anatoly (02:41):

Yeah, for sure. Absolutely.

Brendan (02:44):

I'm giving you the softballs here. And we really do want to get this out to all users. We think, whether you're having a hard time in some part of the world where it's hard to get banks to let you save or borrow, or you're beyond banks like a lot of us are or want to be, Solana is the way to do it. And I mentioned auto earn already, got ahead of myself, but I think this is going to be huge. It takes some skill, you got to make sure if you get on the wrong side of yield farming, you go somewhere where the grass is greener, but we'll make it as automatic and easy as possible. And it's just so much better on Solana. I'm making you blush. And yeah, the dapp ecosystem is growing, but if we do this Solana default on multi chain dapps, I think we'll just pull every dapps that's really popular over that Brave users want, and I hope that's going to be every dapp.

Brendan (03:37):

So here's more NFT marketplaces. There are lots of cool projects in crypto, so we're not doing only Solana, we have obviously Ethereum, we're going to do Bitcoin in the new wallet. It's coming up fast, it's in the Brave nightly builds. And we might do other chains, but I think it's important to pick a chain as default. This is a lesson we learned the hard way with search engines, because when you make a search engine the default, first of all, you can get paid if you get a deal, not always true. And really the user expects to just type keywords into the address bar and search. We want the wallet to have a fast, good default and that's Solana. So enough said. And we're bringing it to mobile too. This is important. I think a lot of fragmentation has occurred due to how wallets are split across mobile and desktop. We're seeing some good mobile first or mobile also wallets. We want to do it mobile and desktop feature parody, evolve at the same time. And we're happy to do that with Solana's partner.

Brendan (04:42):

So the last bit of news is the BAT system is a triangular system that involves privacy preserving ads. And users opt into it to get 70% of the gross revenue. What we've built so far has a part of our BAT ad system requiring us to verify things, to be the trust third party, which is a security hole. And so we started a project called Themus and worked with several crypto projects to see if we could bring it to high speed chains that can do things, like you need smart contract systems for zero knowledge proofs, you need some part of it in the browser because you're measuring attention. You don't want to put your detailed attention log on any blockchain, however fast, because it'll fingerprint you. So we're using black box accumulators in the browser with Themus and we're then minting ZK proofs. And the cool thing about Solana is we can just put those on-chain, no aggregator, no trusted third party. So we're getting rid of ourselves, we're firing ourselves as a trusted third party. And that's something we're excited about.

Anatoly (05:40):

And that's awesome. That was, feels like two years of research. It took quite a while to get to that design.

Brendan (05:47):

And now it's going fast. I think now we've got good working relations with Solana and we can crank out the Rust Co, because we love Rust. Because I was executive sponsor of Rust at Mozilla, so I have a tear in my eye to see my little babies all grown up. And Amazon's hired a bunch of the Rust core team. It's okay, they need jobs. But yeah, we want BAT to be fast, low fee, DeFi base pair and for ads on Solana. So Brave and Solana are doing the new crypto and ad system and it's going to be awesome. Thanks.

Anatoly (06:24):

That's awesome. I'm a huge fan of the web, huge fan of all the work that you guys have done and Brave. And I remember pre-mobile days, I was working on Brew and I was trying to optimized the web and flip phones. And there was a brief moment where the iPhone came out, we had a browser, and it felt like the web has opened up. And then it just got away from us.

Brendan (06:49):

That's right. Jobs said when he did the iPhone one, he said, "The web finally works on a phone." And then the story I heard from somebody who would is that they had to port a bunch of games which were C++ or whatever, and they had to do native apps. And they never looked back after that. But I think the web can always catch up and should catch up. And web 3.0, if you have this evolutionary path with dapps and dapp triggers from webpages, then you just evolve into it.

Anatoly (07:19):

Yeah, that to me is the really exciting part, is there's now an opportunity to have cryptography power the next generation, how web is monetized. Whether it's through advertisement, like with zero knowledge proofs or through direct payments and micro payments. Do you feel like Apple's going to crush us?

Brendan (07:41):

People a few years ago were worried about this Facebook thing, Libra and now DM. And they got crushed because some politicians hate them. But Apple is very cautious, and if they're doing anything with blockchains, it's a ways out. And then when they arrive in, it's going to be the diva at the party at midnight, like, "Start the party now," and the booze has already run out. So we're going to drink all the booze first.

Anatoly (08:06):

All right. I'm down for that. What are some of the challenges with building a browser for general consumers, but also with cryptography?

Brendan (08:17):

This is the problem with browsers is they are universal apps. You spend a lot of your digital life or online life in them. And so if you make the crypto stuff be this expert only area, or it's scary. I use wallet apps, I use ledger hardware wallets, but it's a little bit scary because you feel like, "Did I forget my pin in or did I have to reset it and do the word list?" And there's some anxiety and fear of loss. We want to make crypto be a positive sum, that's why the robo-earn is important to us. Just like with BAT private ads, you could get 70% of the revenue.

Brendan (08:53):

So you're always building up your assets as well as spending or sending them. And it should be slick, it should be for e-commerce. You can even do things like dis intermediate Amazon. I won't give away all my secrets, but we think we can do that without having a bunch of JavaScript user scripts attack every merchant checkout flow. We think there's a way to get into the interchange charge and do it. And crypto everywhere. It should be slick, should be easy, should be comfortable, make you feel like you're going to win, not lose.

Anatoly (09:23):

What about custody and keys? How do I get my parents to understand this stuff?

Brendan (09:28):

Yeah, it's really a little different, but we're looking at Taurus, we're looking at various ideas for backing up your keys that don't just put it on paper and word list in the safe, which we've all been through. And in some ways, the old web went with username and password and had to add a second factor, which often had to be a temporary access number generator on your phone. So at that point you're almost as complex as self custody. I would say you just have this more conventional recovery path. You lose your phone, you know your email, you can try to prove that you're the same person to Coinbase or whatever. But I think self custody has a complimentary role and we want both. We want people to use self custody and be comfortable with it, so we're looking at all these usability challenges. And we think we can get it just almost as good. And then unfortunately the regulators insist, if you want to do Fiat on/off, you're going to go through a custodian.

Anatoly (10:20):

Of course. The challenges, that's the exciting part. No one has figured this out yet and we're going to dive right in and see, how can we actually scale crypto to the general public?

Brendan (10:31):

Make it easy for your parents.

Anatoly (10:32):

Yeah. Yeah, would love to see it. What do you guys see as the tension between the app store on the mobile device and the mobile web?

Brendan (10:42):

Discoverability is always a problem. And we don't want these brutal curators like Apple. So having lots of stores is good, but then you have the need for a search engine, which Brave now has, which is a private engine and also involve users opting into building the index incrementally, that's the web discovery project. So we're going to aim, because we're very crypto first and our ad sales teams, one of who's here, always looks at crypto options and NFT options, we're going to aim at making our search engine best for crypto. It already uses [inaudible 00:11:14] charting, and it's still in beta, but we're working out all the kinks, so I think search, the good old search we remember from 2004 when Google was great needs to come back and it needs to be the way you find stuff in marketplaces and crypto exchanges.

Anatoly (11:29):

That's awesome. What kind of information do you think users would want out of a crypto first search engine or curated environment that's different from the traditional web?

Brendan (11:39):

Search almost gets into, is somebody trying to SEO you and compete for keywords? We're aware of this problem and there's no silver bullet. But we think with crypto, you might actually have a better chance at mechanizing this and having a fair playing field, an automated system for finding the lowest fees and the best yields.

Anatoly (11:57):

Is there hope for a decentralized search engine?

Brendan (12:01):

Yeah. So I had a friend who was involved with pre-research, Rich Scrantom, and pre-research looks like it's running a bunch of nodes [inaudible 00:12:07] Google, which Google does not like. And if they're running on [inaudible 00:12:10] IPs, Google's going to shut them down or use their anti-bot team to take them out. We're building a legitimate search engine, but we can't decentralize the algorithm easily because search is sharing queries, looking for some kind of objective best results like page rank, the eigenvalues of the random walk. And decentralizing that is a research problem as far as I know. But we have an active team, we're evolving search and we need your help because we're trying to crowdsource the incremental indexing of the web, we're not trying to index everything from 1998 on. Only Google can do that. Hats off to them, but their time is passing.

Anatoly (12:49):

When I was growing up as an engineer, the web was just starting, I was really passionate about Linux. And I had this dream of a Microsoft-free personal computer. It feels like the web 3.0 is potentially a dream of ad exchange free, that parasitic Google free web. Is that possible?

Brendan (13:13):

If you don't collect the data you won't go wrong that way. There's still other ways that central powers can turn on their users and take advantage of them. But I think there is, and that means ultimately you might need hardware that's indie or that's user first. And Brave's not capitalized to do this yet, but I know people, including friends from Firefox OS, which actually after it folded at Mozilla, continued in [inaudible 00:13:37] OS. And there's an open source lineage that you can trace back. And people at Qualcomm, we both know-

Anatoly (13:42):

Of course, yeah.

Brendan (13:42):

... We are working on it at the time. So I think there's a chance for a new open source OS that has web 3.0 and none of this Java or swift native stuff. And JavaScript, web 3.0 All the way down.

Anatoly (13:55):

Are we going to end up building a phone?

Brendan (13:57):

Brave OS. I don't know, I'd have to raise some more capital.

Anatoly (14:03):

Yeah. Yeah, that's a way to nerd snipe me for a couple years.

Brendan (14:07):

But people need independent hardware that serves their interest first. Absolutely.

Anatoly (14:10):

For sure. It always feels like that's a really tough challenge. But every two it gets easier and easier, hardware gets cheaper and cheaper and the tools get better and better.

Brendan (14:19):

And then Apple has something new and shiny that the commodity hardware can't match for another year or two, but that's just the nature of the game. So I'm sure we'll have iPhones, but we can probably have BAT phones too. Solana phones.

Anatoly (14:33):

The BAT phone. I love that. The BAT phone sounds really cool. As you guys see the web 3.0 evolving, I think from your presentation, NFTs were such a huge focus as well. Do you think this is the entry point for the Metaverse as people call it or that really interactive rich environment with ownership of the stuff around you?

Brendan (14:56):

Yeah. I think you have to keep running at these problems. And usually if you're a startup and the timing isn't right, or something goes wrong, you run out of capital and then the investors reset, or maybe they try again. With crypto, we have this great ability to just keep leveling up. So we're seeing Bitcoin, now we're seeing smart contracts on Ethereum, now we're seeing Solana. And as you level up, you can start to do some of these things that seemed hard before. Like you want some kind of cryptographic proof of ownership.

Brendan (15:26):

I think one of the demos talked about this. You want to make sure that somebody doesn't copy the pixels. And if you get into VR, there's been interesting research on this. And my friends at [inaudible 00:15:36] have done some work on this. You can actually watermark in a way that's indelible. And if somebody copies your art and tries to remove the watermark, they degrade the quality, because it's been convolved with the luminance and the chrominance. So I have hopes for this being useful in games and connected verses. And to me, that's the Metaverse, it's not going to be something centrally planned at Menlo park by Lieutenant commander data.

Anatoly (16:02):

I hope not. What I see out of the gaming companies that we talk to is that, especially the ones that are crypto focused, is the one to build browser first games. Everyone that I talked to had this idea that as soon as you open the page, you jump right into the game. There's no sign up, there's no friction, your wallet is your identity. And you're just exactly where you left off.

Brendan (16:24):

That took a lot of work at Mozilla, by the way. We did [inaudible 00:16:27] JS and that led to web assembly. And you could show games, in the story, you can start playing them and then you just convert. I think it's a great model.

Anatoly (16:34):

Do you feel like mobile is expressive enough for that? Or is the difference between iOS and Android and desktop is too hard to actually make that work?

Brendan (16:45):

There's certainly a difference. Even with the latest chip sets, you're just not as fast, you have less bandwidth all around. But games can scale down because the view port's smaller, there's hope that you can use the kind of tricks that we see with the remote rendering, cloud rendering. So I think mobile is the future, but I heard this 12 years ago, people would say around Silicon valley, mobile's the future. And then they would say, "That means there's no desktop." And that is very false. Everybody with a laptop or any big enough screen and a keyboard is still very high value. And that means the economics there don't go away, it just doesn't grow as fast.

Anatoly (17:19):

That's true. If you look at the growth of the Solana ecosystem, a lot of the users are basically dust up only.

Brendan (17:27):

Yep.

Anatoly (17:27):

That to me says that a lot of folks, maybe there was a switch during COVID where we went from being so much immobile to where we're staring at screens again.

Brendan (17:36):

A bit of that. You go to India and a lot of people are mobile only, but you need both. And I think as mobile gets stronger, you're just going to see more parody, you won't see this need for apps, which is often artificial. It's like holding the browser back, sandbagging Safari a little bit. This is what my friends at Google, or one of them who went to Microsoft, always accuse Apple of, and it's not wrong. You got to give the browser it's due and then it can compete with native better.

Anatoly (18:00):

Got to ask you about languages.

Brendan (18:03):

Okay, [inaudible 00:18:04].

Anatoly (18:03):

How do you see smart contract development in the future as somebody that had incredible depth and understanding how application development happens on the web?

Brendan (18:12):

Yeah, I think the thing you're seeing with type script, especially with large teams, is more information that you need some kind of proof system or it could be just a warning system, but it's based on model checking. Often it could be based on higher level models than you can express in sound type system, which is something where there's just this timeless world of types that's potentially syntactically checked and prevents bad things from happening at runtime. You need dynamic systems, dynamic code, JavaScript, and the static checkers.

Brendan (18:44):

And you get the best of both worlds if you have really good ones. So I remember at Mozilla, we were investing in model checkers for C++ because it's memory unsafe. And you could build these higher level checks that knew about security properties you wanted to enforce. And I think this is what you're seeing with smart contracts. I was talking to somebody I met at the hotel bar about this, because it's still a very fruitful area that's had good research in computer science, programming language theory. And it hasn't always been brought to the programming masses like it should. There were companies like [inaudible 00:19:17] Covarity and others like that. The compilers themselves grew the ability to do plugins for static analysis. And now [LOVM 00:19:26] is there.

Anatoly (19:27):

Do you think that smart contract development needs to have a high level, easy to use language environment? Or can it be driver code?

Brendan (19:37):

Yeah, exactly. Driver code in the era of C was the worst code in the kernel.

Anatoly (19:42):

Driver code with Rust is a little bit less frightening.

Brendan (19:45):

In fact, a friend of mine who was at Microsoft at the time went to Mozilla and has his own startup now, did it at Microsoft, a checker for driver's C code. Which he could skirt the halting problem and kind of statically reason about it and say, "This is garbage driver code, send it back to the vendor." But yeah, I think you don't want to have happy, fun, JavaScript looseness if there's big money at stake. So I think it's important to have the right tools with the right static and dynamic checking.

Anatoly (20:13):

Do you think smart contract development is strictly financial or are we going to see things that are not financial that you can actually [crosstalk 00:20:21]?

Brendan (20:20):

You'll see things that are not obviously financial, but they'll turn into reputation in a game or gifting and those tend to matter too. So you still don't want too many dynamic errors.

Anatoly (20:32):

That's true.

Brendan (20:33):

So I talked about this in my chapter in coders org, I'm still a fan of static, even if it's unsound semi-static checking.

Anatoly (20:40):

What do you guys see as like the opportunity for us to grow crypto to a hundred million users, actual signers?

Brendan (20:49):

Yeah, I'd to get Brave to that scale in a year or two. It depends on everybody here and others. It also, I hate to say it, depends on the nation states of the world not doing something adversarial. But I think given the state of the world, not a great state, but there will always be options to do things with crypto. The internet routes around censorship, and that's true in the web 2.0 And the web 3.0 world. And it's true with blockchains. You still have concerns you have to fork to undo the censorship, but at least you have options. Do

Anatoly (21:26):

What kind of applications do you envision will actually drive that growth?

Brendan (21:30):

I think at first it's going to be people using crypto for payments and for DeFi. And some leading edge of that user base will be getting more sophisticated in doing other things. But just having things like gift cards, where we often find that they're useless points, even if we can use them or Congress passed the law to don't expire, we still just don't use them. We should have much more liquidity. We should have liquidity across all kinds of assets. And this is where you start talking about tokenized securities, and can you have primary and secondary liquidity for companies? I think if you're as old me, you all had a tiny piece of some startup that went sideways for 10 years and then sold. And you couldn't trade it easily. And you might have wanted to do that because you might have been squeezed out when it sold. So there's lots of room for blockchains to solve these problems. I think in general, connecting people more directly getting rid of these officious or censorious intermediaries. A lot of room for application.

Anatoly (22:29):

In this new evolution of the web, I often describe crypto as a fully connected network, as opposed to a social graph, like on Facebook.

Brendan (22:40):

Yes.

Anatoly (22:40):

Do you think that's true? Do you think we're going to enter a stage where I am effectively with my cryptographic signatures, I'm in this true global peer to peer environment?

Brendan (22:50):

I hope so. I showed at web summit last week, I showed the slide with the correct diagram, which is more like a mesh for decentralized, and the incorrect one, which sometimes is called decentralized, which is really distributed, but it's mostly tree structured. Or if it's a graph, it has a dominating spanning tree. That's Google, that's Amazon. So with projects like Helium, with web RTC making it so you can make connections into the endpoints instead of only out. In the old days in the nineties, we could only make TCP connections out from the browser. I think we're heading toward this world. We have to build it iteratively and collaboratively, we have to get around the concrete firewall problems that web RTC mostly got around, it's still a little dodgy. And I think that is the future. I think we should all have Helium nodes if we can. I'm a fan of the project.

Anatoly (23:38):

That's awesome. The idea of decentralized browsing on an open source phone connected via an open network.

Brendan (23:49):

Low raw radio.

Anatoly (23:50):

Yeah, run by the people. Accessing Solana, that would blow my mind.

Brendan (23:55):

It sounds too good to be true, but I think it could be true, especially if we build it carefully and quickly enough and get it out there and make it usable, which is why I've always wanted to make Brave be about crypto. Even when we started using Bitcoin for our prototype, it was clear once you shield the user by blocking all those trackers, you break all the economics that pays advertising money into the publishers after taking a big slice out for the middlemen like Google. And if you cut that out, how are you going to reconnect it? It's crypto, peer to peer.

Anatoly (24:26):

All right, let's do it.

Brendan (24:28):

Awesome.

Anatoly (24:28):

I'm excited. So thank you, Brendan. Thank you so much for doing here, for working with us.

Brendan (24:34):

Thanks.