#8 - The Solana Phone
[00:00:00] Balaji: Welcome to the Network State Podcast. I’m here today with Anatoly Yakovenko, the co founder of Solana. Today we’re talking about why the decentralized crypto phone is central to both digital civil rights and digital property rights. To understand the crypto phone, you need to understand the concept of private keys.
[00:00:16] Balaji: These are the keys to your crypto wallet. A crypto phone is unlocked with those keys just like a normal phone is unlocked with your face ID or passcode. Because these private keys are kept local, the crypto phone is built around protecting digital civil rights and digital property rights to a much greater extent than a typical phone.
[00:00:33] Balaji: It protects digital civil rights because it allows you to send provably end to end encrypted messages and protects your account from centralized search and seizure. But the cryptophone also protects digital property rights by preventing centralized services from freezing transactions or seizing funds.
[00:00:48] Balaji: And that’s why the cryptophone is important. Protection of digital civil rights and digital property rights. Anatoly is with me today to talk about Solana’s efforts on their phone, and how as a former Qualcomm engineer his experience in low level [00:01:00] hardware optimization was actually very helpful in building it.
[00:01:02] Balaji: We talk about some of the nitty gritty technical details in building the phone, and then go all the way up to the big picture of the society we’re building, namely how the cryptophone is perhaps the crucial enabling technology for the sovereign individual, and thereby a crucial enabler of the sovereign collective, what we call the network state.
[00:01:18] Balaji: Let’s get started. Welcome to the Network State Podcast, Anatoly. Thank you. Should I say Toli? Go ahead.
[00:01:23] Toly: Toli Wurtz. Thank you for having
[00:01:25] Balaji: me. You know, we’ve been, I think, friendly for a few years now. I’ve seen the rise of Solana from very humble beginnings. I actually knew Raj. All the way back in the day when he was at Omada Health, and I was a board member there.
[00:01:39] Balaji: And, uh, you have gotten to know more recently, and you were formerly a Qualcomm engineer before a relatively unusual late in career migration to crypto. What I thought we’d talk about today is, you know, the Solana ecosystem in general, but specifically why the Solana phone, a decentralized crypto phone, is central to human freedom.
[00:01:58] Balaji: Uh, but maybe you want to, [00:02:00] you want to just give the, for, I, most people know what Solana is at this point, but we want to give the background yourself, background on Solana and so on, just bring us up to the
[00:02:06] Toly: present moment. Sure. So the, my origin story, you know, I’m a, went to school in Illinois. I was a nerd, loved computer science, loved performance optimization.
[00:02:19] Toly: I don’t know, for whatever reason, I like mentally got into the, the love of like looking at code and trying to make it run as fast as possible on a piece of hardware, like optimizing assembly, all that stuff. And I ended up working on kind of that part of Qualcomm, like looking at like various parts of the stack and like doing optimizations.
[00:02:39] Toly: And I was there from the start of the mobile revolution, basically 2003 for about, I can’t remember for sure, but like 12 or 13 years. So I went from like these two megabyte, two inch phone flip phone devices. So, like,
[00:02:53] Balaji: 2005 ish, pre iPhone. Yeah. Yeah.
[00:02:55] Toly: Yeah. Way, way before iPhone. Um, and like, I saw the [00:03:00] iPhone revolution.
[00:03:00] Toly: I saw the feature phone market drop from 80 percent to 20%. I saw the hardware improve every year by more than a factor of 2X, which is pretty cool. Um, and, uh, 2017 and like I’ve been, I was aware of crypto, I think since the origin, I thought Bitcoin was kind of neat, but I never understood the social impact of it.
[00:03:20] Toly: Even in retrospect, it seems so obvious, 2008 financial crisis, this sound money thing comes out, it just like didn’t click. It seemed like, oh, this is kind of neat, but would never scale kind of thing. A lot of engineers get very skeptical about like. This is going to change the world. And like miss the whole social implication of it.
[00:03:41] Toly: In 2017 is when I got into crypto. I saw fees on Bitcoin hit like 60 bucks a transaction. And, um, I literally had like a Eureka moment that there’s a way to encode passage of time as data. And that was the foundation of. Proof of history, but like that, I [00:04:00] didn’t know what to Google for. This is kind of funny.
[00:04:02] Toly: Like the reason why Solana started is because I couldn’t, I didn’t know this was a verifiable delay function. There’s a bunch of folks like Dan Bonet working in these things at Stanford. And I didn’t know what to Google for. And I thought, Oh man, I found something really, really cool that no one has thought of.
[00:04:16] Toly: And it has this unique property where I can take this data structure and I think I can apply it to build a source of time and create time division, multiple access, which is how. Old school 2G cellular networks, like what I started with. Before
[00:04:31] Balaji: CDMA. Yeah. Before CDMA. It was all CDMA before
[00:04:33] Toly: CDMA. That’s right.
[00:04:34] Toly: Yep. Yeah, so I, and that’s like the fundamental way you scale bandwidth, uh, in a shared channel on radio is you split everything by time and then you get. You can utilize basically almost all the bandwidth available to you. And once you have that, you have a scalable blockchain. So that was like the Eureka moment.
[00:04:52] Toly: I didn’t know what to Google for. Uh, my wife basically said, you better go start this and do this for real or stop [00:05:00] talking about it.
[00:05:02] Balaji: That’s great that she encouraged you though. That’s
[00:05:03] Toly: really great. Yeah. Uh, and that’s when I met Raj and, uh, he was like kind of the perfect co founder because he was not an engineer, he’s raised money before.
[00:05:13] Toly: He kind of understood a large part of tech in Silicon Valley that was really new to me. His quote is that like, he was like, okay, I’ll help you for six months and then, you know, off you go. But five years later, yeah, five years later, we’re still, uh, working and building.
[00:05:28] Balaji: The sort of low level optimization stuff that you’re talking about, you know, you know, Duff’s device.
[00:05:32] Balaji: I’m sure you know Duff’s device, right? That type of stuff, right? Where, in theory, you can just write a for loop and, you know, the compiler will take care of it. But in practice, you might need to actually manually unroll the loop and you have to have a familiarity with the underlying. Yeah. You know, it’s interesting because, you know, a couple of just remarks on what you said.
[00:05:48] Balaji: One is, you know, Feynman, I think he used to say that he didn’t like to read the literature before he attacked a problem. He only did that afterwards if he ever did it because then he would try to take an original [00:06:00] approach from first principles, which is sort of what happened with you in proof of history.
[00:06:03] Balaji: And now obviously there’s an intersection with VDF and you can pull some ideas from that literature, but you, you came at it from an original way of thinking of, you know, essentially VDF says an alternative to proof of work or proof of stake,
[00:06:14] Toly: right? Yeah, I felt. Like, I got lucky in solving a hard computer science problem.
[00:06:19] Toly: You can, you, you basically, as an engineer, you never want to feel like, okay, this is like . I’m gonna solve this really hard computer science problem. You’re very likely to fail. And what I felt was like, oh man, I think I got lucky. And they have like a solution to something really clever, to a very hard computer science problem.
[00:06:36] Toly: So let’s see, like how far we can take it and how far we can scale it. Um,
[00:06:41] Balaji: and one way of thinking about this, and you tell me if I’m off base, but like. The way I sort of model it is that part of how Solana scales is because you assume that nodes can have heavy compute, whereas, uh, you know, Bitcoin, for example, assumes like a one meg block size [00:07:00] constraint is a hard constraint system.
[00:07:01] Balaji: The whole Bitcoin civil war is fought over that. And then Ethereum now obviously has a totally different model with stake and so on, which is, which is different. Right. But, um, would you, would you agree with that?
[00:07:11] Toly: Yeah, and that comes from this idea that they don’t have a source of global, they don’t actually have a global clock and they have to rely on other forms to, to get their notes to coordinate and the problem this is like TDMA solves is like pretty intuitive, like imagine you have two radio towers that transmit at the same time or the same frequency, you get noise.
[00:07:31] Toly: So if you give each one a clock, they alternate by time. Yeah. And they, there’s no longer interference and information passes through. This is actually exactly the same thing that happens with blockchains. You have two block producers that produce a block at the same time, they’ll produce a fork. So now the chain doesn’t know what the next state is and has to resolve it.
[00:07:49] Toly: And that’s effectively noise because now you’ve dropped the amount of information that this, this thing is, is handling per second.
[00:07:56] Balaji: Yeah, and actually, I have a signal processing background, [00:08:00] I didn’t do certainly as much as you’ve done at Qualcomm, but, you know, for the, for the viewers or what have you, the, uh, you know, the noise that you’re talking about is if both of those stations try to compete with each other by blasting at the same time, there’s a beggar thy neighbor thing where I’m broadcasting.
[00:08:15] Balaji: Okay. No, you can only hear me. Okay. I’m going to broadcast and then they compete and then people just hear static because they’re just kind of competing with each other. Right? Yep. Uh, but if instead they take turns, which they can only do if they’ve got a very reliable clock, then I can broadcast stop.
[00:08:29] Balaji: You broadcast stop and we cut up the spectrum by time. But what’s interesting is I don’t think, you know, maybe it’s just, I’m remiss or derelict in this, but I don’t think I’d actually heard the analogy between TDMA and proof of history before. And the thing is that CDMA is much more efficient than TDMA because you have, you know, instead of, um, I broadcast and then you broadcast and then person C and person D or whatever, uh, you use orthogonal functions, you know, wavelets or, you know, hard bases or something.
[00:08:54] Balaji: And then everybody can broadcast at the same time, but their signals are actually have zero inner product with each other. And then you have a much, [00:09:00] uh, you know, much better use of bandwidth. I wonder, is there, is there something, I’m sure you’ve thought about that. Maybe it’s a really trivial question, but is there some analogy for that with the, Um,
[00:09:09] Toly: there is, it’s, it’s pretty complicated.
[00:09:12] Toly: So there’s a way to broadcast multiple block, have multiple block producers broadcasts at the same time. And you’re kind of dealing with a similar problem. If I see, if I have to guarantee that when I receive the information that is partially from A and B that I can decode it. Um, and I think in a very general way, that’s very similar to how these wavelets don’t interfere, you can solve it with erasure coding.
[00:09:36] Toly: You can pad it with enough erasure coding, which is a polynomial, and you receive parts of the message and then you recover it. Um, that’s like on a roadmap. Um, eventually this is something that we want to build, but. This is like one of those nonlinear hard engineering problems. Sure, sure.
[00:09:53] Balaji: I think it’s my, you know, my recollection is CDMA was not done in an adversarial environment.
[00:09:58] Balaji: And so probably [00:10:00] there’s some assumption that breaks down when folks are trying to broadcast, because you have to, you have to nail that basis exactly for it to work, right?
[00:10:08] Toly: The concepts of like having, um, polynomials interpolate the data and give you this like, like now, now you’re no longer looking at like discrete chunks of data.
[00:10:18] Toly: You’re looking at samples of a function, you know, you’re, you’re sampling a function that actually works really well in an adversarial environment because it has this natural error direction. So you can make an assumption that as long as X percent of the participants are like honest, then I will recover the sample.
[00:10:37] Toly: And like in general, you know, the assumption that all these networks run on is that two thirds of the network are honest and therefore that’s a pretty valid assumption. What you want to avoid is that if they are dishonest, what you want is that the natural status of the thing halts. But that’s the result.
[00:10:54] Toly: Like not that it can like steal money or issue like new tokens or something like that. And that’s also [00:11:00] pretty easy to achieve. So I think, I think we have like a real realistic engineering plan to, to build that. Uh, it’s just deploying these systems like on the global internet with random people running this random hardware on random connections.
[00:11:16] Toly: It’s like a massive pain.
[00:11:19] Balaji: It is. Yeah, I know. Of course. Well, that’s the funny thing is even, even with the hardware requirements that you guys, you have fairly beefy requirements for the nodes, right? So even with that, it’s still like, uh, you know, the Dunkirk evacuation? Uh, no. That Well, so basically in, in World War II, the British, uh, were able to evacuate France with a, it wasn’t like a nice military set of ships, it was just this boat and this dinghy and so on.
[00:11:51] Balaji: And they just kind of used everything that they could get together. And that’s kind of how I think about a lot of decentralized computing is it’s like, it’s like the Dunkirk evacuation, just get whatever boat you [00:12:00] can to, to solve the problem. So even with that baseline, that’s interesting. So you still have a lot of.
[00:12:05] Balaji: You know, just a lot of folks in the consensus network.
[00:12:08] Toly: Well, yeah, well, Solana is unique in that, like, um, I think we’re the only network that has a single quorum where there is no sampling. And it’s very large. It’s like about 3, 000 computers all participating in consensus. And they’re voting every 400 milliseconds.
[00:12:25] Toly: So we’ve kind of like, when you look at the Trellama problem that Vitalik kind of laid out that there’s trade offs between performance, uh, decentralization and security, the properties, um, you can’t really scale that without hardware. Oh, or, or you basically take the sharding approach where you start splitting the network and you’re making a trade off there a bit on security like, well, there’s
[00:12:51] Balaji: a third approach, which is zero knowledge or something along those lines, right?
[00:12:54] Balaji: Like where you are, you’re, it’s, that’s almost like the CDMA of R space where you’re using, you have an [00:13:00] algorithmic gain potentially. I don’t know if you’d agree with that. Zero
[00:13:02] Toly: knowledge helps you on the compute side so you can improve how much. Uh, computation, a verifier needs to, like, you basically lower the CPU count, but Well, I was saying
[00:13:12] Balaji: like roll ups, like ZK roll ups, like, you just store less, you use it as a compression thing, you store less information on chain or something like that, that has its own trade
[00:13:19] Toly: offs, of course.
[00:13:19] Toly: You can get rid of signature verification, but you can’t get rid of the state and the ledger itself. Sure. Right. And, fundamentally, that’s, that is the bottleneck, like, there’s tricks, you can apply these tricks to layer one, and kind of compress a bunch of the signatures and do a bunch of other tricks. But at the end of the day, the bandwidth that’s shared between the network, in a very, like, radioed analogy, that channel that’s shared between all the towers.
[00:13:46] Toly: That’s the limit and totally either you build two different networks on different spectrums, right? And then have like algorithms to, to make sure that they can reliably assume that the other one didn’t [00:14:00] create a fault or you just make it bigger, right? You like increase the spectrum and therefore you can increase the number of participants.
[00:14:07] Toly: And that’s the approach we’re taking. And that means that we are the optimization nerds that are looking at like Duff device, like, okay, what if we like. Pre fetch this into memory and like do stuff like this. Can we handle more messages per second and therefore have more participants? And the result of that, I think, has like been awesome.
[00:14:26] Toly: Like we have nearly 3000 validators all in a single quorum, all voting every 400 milliseconds. So this is a single replicated global computer that’s synchronized every 400 milliseconds in the world. So when you drop a piece of data into it. It’s replicated around the world. Which is pretty amazing.
[00:14:44] Balaji: Yeah, yeah.
[00:14:45] Balaji: Yeah. Actually, that’s an interesting way, you know, it’s an interesting way of thinking about it. It’s not just a CDN in the sense of data replication. It’s also compute replication, right? Yeah. And it’s, and, and it’s got verification that the data was replicated consistently across all those places where a CDN, a normal CDN like [00:15:00] Akamai or something like that doesn’t have, doesn’t have the same thing.
[00:15:02] Balaji: That’s just an interesting way of thinking about it. I hadn’t, I, for some reason, I just hadn’t thought about the latency aspect in quite that way. Um, like the, uh, a CDN 2. 0, um, okay, well, so, uh, well, actually, do you have any more remarks on that? I wanted to, I wanted to talk about the, go ahead. Yeah.
[00:15:17] Toly: There’s like kind of a fundamental difference between in, in like values between what Solana is working on, what Ethereum folks are working on, and you can kind of think of it as like, Both care about trust minimization.
[00:15:29] Toly: There’s some amount of that the network needs to minimize the amount of trust that people place in it. But a theory like Solana really cares about global synchrony, that data is replicated globally as fast as possible. And everyone has the exact same state, like as fast as possible. And atomic composition that me as a developer, me as a user, I can like reuse the references, any pieces of this data immediately and like atomically.
[00:15:57] Toly: And those three things are what Solana cares about, and [00:16:00] Ethereum folks have been able to sacrifice the other two while improving trust minimization to a degree, and that’s
[00:16:08] Balaji: I think it’s reasonable. I actually think, I mean, I’m a multi, I mean, I’m a fan, obviously I’m an early Bitcoiner, I’m an early Ethereum holder, and I’m a big fan of Solana, and I think actually, um, I know it’s hard to, without getting into all the coin wars and so on and so forth, My mental model is that you have, um, just like you have different levels of security for different kinds of things, you know, you might have a certain level of security for the president of a country and another level of security for, I don’t know, the, um, I don’t know, the mayor of a city or something like that, right?
[00:16:41] Balaji: And then you have another level of security for just somebody walking down the street, um, who may not be as much of an assassination target or what have you. In the same way that you have, uh, you know, Bitcoin has a certain level of security where it’s like it’s trying to be nation state resistant and then Ethereum and, and completely immutable and Ethereum has relative immutability in the sense [00:17:00] of infrequently mutated smart contracts and they try to keep the chain fairly thin, but not as thin as, or as Bitcoins.
[00:17:08] Balaji: And then you guys are, you know, like, it’s like a third gear on a bicycle and, um, you know, each gear on a bicycle is like useful for different things. And then you guys have. What I call a fatter chain, right? Which makes actually a lot of things very convenient. You just have all the state there. You can just program against it.
[00:17:23] Balaji: You don’t have to have anything off chain or what have you. But it does increase the weight of the chain and you need, um, you know, heavier compute to deal with it. Does that sound like a fair
[00:17:32] Toly: description? Yeah, and like the difference there isn’t that the security is worse or better. It’s that security is more expensive.
[00:17:39] Toly: Like you have to pay for a more expensive full node to get the full trust minimization
[00:17:45] Balaji: guarantees. So what I wanted to talk about today, you know, maybe primarily, is the Solanophone. You know, the decentralized cryptophone is central to liberty and freedom in the world. Um, to civil rights, to property rights, [00:18:00] in this age of, you know, centralization of, of things.
[00:18:03] Balaji: And I can give my brief thesis on it, and I’d love to hear your thesis on the Solanophone as a specific embodiment of this. Essentially, if you, you know, now I’m about 15 years into the smartphone revolution. Certainly our wallet is on our phone, um, you know, with your Apple Pay or what have you, your boarding passes are on there, um, in many countries, uh, to get through like a COVID checkpoint or something like that, you needed to use your phone, they just assumed everybody had it.
[00:18:28] Balaji: You call your Uber or whatever on your phone, you get to your Airbnb on your phone, your data is on there, your communications are on there, a large part of your life is on there, your digital life is on there, but it can be remotely accessed, like for example in China not too long ago, Um, videos of those, you know, protests against, uh, the, like, you know, the third year of lockdowns are videos of those were just like remotely I’ve heard deleted, you know, from like WeChat, they just reached in and deleted them.
[00:18:56] Balaji: We’ve seen, uh, for example, with Google, there was a [00:19:00] report that, uh, some parents were taking photos of their kids for the doctor, for the pediatrician. And they got flagged as CSAM and, you know, their life was turned upside down as a function of that. Okay. And then there’s stuff where, you know, of course people, you know, there’s various hackers that can do remote wipes and, and so it’s over.
[00:19:17] Balaji: And so this thing that really is your sanctum sanctorum, your phone, there’s some being some good moves in the sense of like end to end encryption is now out there. Apple has done some very laudable stuff recently with their, you know, iCloud actually has private keys. It’s getting all these people.
[00:19:31] Balaji: Everyone’s like, Oh, no one will learn private keys. Apple has actually got this new thing with iCloud where you can have private keys and actually get end to end encryption locally. But, uh, we’re still not all the way there towards a, a sovereign phone, you know, where you have private keys that you hold locally, that encrypt cloud data, and they give you not just end to end encryption that’s asserted, I mean, I like the fact that WhatsApp asserts it, but that’s provable because you’re encrypting it locally, right?
[00:19:55] Balaji: And so, um, so that is kind of where I’m at, why I wanted to see a crypto [00:20:00] phone for like 10 years. And I’m so glad you’re doing it, but I’d love to hear your thesis on it and then maybe we can kind of go from there. I think
[00:20:05] Toly: like the really coolest thing about crypto is not the blockchains or the layer ones, is the, it’s the self custody cryptography piece.
[00:20:14] Toly: Yes. Because it’s the, it’s like a, I think when you kind of think of that combined with like the first amendment and like the fourth amendment, you effectively have. The base layer for a stateless society, like I have, I have my cryptography, nobody can take it from me. There’s an, you know, an atomic weapon cannot break the encryption.
[00:20:36] Toly: And like, that’s a very, very powerful tool. And I think it’s because it’s secured by math, it makes it, I think, kind of outside of almost human facilities, like magic effectively. Yes. And I think that was like the revolution what crypto created is that like, you now have consumer grade crypto that people are able to install and download through MetaMask, through Phantom, [00:21:00] through like Chrome extensions or through Ledger.
[00:21:02] Toly: And that didn’t exist before 2008. Like I don’t know if you ever worked at a big tech company prior to that designing something required. They were like armed guards by the Verisign security. Oh really? I didn’t know that.
[00:21:14] Balaji: That part I did know. Yeah, go ahead. How did that work? Tell me about that.
[00:21:18] Toly: Uh, yeah, so the way Verisign works is there’s a certificate chain, they have a cryptographic key that is the root and then that key signs a child key that then gets passed down to like Google and then Google has its own certificate chain down to like its own people, but it’s really,
[00:21:36] Balaji: yeah.
[00:21:36] Balaji: Click the lock symbol near an HTTPS site and it’ll show you the certificate chain. But the armed guards part. I wasn’t aware of it. I knew about like DNSSEC and I knew about like the people who are like the root domain, you know, holders and so on. But tell me about this. Well, if you
[00:21:50] Toly: steal the root, you can reassign certificates and then you can fake all the connections and you can pretend to be Google.
[00:21:56] Toly: So what was
[00:21:56] Balaji: the armed guards thing? So VeriSign basically So,
[00:21:59] Toly: so like [00:22:00] at a big tech company, like if you’re signing code that says, Hey, this is like Qualcomm’s code, or this is Google’s code, that’s now running in your device, and the device trusts that code to have certain privileges, you have to sign it with a certificate, and that signature process is extremely like, it’s a ceremony.
[00:22:20] Toly: There’s guards that guard the key for my decoded, like, a Faraday cage. It’s a whole thing. It’s a whole thing.
[00:22:28] Balaji: Interesting. So, I mean, you know, what that does remind me of, actually, is the security for a crypto exchange, where you do use a Faraday cage. And you do like, you know, you just don’t take any, you know, you, you, you sign offline and you air gap and you cold story.
[00:22:43] Balaji: You just all of the different defenses in depth. Right. Um, but go ahead. You’re going to say.
[00:22:48] Toly: Like consumers basically didn’t have access to this. It was very likely PGP was like the encryption tool for signing emails and all of this stuff never really worked at scale, but [00:23:00] all of a sudden with crypto, because there’s some.
[00:23:02] Toly: You know, need for like people to have fun, like either there’s for speculative reasons or just because they’re developers that are messing around with Ethereum and looking at it as a new programming environment or Bitcoin. People started building all of these tools to make it easier. And you started MatterMask, I think was a massive step up from BGP.
[00:23:24] Toly: Like if you ever, as a, as a human, like. I could not figure out BGP. Right, well.
[00:23:30] Balaji: And I’m an engineer. That’s the thing is, you know, I have a rule of thumb, which is, if it’s hard for you and it’s hard for me, it’s usually too hard. Yeah. Right? Like, like if it’s hard for like basically, you know, our friends, whatever, it’s usually too hard.
[00:23:42] Balaji: Right? And, uh, but, but your point is actually a really good one because. The entire, you know, PKI problem, right? Public key infrastructure. There’s this whole slew of actually, nowadays, as I’ll get to, very useful papers. But at the time, all these papers were written under the assumption in the 90s and the 2000s of, [00:24:00] assume the PKI problem is solved, right?
[00:24:02] Balaji: Assume, just for the viewers, that everybody has a private key that they keep secure and accessible at all times. And a public key that’s, that’s available to everybody on the internet. If everybody on a network had that, there’s all kinds of amazing magic that you can do with things that are, you know, like, like secure email and messaging and all this stuff, right?
[00:24:22] Balaji: Transactions. The problem is, I think the part that was not appreciated is that to have something both A, secure, and B, accessible at all times is actually very costly because you could have it secure without being accessible by just going and putting in some safe deposit box that you never, you know, it’s dug up in the ground, right?
[00:24:41] Balaji: Or you could have it accessible without being secure by just having it, you know, like pinned to the top of your Twitter or like taped to the top of your computer or something like that. To be both A, secure and B, accessible is like on your person. It’s like, it’s like literally a wallet, right? Yeah. And, and so you needed a strong incentive for something like that.
[00:24:59] Balaji: And you need to [00:25:00] teach people that if it was not secure or not accessible, they would lose money because it wasn’t accessible. Then they couldn’t find the key. And if it wasn’t secure, someone else would take the key. And so cryptocurrency incentivized the build out of public key infrastructure. Without it, we wouldn’t have had millions, hundreds of millions of people trained and incentivized to keep their private keys private and secure, right?
[00:25:21] Balaji: Secure and accessible. And so that’s enabled this entire, all this magic of, you know, this, this like world that these, these papers assumed into being now that assumption is actually manifest. I’d love your thoughts on
[00:25:33] Toly: that. Yeah. And that, that’s exactly what I was kind of my realization too. Like once you have PKI, that’s like consumer scalable PKI, all of these ideas, like become possible.
[00:25:44] Toly: You can like, you late, what does, it’s not even going to matter which layer one people are using when you have 8 billion people that can sign with cryptography. That’s
[00:25:53] Balaji: right. And it’s like, it’s like, you know, in the early days when people started using the term crypto, it was something where [00:26:00] cryptographers, a lot of them were mad.
[00:26:01] Balaji: They’re like crypto means cryptography or whatever. And it’s a little bit like saying, you know, of course you could do software before the internet. I mean, there were personal computers. There, there, you can do disconnect computers, but software and the internet are almost synonymous today. They’re not of course, but because so much software runs on the internet and so much software depends on the internet and so much software assumes the internet.
[00:26:21] Balaji: And in the same way, like so much cryptography will assume cryptocurrency because if you’re going to assume that a private key is there locally, I mean, this is the thing where Apple introducing this thing for iCloud. Or WhatsApp doing intent encryption, or, you know, Google has like DNS verification for you to confirm you have your domain, right?
[00:26:40] Balaji: Amazon, of course, has its EC2 keys. All of the majors understand that you need, like, some sort of, you know, key challenge or some sort of digital verification. And all that will eventually, in my view, and probably yours, eventually go to some sort of, you know, crypto based thing, right? Because it’s just the platform that has the most [00:27:00] adoption.
[00:27:00] Balaji: We’ve got 300 million crypto users now. In five years, 10 years, we may have a billion or more, right? Go ahead.
[00:27:05] Toly: And it’s much easier once the, like the, all the tools that we built for crypto make that communication challenge response process much, much easier and much faster. And like, you’re now no longer need to be like PhD in cryptography and cryptographic secure protocols.
[00:27:23] Toly: You can actually use a bunch of JavaScript to go punch it out. And all of a sudden you’ve got something that works. And people like, we’re still not at a stage where you can make that secure, like, like, uh, entry level engineer can build a secure system, but we’re at a stage where you can build these systems and they’re somewhat straightforward to audit.
[00:27:44] Toly: And you can get to that, like, okay, this is a secure protocol. We’re doing the key exchange over Solana or whatever. And now we’ve set up the communication channel. Everyone can go and talk in a secure environment.
[00:27:55] Balaji: That’s right. And actually, you know, one thing you touch on, which is User interfaces, like really [00:28:00] nice and clean user interfaces.
[00:28:01] Balaji: People take them for granted because they’re simple to use. They think they were simple to develop and it’s actually quite the
[00:28:10] Toly: opposite. Go ahead. And it’s a, the simplicity comes from like, uh, consumer scale. When you can make an assumption that the internet connection’s always there, it always works. And when you, when you get to a point when developers can make the assumption, okay, everyone’s going to have a wallet.
[00:28:26] Toly: And when they talk to my website, there’s a wallet. You actually blow away a massive part of like normal developer user account management. I don’t know if you ever had to build a SaaS or whatever. Like managing accounts, managing their credit cards, doing all of this stuff is like a team of 20 or 30 in any decent like size, like SaaS company.
[00:28:49] Balaji: And, and, and you basically, I mean, you can rely, you can try to use Facebook login, Google login and so on and so forth. And you can actually get quite far with that. But eventually you’re almost always going to need your own account system because [00:29:00] it is the central node in everything you’re doing, right?
[00:29:03] Balaji: Like you have types of users and all that type stuff and you have admin users or whatever. But now we have the portable login. We actually have this article called the billion user table, which I’ll flash up, but basically, you know, the advent of ENS and SNS and so on. Like, now you can have a portable name between sites, which is the portable social profile that people wanted.
[00:29:23] Balaji: Um, it is, is locked to a blockchain, but you can import and export your profile data in and out of sites, which is kind of what you’re talking about, right? It’s like a third model for accounts, right? Not username, password, not log in with Google or Facebook, but log in with your private key effectively. Go ahead.
[00:29:38] Toly: Exactly. So once, once you assume that consumers of PKI, it’s just. Like, not only can you build like just normal applications much faster because you’re not managing accounts. But you can now also build financial applications, which has really been impossible before. Yes. And like, you can build a bank, right, like two, two devs can build a bank in three weeks.[00:30:00]
[00:30:00] Balaji: And you know, the thing is, one thing I find, and you’re probably aware of this as well, is folks who are first or second gen immigrants are just so much more conscious of how big a deal that is. Because we’ve all experienced some problem with the banking system, either in our own countries or, you know, with family abroad or something like that.
[00:30:20] Balaji: And the more you’re kind of, you know, more the developed world rich country, whatever, you know, legacy infrastructure works for you, the less you see the need for crypto because you don’t, you just don’t see the problem. You’re like, you know, my Starbucks thing kind of works. Why do I need this thing or whatever?
[00:30:33] Balaji: But if you’re sending money overseas, you know, Murtaza Hussain had a good post on this where, you know, he’s in, he’s got like family, totally innocent family, they happen to be in a country. Which is just very hard to send money to and boom, he’s just able to do it. And so that kind of thing, like now, boom, you can just instantly set up the equivalent of a bank account and send and receive money.
[00:30:51] Balaji: It is actually like, it’s not a sm that’s the thing is people think, Oh, where’s the use cases? Well, it’s a thousand X, it’s a hundred thousand X easier to set up a bank account in terms of time [00:31:00] and to send money across borders and to do accounting on chain and, and, and all these different things.
[00:31:04] Balaji: Whatever. I, I know I’m preaching to the choir, but go ahead.
[00:31:07] Toly: So, so to get to kind of like the mobile phone, why a mobile phone? Yes. Well, like Me as an engineer at Qualcomm, this is while I was there, this is basically, I think, before even Bitcoin was a thing. We were already working on this idea of a trusted display and a trusted computing environment on your phone.
[00:31:27] Toly: And that means that it’s, it’s, by trusted, what this means is that it’s an environment that the operating system that runs all these user applications can not, can not touch. It’s like, it’s like, exactly, SGX,
[00:31:41] Balaji: right, but SGX did have a problem with it, right? If I was like a while back or, go ahead. These,
[00:31:46] Toly: these systems are hard to build and it takes a lot of iteration and like the SGX design was more ambitious than like something like TrustZone and ARM.
[00:31:54] Toly: Um, do you basically the way kind of from a very high level the way it works is it segments [00:32:00] memory. Is it, there’s some chunks of physical addresses that are accessible to the TrustZone and nothing else in the operating system can read it. And you need to also make the display trusted environment. Uh, that’s a very kind of unintuitive thing, but like, let’s say I had like a, a widget for you to enter your seed phrase on the phone.
[00:32:23] Toly: An application from a malicious developer replicated, yeah, mimic the exact same screen that asks you to reenter it. Users would get phished and accidentally enter it and now your seed phrase is gone. So you need a very, a way to override the display and make sure the operating system cannot mimic that and like create that environment.
[00:32:42] Toly: Very
[00:32:43] Balaji: similar to the HTTPS lock symbol.
[00:32:45] Toly: Exactly. Once you have those pieces, you can then effectively treat. This device, very similar to a hardware wallet, like a ledger. I would say that like, most people should not use this for [00:33:00] cold storage, like storing their life savings on it. But even if, even if the, if we like prove that supply chain and everything else and audits was as secure as a ledger, you still need to get into this like idea that I have my cold wallet, which is my savings account.
[00:33:15] Toly: And that has air gaps, security, and this is my hot wallet that I use day to day, and I use it for talking to applications, and that if I lose my wallet, it’s not catastrophic, right? Right,
[00:33:27] Balaji: right, right. No, I think, I think it’s very important. I mean, the, um, the thing about it, though, is It is much, while you should still have cold storage, this is, this is putting up a much higher bar.
[00:33:41] Balaji: Uh, a, a, a warm wallet is at least not remote shutoffable very trivially, right? Exactly. It can be hacked, but it’s like, it’s still far more secure than the status quo. Um, and you know, there’s something else you said, by the way, which, uh, is I think really important. I didn’t want to skip over. When you said restoring the first amendment and the fourth amendment, I actually wrote an article on this called, uh, Bitcoin is [00:34:00] civilization.
[00:34:00] Balaji: That’s actually the, uh, domestic policy version. And then also, um, Uh, you know, the, uh, you know, uh, an article for foreign policy on the international policy version. But to your point, free speech, how are you going to do it? Well, you need to actually encrypt it speech. And how do you have protection from search and seizure?
[00:34:16] Balaji: The TSA is like patting everybody down, searching everybody without warrant. Well, encryption restores your ability to walk around without search and seizure, especially in those countries that don’t have even those. Uh, traditions of civil liberties, you know, unfortunately, in many places around the world.
[00:34:30] Balaji: So we’re basically, wherever there’s a phone, we’re parachuting in civil liberties as well. Now, the only thing I would maybe edit on what you said is, I do believe that what that does is it gives people the ability to opt out of, you know, like bad states, you know, failed states, surveillance states. I think the next step is going to be them opting in to good jurisdictions.
[00:34:51] Balaji: And I, I think we’re both, you know, in a sense, pro civilization because that, that’s actually the third bit in, in, in my book, The Never Say I Was Like. The least popular word in [00:35:00] the world is re centralization. You know, because the people who are currently running centralized systems are like, why aren’t those okay for you?
[00:35:06] Balaji: And the people who are pro decentralization are like, why would you want to re centralize? That’s where we’re escaping from. But the way I kind of think about it is, you know, decentralization is like hitting the fire alarm, you know? Everybody runs out the building and in different directions, but then they want to like regroup at the like, you know, the, the union point, right?
[00:35:24] Balaji: And maybe not in exactly the same way, but, you know, you, you need to have those sovereign individuals, they, they unbundle from the system that’s failing them, they use the civil rights, the property rights that the crypto phone delivers to do that, right? And it’s really both. It is, I mean, it’s like, you know, punk6529, you know, on Twitter, right?
[00:35:43] Balaji: So he has this really great thread, which actually expresses that so clearly, which is most civil rights don’t exist in practice without property rights. They are basically the same. Like, You want to buy a sign to protest? You need some money, right? You know, almost all kinds of things. If you can shut off somebody’s banking, you can basically like, throttle them and they don’t have, you know, the [00:36:00] cost of speech.
[00:36:00] Balaji: If you raise it too high, they don’t actually have freedom of speech, right? So the, sort of, so the, the left libertarian, the right libertarian have something very strong in common here, you know, right? And, um, so, so putting that together, basically, Uh, that sovereign individual who has those civil rights and property rights that are restored, in my view, protected by the cryptophone, can then bundle with other people to have a sovereign collective, where now they can actually, because they have property rights, they can do collective goods together, because they have civil rights, they can speak freely together, and they can come up with a polity that suits them.
[00:36:30] Balaji: Go ahead. Let me know your thoughts on that. These are like the
[00:36:33] Toly: freely associated communes that like mikhael Koonin dreamed of. Like there’s no, there’s no oppressive state. Everyone associates at their own will, they bring whatever they can and everyone kind of works together and connect it.
[00:36:46] Balaji: Yeah. Well, so, so, but, so it’s funny, there’s, there’s that utopian version, but there’s actually the actual version.
[00:36:53] Balaji: And the actual version is like, uh, mid 18 hundreds America. So it’s called, uh, The Communistic Societies of the United States. It’s a book by this guy, [00:37:00] Charles Nordhoff. And he goes through and, for example, like the Oneida Commune in, in upstate New York was actually a group of people, because there was all this land, right?
[00:37:07] Balaji: So people could just go to wherever piece of land, and if they liked it, they stayed. And if they didn’t like it, they’d apply to another community, and they’d move there. And some of these things failed. But some of them really worked and like Oneida today is actually the name of like a glassware manufacturer, as you might be aware, but back in the day, it was like a, it was like an opt in society where there’s enough geography.
[00:37:23] Balaji: And actually the kibbutzes in Israel were also kind of like this. And some of them now today have become like big companies like Netafim is like an agricultural thing. So the reason I say that is people, there is the utopian idealistic version that doesn’t actually get anywhere, but there actually have been successful experiments in opt in living, of which arguably the United States is one of the best.
[00:37:42] Balaji: Where, you know, it’s only 4 percent of the world. Most of the people who came here came here voluntarily, of course, you know, sense of slaves notwithstanding, Native Americans, et cetera, that’s important, but like, let’s say 90%, 85 percent of the population came voluntarily, and it did build something great, it is an opt in society, of course, there’s lots of other [00:38:00] dimensions to it, but that, that is something, so the only reason I just say that is, otherwise people will think, oh, biology is just so unrealistic and utopian, but you’re right, there’s the Bakunin aspect, but there’s also the American aspect, there’s the, there’s a realistic aspect, go ahead.
[00:38:13] Toly: Um, and, So like, why a phone, I think is the engineering aspect of the phone for PKI is just really, really obvious. Yeah. This, this device already has the hardware to support it. Like where Apple and Google store your biometric scans, it’s in the secure element. Like when you do Apple pay, that’s a trusted display.
[00:38:34] Toly: Nobody else can fake it. And like all this stuff already exists. So we’ve taken. Basically it’s stock Android builds, uh, and like worked with like an OEM and five engineers like made the changes to enable cryptographic signing in this device. This
[00:38:49] Balaji: is, I’m honestly, I’m so happy you’re doing this. I actually, you know, the thing I hadn’t connected until just now is I was like, you know, phone’s pretty ambitious and you guys have shipped it, man.
[00:38:58] Balaji: It seems to work. And I [00:39:00] realized your Qualcomm background is probably helpful in this, in terms of, you know, it’s like knowing what to turn and who to talk to and so on and so forth. Is that correct? Like, yeah. Yeah.
[00:39:09] Toly: I think understanding the scope of the problem, like we were able to minimize it to like, we need the smallest change inside the trust zone to support this.
[00:39:18] Toly: It’s not going to take a team of a hundred people, it’ll take five and you can work with any OEM to enable it. Wow. That’s.
[00:39:26] Balaji: Go ahead, finish what you’re saying. I’m really psyched by this.
[00:39:29] Toly: And that means that, like, this technology, like, we, we have a really good relationship with, like, the Google folks and, like, it runs all the Google services.
[00:39:37] Toly: There’s no, like, objections on their side, like, Oh, you can’t do this or something like that. It’s just Crypto is really, really small. There’s not a, like, I would imagine that like the number of Twitter users that have enabled the NFT search is probably under 50, 100, 000 that have, that have. And like, when you look at Twitter, it’s like 300 million people to them.
[00:39:59] Toly: [00:40:00] Like, okay, crypto is a buzzword. We can enable this feature and do a whole bunch of work to support self custody or whatever. And then only 100, 000 people opted to it. It’s a failure.
[00:40:11] Balaji: Well, so here’s the thing, actually. I think that that is going to be pretty important for them in the future, and here’s why.
[00:40:18] Balaji: 100 percent Right? So, so, like, lots of things are like this, where, like, for example, CryptoKitties, you know, it was like this small experiment, NFT thing, in 2017, it was like a little bubble, and then NFTs became a huge thing, and then they crashed, but they crashed back down to a much higher level than they were before.
[00:40:33] Balaji: One thing, for example, um, you know, there’s an app, Solana app, uh, I think it’s called Glow, which you’re probably familiar with, right? And you can just take a photo of something and NFT it. And people think, oh, that’s so stupid, what are you doing? The answer is, it’s not just, okay, yeah, you can sell it, maybe, or whatever.
[00:40:49] Balaji: They’re thinking of NFT as speculative art. I’m thinking of it as a crypto instrument. It’s a crypto camera that at least that person at that time can assert on chain that [00:41:00] this thing was photographed by this device. And maybe you include a device signature with it. And there’s things you could do, maybe even like a hardware fingerprint or something.
[00:41:06] Balaji: Why is that so important? Because AI is coming online and everything will be considered fake online or on Twitter unless there’s a crypto signature of it. So it’s like, you know, um, one technology is going to create problems and everyone may at least mitigate them. Right. And so that code that Twitter has for, you know, code paths for handling NFTs.
[00:41:25] Balaji: It may be something where you have a checkmark, not just on the person, but on the image that it came from a crypto camera and you can actually show it’s, uh, it’s, uh, it’s supply chain, like it’s digital supply chain. It came back from this block at this time. And that doesn’t mean, by the way, to be clear, that doesn’t mean it’s necessarily 100 percent real.
[00:41:42] Balaji: It just means at least you can have some, some things you know about it. You know, some of the metadata, some of the China custody. It had to have been faked by this guy at this time. You know, on this chain that narrows it down a bit because with that person about the motive to fake it, you know, so I’d love your thoughts on something
[00:41:56] Toly: like that.
[00:41:57] Toly: So I, uh, yeah, absolutely 100 [00:42:00] percent agree with you. I think like in the post product market fit crypto world, like every company is going to have to have some interface to crypto and like, but the hard problem for them right now is that like, The number of active users with PKI is actually pretty small.
[00:42:15] Toly: You have a bunch of people that hold Bitcoin, a bunch of people that may have traded on exchanges and just buy and hold it, but like when you look at the number of active transacting users in NFTs and all these other stats, when you combine them, OpenSea’s global total number of users that have used OpenSea is around 5 million.
[00:42:35] Toly: Number Listen,
[00:42:36] Balaji: I have an argument. I have an argument on this. Historically. No, I understand. Yeah. I understand. I understand. And I’m not, I’m not One thing that I’ve, uh, thought about, and this observation isn’t original to me, I think Dixon has made it, other folks have made it, you know, social media arose where it had lots and lots of users, but then for a long time people said, how is this thing going to make money?
[00:42:57] Balaji: And, for example, Instagram had tons and tons of [00:43:00] use, tons of use cases, and when it was acquired in 2012, You know, Jon Stewart, because it had no revenue at the time, by the way, it was actually a huge, huge, uh, coin flip by, by Zuck and Facebook was not, you know, they only had like 4 billion in cash. They were a few weeks away from IPO.
[00:43:15] Balaji: Facebook itself, it was questionable as to whether they were going to make money on mobile. They spent a billion dollars acquiring Instagram. He didn’t consult the board. Now, today, it’s a 100 billion company. Everybody thinks it’s such a genius move, obvious move at the time, antitrust. It was a very non obvious move then, right?
[00:43:29] Balaji: Point is, Instagram had zero revenue, but it had lots of use. And crypto has, in some ways, the opposite emergence mode from social media, where it generates tons and tons of money, and it’s the use cases that are second, as opposed to social media, which was use cases first, revenue second, right? So that’s kind of one angle on your argument.
[00:43:46] Balaji: Maybe you have some thoughts on that.
[00:43:48] Toly: That’s actually, like, You, you almost led me to the opportunity for us. Great. How can we, how can like a tiny startup with like five engineers messing around with Android take [00:44:00] on like, like Google and Apple? Cause like the thing that we’re really delivering here is the PKI for users, but also an app, a dApp store, like an application store for developers that does not have the fee model that the Google and Apple stores have and does not have the restrictions on treating PKI as a feature.
[00:44:20] Toly: On the device for token gating, for NFTs, for minting these NFTs, you know, and like submitting them to a chain and selling them. Like that’s basically prohibited on both Google and Apple. Like, and that’s, that’s kind of like the feature set. And the opportunity for us is that While crypto has a small user base, it’s probably the spendiest user base on the internet.
[00:44:42] Toly: Like I would say it’s with, it’s like the 0. 1 percent of people that like generate the most like volumes, the most trading, the most like financial activity on the internet right now are the crypto users. So if we can get this device to 25, 50, 000 of those users, that’s [00:45:00] actually a better distribution channel for devs.
[00:45:02] Toly: Then the big app stores are today. Anyone that’s building in crypto would rather get their application to 50, 000 active Web3 people that have used OpenSea or Magic Eden on a day to day basis than try to, like, find some way to sneak in a feature in the apps in the Apple App Store behind the scenes.
[00:45:21] Balaji: There’s a couple of questions about that. A, um, I assume you did have multi wallets on there, so it doesn’t just hold Sol. It holds ETH and BTC and so on, right? So,
[00:45:29] Toly: like, uh, V0 right now would just Why, the reason we shipped is that,
[00:45:33] Balaji: sure you’ve got features, but it’s on the roadmap to add. Yeah, absolutely.
[00:45:37] Toly: Yeah. And like there it’s open source and we have like external contributors working on adding ETH support. There’s actually an EVM equivalent runtime called NEON on Solana, which requires ETH support. So like, even if we were just like trying to box it out, like it would eventually happen because all this stuff is
[00:45:54] Balaji: interoperable.
[00:45:55] Balaji: I’m glad you’re taking that approach because I think once a Solana phone holds both [00:46:00] ETH and BTC, I mean, maybe it’s not a. I mean it’s not a exactly a blocking requirement or something like that I think it’s like worth checking out even now and that’s a software upgrade so you could get it now and try to but um the thing is that basically uh this is I I’ll I’ll make a couple points these are obvious points to you maybe not points that folks know in the ecosystem Web development was actually held back through most of the 2000s by Internet Explorer.
[00:46:27] Balaji: Absolutely. Like, right? Yeah. That whole time period has been erased from memory. Right? Everyone thinks, oh, go ahead. I remember it. You remember it. I remember it. Right?
[00:46:35] Toly: It took literally like an engineer at Google to like Accidentally deprecate Internet Explorer 6 and post a link, like, Hey, update your shitty, shitty browser.
[00:46:47] Toly: And like, it wasn’t a PM that made that decision or a director. It was like, there was like a fire in Google management. Holy shit. Who, who pushed this feature?
[00:46:57] Balaji: Yes, it, well, it was, it’s something where like [00:47:00] that whole story is worth telling. And the reason I just say it is because web development, as we currently know it.
[00:47:06] Balaji: It was held back by Microsoft because they beat Netscape. So just for those who don’t remember this, you know, cause they’re too young or whatever. In the nineties, Bill Gates saw the internet thing coming to his credit. Okay. And Netscape had just had their huge IPO, I think 1995 and Netscape Navigator was coming and he decided, okay, we’re going to ship a browser.
[00:47:26] Balaji: We’re going to bundle it with the operating system and we’re going to asphyxiate Netscape. And, and basically that’s what they did. And that was actually part of the basis for the antitrust trial in the late 90s, early 2000s, was they were bundling their browser with the OS. And it was, but it was worth it for them in a sense because Um, they took all the revenue out of the space.
[00:47:44] Balaji: Netscape did sell to AOL. It was like, but, but basically what happened was once Microsoft had killed Netscape as a, a competitor, InterExplorer had like 90 something percent share. And they basically just like kind of stopped development after that, right? Because they didn’t want the web to become a big thing.
[00:47:58] Balaji: They wanted everybody to run on [00:48:00] Microsoft Windows. Now 2000s, there were two alternatives to Windows, which were A, Apple and Mac OS, um, which is based on BST and B, Linux. And Linux was a server side technology. That had an ethics around it, open source and whatnot, which we forget today because Linux is now so commercial.
[00:48:14] Balaji: By the time is a big thing. And Apple was still a shadow of what it is today. It was pre, it was even pre iPod. Yeah. Right. Pre iPod, let alone iPhone the monster it is. Right. So Microsoft was very dominant in the early two thousands and it was basically choke pointing the entire web and Google just bided its time because it didn’t want to get killed like Netscape.
[00:48:36] Balaji: And, uh, it, you know, just did search, search, search, search, search, all the way through 2004. Focused on that completely over six years. And, by the way, people wrote off the internet after the dot com crash, right? For years, this was, it wasn’t, it was like so out of the news. That it wasn’t even something that people attacked.
[00:48:54] Balaji: It was just thought of as obviously, you know, a failure, whatever. They were all focused on Iraq. They were all focused [00:49:00] on terrorism. That was the 2000s, right? Even in 2009, somebody wrote this article about, like, the power centers in the U. S. And they didn’t list Silicon Valley among it. Okay. They had like, you know, the Pentagon and they had Wall Street and all this stuff.
[00:49:13] Balaji: They did not list Silicon Valley because it was still gadgets. It was like, you know, maybe you have your iMac and there’s Microsoft Windows. And that, Facebook was still in the 2000, 2009, still like for college students. All that stuff was just not even there. Point being, that it was only in 2004 with Gmail, which actually pioneered Ajax, and 2005 Google Maps.
[00:49:33] Balaji: And it took Google’s engineers, which were the best in the world, to figure out this backdoor. Intro Explorer left this backdoor, right? For XMLHttpRequest. And then, you know, Firefox started gaining some traction in the mid 2000s. And then Chrome came out. I think it was only like, what, 2008 or 2009, right?
[00:49:47] Balaji: Was that right? Okay. And HTML5 was like the late 2000s. And HTML5 was this exciting thing. Oh, you can do the web again. But that was like 10 years from the dotcom crash in 2000 to actually being able to do web [00:50:00] apps kind of sorta in 2010. And now the iPhone app store only came out in 08. All of this stuff is like extremely recent.
[00:50:08] Balaji: And so the reason I bring all that up is I think that, you know, the internet only came back by public acclimation as an important thing in February of 2013. And that’s when the tech lash began, but it’s basically about 12 years between the dotcom crash in 2013. Before tech was taken seriously again, the reason I say this is a big part in my view of what’s been holding back people like, oh, where are the applications for crypto and so on and so forth.
[00:50:32] Balaji: My first answer to that is, well, there are applications. We’ve got digital gold. We’ve got crowdfunding. We’ve got wire transfers. We’ve got setting up bank accounts. We’ve got a lot of stuff. It’s like the, what are the Romans ever done for us? There’s all these things, right? The second is, well, block space has been a constraint for a long time.
[00:50:45] Balaji: And so the apps that are being developed are those that minimize use of block space, like Bitcoin just hold for a long time. Where Ethereum is buying to the crowd fund and then still hold for a long time. It’s minimal on chain. And only now with what you guys are doing and what other chains are doing is block space [00:51:00] much less of a constraint.
[00:51:01] Balaji: Right? So that’s the second thing. And the third thing is what you’re doing with Solana phone. Because Apple, as you know, was just like ridiculous for many years in terms of filtering apps and you couldn’t do crypto stuff on the phone. And they’ve gotten a little better on that now. But it’s like a, it’s like one step forward, two steps back sometimes.
[00:51:18] Balaji: And now you’ve got a phone which is totally unlocked and so you can do all the crypto stuff on there. So, so those three things I think are why like Solana has like this really interesting mobile app culture now, those things have all like come together now finally, go ahead. Yeah, that’s my take on it, maybe idiosyncratic.
[00:51:32] Balaji: Um,
[00:51:33] Toly: there’s like a tidbit of history there that I think, at least in my memory, this is kind of how I remember it. The web sucked. Until the iPhone released and Apple decided to actually fork KHTML into WebKit, which is the open source, uh, like web browser that was built for KDE, which is part of the desktop suit for Linux.
[00:51:56] Toly: And this is like. The people, like at the time I was like a huge Linux [00:52:00] nerd, I ran Linux and like the year of Linux and desktop was always going to be next year. Yeah, yeah, yeah,
[00:52:05] Balaji: it’s actually arguably come, it’s actually come because Android is the new desktop and that’s the mobile Linux, right? But, but, but it came in a different brand, go ahead.
[00:52:12] Balaji: And, and like
[00:52:13] Toly: the Apple engineers working on WebKit and make, and that thing being open source. is when Google then forked WebKit and made, and made Chrome. And that was like Google and Apple both working together to kind of like build a browser that was equivalent to Internet Explorer, and then like could kind of drive the web to like next generation, like these dynamic application environment.
[00:52:38] Toly: Was like, I think what saved the web and there’s kind of a deep irony there that now, like both Google and Apple run these like app stores that are anti web arguably that want to, in some ways like neuter the web and prevent these like rich applications running in like a browser only environment. So, we’re like back
[00:52:57] Balaji: full circle.
[00:52:58] Balaji: We’re back full circle. And actually, [00:53:00] Microsoft, I mean, I give a lot of credit, immense credit, to Satya for turning around Microsoft and so on. But we may literally be in this weird history of repeats thing with Microsoft about to like choke points the entire tech ecosystem through an API running on its servers.
[00:53:14] Balaji: Do we want that again? I mean, come on, right? You know, like with all the props, I hope they make tons of money. I just don’t want them to have the same control that they did, right? And, but, but yeah, your point is very good because basically it took Google and Apple and Firefox, right? All, I mean, you know, like a really good startup because Firefox had Blake Ross, who’s like a super genius, you know, like browser developer, like this.
[00:53:34] Balaji: Yeah. I, you remember him, right? He, I actually, I don’t know what he’s doing now, but he’s like an early engineer at Facebook also. And he like coded Firefox when he was like a teenager. So Blake Ross at Firefox and, uh, and Google and Apple, all three of them took years before finally IE6’s market share came below 50%.
[00:53:51] Balaji: I don’t remember what year that was, but it was like, I feel it was the late 2000s. That’s my guess. We’ll go, we’ll kind of look it up. I might be wrong, but I feel like that, right? Yeah, I think you’re right. Right? [00:54:00] So, so basically, and then, and only then, could all the stuff that Paul Graham was talking about with web apps and Y Combinator, only then, late 2000s, early 2010s, could that stuff finally become feasible.
[00:54:11] Balaji: And, and even like Netflix, for example, streaming Netflix, House of Cards, I think was only 20, 2012, 2013, right? They were doing mail order DVD. The web just didn’t work for long. The reason I just say all this is, people have retconned that. Because it works today, they think it instantly worked.
[00:54:28] Toly: And this is where I think like PKI is going and I don’t know if it’s going to take like Solana like SMS showing that there’s now developer traction and devs are shipping apps and making real revenues to get like the Google and Apple to change like my, my, my kind of bet is that like in the, in the best case we succeed, we get one to 2 percent like market share of like global users using SMS.
[00:54:57] Toly: And these are the web three users and all of [00:55:00] a sudden developers that are shipping applications to this user base are generating more revenue than they do in the other app stores. And that’s going to be the inflection point that forces the big companies to be like, okay, we got to like change your business model, which is going to be really painful because the career is built around that business model.
[00:55:19] Toly: Right? Right. And support
[00:55:21] Balaji: PKI. I, I, it’s, it’s funny, I think, well, you’ve, you’ve done a very good job at Solana. Uh, I mean, you have a, you have an Instagram deal, you have a YouTube deal, right? Um, it just, you know, Twitter has got NFT support and so on. Um, so you’ve done a good job of interfacing with the big tech companies, um, with the open minded people there who at least are putting a toe in the water, right?
[00:55:44] Balaji: Reddit also has NFTs now. Um, Discord is, uh, you know, they had some pushback last year, but they’re, I think, gradually adding some crypto stuff and so on. Um, and, and I’m also starting to see apps like, uh, like Luma, for example, that have, um, you know, [00:56:00] Ethereum and Solana support, um, or, uh, like Spatial. io has some like crypto support.
[00:56:04] Balaji: So you’re starting to see it not as like necessarily the, the be all and end all of the app, but as a feature in the app, which you can log in with Metamask or, uh, you know, um, Phantom or Google. So that’s like just starting to happen, but it’s happening kind of out of public site in some ways. It’s only like the hardcore we’re seeing it.
[00:56:23] Balaji: Go ahead. And these
[00:56:24] Toly: are the, the small stages of adoption. And I think these are the kind of the, the early, like, you know, it’s like really painful to zero to want to start up. It’s really painful to zero to want to make an ecosystem. Yeah. So like a whole like new programming paradigm. It’s, it’s also really, really hard.
[00:56:45] Balaji: Oh, oh, and Rust. And Rust, by the way, yeah. I like Rust, but Rust also adds a degree of, uh, a degree of difficulty to something said, right? It’s great, great language, but, um, the, you know, it’s funny though, one thing is, one of my theses, I love your thoughts on this is, [00:57:00] after mid 2017, I realized that we weren’t going to have another trillion dollar American tech company.
[00:57:09] Balaji: The reason is, the backlash against Uber, you know, backlash against Uber, but Ethereum was going vertical at the time. I realized, okay, we’re going to have the American tech ecosystem. We’ll have the Chinese ecosystem, and then we’ll have the global ecosystem where the only other thing that can get to global scale will be those things that have crypto backends.
[00:57:28] Balaji: And actually that is, you know, the, you know, in a sense that competitors to Google and Facebook and so on are actually already in plain sight. They are. Bitcoin and Ethereum and Solana and you know, the, basically the things that are built around those chains as opposed to companies. It’s like communities and currencies as opposed to companies.
[00:57:46] Balaji: I’d love your thoughts on that.
[00:57:48] Toly: I think that the, maybe I’m like too much of like an anarchist. I think it’s going to look more of like smaller companies that are very [00:58:00] hyper optimized for their user base. Yeah. I think it’s going to be very hard to build a global brand that scales up. Maybe protocols can do it, but it’s going to be like local companies that are like, Hey, I’m using Ethereum and people have some trust to, to that brand, or I’m using Solana, or I’m using like this protocol for matching drivers to, to passengers and these small companies all around the world can all coordinate around this one.
[00:58:31] Toly: Protocol and use that to bootstrap their like initial user base, initial like set of drivers and stuff. And like, cause like, I think what, what chains provide are like basically marketplaces and that’s a very, uh, small thing, but very powerful thing because the web runs in a marketplace. It’s the ad exchange marketplace.
[00:58:51] Toly: Everything runs in a marketplace somewhere. Right. When you’re buying stuff, somebody has to match you with a seller. And like having [00:59:00] these decentralized transparent marketplaces with branded protocols, I think could help bootstrap these like small communities that are like now running effectively like Uber, but now Uber that’s built on top of a protocol.
[00:59:12] Toly: That protocol may not actually be as extractive as a trillion dollar big tech corp. You
[00:59:18] Balaji: know, you mentioned something which makes me think of the FAT protocol thesis from like a few years ago. The thing is. The FAT protocol thesis, I think, is in the long term, may be partially true. We may see some app chains, but I think in the short term, it is not true because what will happen is you’ll put the minimum necessary state on chain, correct?
[00:59:39] Balaji: Yeah. And I think this is the part, you know, Moxie Marlinspike is a great engineer, you know, did Signal and so on. He had this post a while back, like about a year ago or so, where he’s like, it feels like all the web three stuff is just trying to put, it’s going to, it’s trying to have the absolute minimum amount of decentralization.
[00:59:54] Balaji: I’m like, you’re absolutely right. Because block space is the key technical constraint of the whole [01:00:00] thing. Just like bandwidth was a key technical constraint of the 2000s, right? So like in the 2000s, the whole name of the game was minimizing bandwidth. That’s why Netflix did Miller DVD, right? That’s why Google did search for a long time, uh, because it was just keyword in 10 links out.
[01:00:15] Balaji: That’s why Amazon had a card, uh, basically a catalog interface, and, you know, AWS came only much later. And it’s because they were minimizing, you know, like, bandwidth use, because the early internet, most people didn’t have T1 connections. Even Facebook, you probably know this story. You know why Facebook started at Harvard?
[01:00:30] Balaji: I don’t. Because, like, people had tried to do social networks before, like six degrees and so on, um, and that didn’t have photos. Because. People didn’t have enough digital cameras, didn’t have high enough ban I mean, think about that time, right? Like, digital cameras couldn’t be taken for granted, right? Yep.
[01:00:45] Balaji: And, most people didn’t even have that many digital camera photos, and you had to have a special camera that would like Go with some like printer cable like thing to your computer said like one one digital camera photo and at Harvard You had just like you talked about like the 50, 000 crypto users as like the starting [01:01:00] base for something Harvard had that group of like in 2004 2003 these relatively elite users that had t1 connections They’re well off enough to have digital cameras.
[01:01:09] Balaji: They were young and early adopters and those the hothouse environment Where like this could take off sooner and then going to colleges, it was not simply a strategic thing. It was like an act of necessity. I mean, even in 2009, like something like only 5 percent of India was online. So like five years before then, you couldn’t have had Facebook in India.
[01:01:29] Balaji: And so, um, I think in the same way, the strategy you have where you get 50, Solana mobile users, Is the same emergent strategy for, you know, starting at Harvard or starting in San Francisco, but rather than starting in a geography, you’re starting to the global community.
[01:01:45] Toly: Getting that like small user base is, is a very counterintuitive, like go to market, I think for like phone companies.
[01:01:52] Toly: And I worked on. The Amazon fire, the Facebook phone, like a whole, and like the windows metro phone [01:02:00] at Qualcomm, we were like, okay, this big company is going to build a smartphone and we’d throw a bunch of engineers at it. And I would be one of those like working on subcomponent, very similar to like that being lucky, being at Harvard to start Facebook, if, if we can get, if we can be that lucky and get the phone to the, the crypto users that are the.
[01:02:19] Toly: The number one, like I would say spenders on the internet and the daily active PKI users. Uh, and if we get enough of them, that’s going to be, I think, a very competitive marketplace for developers. Can devs build applications that are like so high value to users that somebody will decide, okay, I’m buying an SMS enabled phone that because of this application.
[01:02:45] Toly: And if we see that, like we see actually somebody doing that. That means that we actually have product market fit for crypto. The, this idea of that, like PKI and the interaction between PKI and a public blockchain, whatever that data structure [01:03:00] that’s in it’s interacting with is now so valuable that users are willing to like make a.
[01:03:04] Toly: Purchasing decision. I think that that’s going to be like the big unlock and that’s the big unknown. Cause I think while we’re there a bit, unlike NFTs and like crypto trading and stuff like that, I would say in my, in my mind, the only kind of non like more consumer ask use case that we’ve seen over the last year was.
[01:03:23] Toly: Stepin, which is like crypto, like marketplace based fitness app and that had like a breakout, but like not a sustainable one, but like we need, we need to see more attempts at this. Like this is a consumer application. It’s got, it’s using crypto rails and PKI for like user identity and cryptorails for finance and like whatever else, but it has like real value to consumers in the, like the real economy outside of crypto itself.
[01:03:51] Toly: And. And I think, I
[01:03:53] Balaji: think part of that is now enabled by block space being cheap enough that it’s, [01:04:00] uh, you know, it is something where it’s not, it’s not the entire be all and end all to do an on chain transaction. Um, and, and that’s relatively new. That’s like about a year old or so, right? Technologically.
[01:04:10] Balaji: You know, one other thing, by the way, to your point is, uh, there’s this post at, you know, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, Um, it’s, it’s kind of good long form version of an argument that, uh, that I’ve discussed on Twitter. We’ve discussed with him, but it’s basically that, uh, in the longterm, I’m pretty sure that people will log in with their crypto account as opposed to, you know, Facebook, Google, et cetera, for the simple reason that when you log in with.
[01:04:38] Balaji: Um, your Solana name or your Ethereum name, you have a balance, ta da, right? That alone, you know, if you, if you just had 10, 000 users who logged into Facebook and all you’ve got is their profile photo and their name, or you have 10, 000 users that logged in that have non zero balances, and then more than that, they have the whole vector of all NFTs, whatever else they’ve decided to share with the site, right?
[01:04:59] Balaji: Because they’ve [01:05:00] logged in with that particular name. You just like, you know, that’s just a much more economically valuable user. And, uh, you know, one way of thinking about it, you know, you mentioned, you know, 50, 000, 100, 000. Well, do you know, for example, do you know how many people the Cayman Islands has?
[01:05:14] Balaji: Like as a, as a country? It’s only like 68, 000 people. That’s
[01:05:18] Toly: believable. So, Yeah.
[01:05:23] Balaji: That’s kind of interesting, right? Estonia’s only a million people. Like, uh, there’s a lot of countries. One thing I think, you know, uh, it’s a theme of the Network State book and, you know, the pod. Just like there are startups that punch way, way, way above their weight, you know, like WhatsApp had 13 people, I don’t know, 55 people, and Instagram had like 13 people, you know, to get to hundreds of millions of users.
[01:05:41] Balaji: And you guys had 5 people to go and do the salonophone when it would take a huge group of people. Um, there’s small countries that punch way above their weight. And they just do way more with a few tens of thousands of people. I mean, the Cayman Islands is like a globally important thing, right? Um, Estonia is like pretty, pretty important given it’s only like a million people.
[01:05:57] Balaji: So I think actually it’s not [01:06:00] just, I, I, I know you know this and I know you’re saying this, but it’s not just the numerical base. It’s kind of their commitment and influence. I do think that the kinds of applications I would run on the Solana phone, um, you know, one graph that has always stuck with me, have you seen the graph of Facebook’s user base and the smartphone install base?
[01:06:18] Balaji: If you, if you compare both of them, I believe that smartphones actually grew faster than Facebook. Even though you had to pay. Money for smartphones, like as useful as Facebook was and how big has got it, right? Smartphones were so ridiculously valuable to people, right? And so an interesting question will be, um, when would your Cryptophone get so valuable that people would pay for it, you know, what is like the killer app, right?
[01:06:47] Balaji: And it’ll you know, it might be I can I can think of negative models, right? I can think of negative stimuli things where there’s Um, you know, threatened or actual seizure or search of, of things, right? I want to think [01:07:00] of positive things, uh, you know, obviously the crypto economy and tasks and other kinds of things.
[01:07:05] Balaji: Maybe you can earn more within that environment, right? Maybe it’s just like a crypto first environment, you know, like Farcaster, do you know what that is? It’s, it’s like, in a sense, it’s like an Ethereum social network, right? Um, Nostr, though it’s not, it wasn’t necessarily created for this purpose, is like now becoming a Bitcoin social network, right?
[01:07:21] Balaji: It’s got like lightning integration and ordinals. Ordinals are coming in. Bitcoin is actually now much more interesting, again, from a dev sample with Nostr and Ordinals, right? And then there’ll be a Solana network, maybe, among these cryptophones. That seems like an obvious thing to do, just like other Solana users, integrating with other Solana users, right?
[01:07:38] Balaji: So I, so that may be just something where, to participate in the Solana crypto economy, That’s maybe why you, you, you get that phone, right? So maybe you don’t actually lean out of it. You just grow the Solana country, right? Let me, maybe, maybe disagree, maybe you agree with that.
[01:07:53] Toly: I, I mostly agree with that. I think, I think we, we will see like, these kind [01:08:00] of like PKI interactions like start happening with applications that are like start, they’ll, basically like the complexity of like dealing what chain you’re dealing with, where the data structures are, how the users log in.
[01:08:12] Toly: These are starting to get more and more kind of melting away with libraries and developer tooling and like the PKI aspect of it, like for a user’s perspective, like if they have this phone, this, the, what the phone does is it secures a seed phrase, but then the wallets themselves can connect to Ethereum, can connect to Solana, can connect to Bitcoin and whatever, and it’s all based in the same seed phrase.
[01:08:37] Toly: And that’s where your root identity is generating from kind of going back to that VeriSign world. That secure element inside the device are the armed guards for the certificate. And then you end up with like infinite, yeah,
[01:08:54] Balaji: better than armed guards because
[01:08:55] Toly: encryption is that can be on any network and, um, [01:09:00] the tooling and like the.
[01:09:02] Toly: Being able to kind of like move value and like move maybe like key points of like identity between them, that stuff is going to get better and simpler across, like, you know, it’s just time progresses like on all the major networks. We’re kind of seeing some of that. Like you see like major wallets, like Phantom start supporting multiple chains.
[01:09:24] Toly: But do it in a way that everything is derived from a single identity root and you have accounts that can be shared between chains and like NFTs that are associated with one, but you can then send proofs like for smart, for like applications that are kind of becoming more agnostic. Like I own this NFT, it may be on Solana, but it’s a DGOD and DGODs are doing stuff on Ethereum and stuff like that.
[01:09:47] Balaji: So I’m really glad that you’re Thinking cross chain and multi chain eventually, uh, even, I obviously have to ship and so on, but that actually, I think, you know, like things like neon EVM compatibility. I think that’s really good and [01:10:00] positive. Some, you know, I have, uh, so I have a couple other things I want to talk to you about with the phone.
[01:10:06] Balaji: May one is just on like the logistics of building it. Right. Cyanogen, a long time ago, they tried to fork Android and they got, I mean, they’re, they’re pretty good technically, uh, strong technically, but, um, they just, uh, didn’t have like the business model for it at the time. And, uh, then a bunch of Chinese manufacturers successfully forked Android and, uh, like Xiaomi being, you know, one of them.
[01:10:27] Balaji: Uh, and then one of the things Google did is it sort of pushed more and more stuff into Google services to kind of discourage that. So Android was initially very open and everyone like free ride on Google and Google’s like, Wait a second, I didn’t, I didn’t really want you to do that. And so more and more stuff became like Google services where they, they didn’t close Android source, but they closed the state of the application and pushed it server side.
[01:10:46] Balaji: Um, and the third was Facebook using Android for Oculus Quest. And of course, you know, Facebook’s a big competitor of Google, but they, they found that that operating system. Was a mobile OS that they could fork and do something with. Um, So, you [01:11:00] know, I don’t know, I mean, like the, uh, it’s interesting because iOS in a, in a sense is also built on BSC, but they have not made it anywhere near, uh, not even anywhere near, they have not made it publicly forkable at all.
[01:11:11] Balaji: How did, you know, how did you guys think about that in terms of those other three that I mentioned, like Cyanogen and the Chinese manufacturers and, uh, and Facebook are all competing with Google, but you’re, you took a more cooperative approach. Am I correct about
[01:11:24] Toly: that? And this is for practical reasons. I use both my Saga device and my iPhone.
[01:11:30] Toly: But even before I had the SawGuy, I was already using Google Maps and Gmail and, and Chrome. So like, like those are pretty decent services and I kind of know that they’re taking my data, but that that’s just like competitively, I think it’s really, really hard to build products that are as good, but it’s getting closer.
[01:11:50] Toly: You start seeing like things like HiveMapper builds an open data set for mapping. Uh, and that’s definitely going to end up being competitive to Google Maps [01:12:00] eventually. I use Signal like more often than I use iMessage. Like my iMessage is filled up with spam from ran Oh yeah. From randos. Uh, my Signal I actually use to talk to like my friends and family on a day to day basis.
[01:12:14] Toly: Um, so like that, that stuff was already actually happening at some point. We’ll get to parity to where you have encryption first, like privacy, preserving applications that have the exact same properties as the Google applications. And then it won’t really matter if the Google stuff is there anymore. And for now, my guess is that like the GMS licensing.
[01:12:41] Toly: And OEM licensing is set up in a way due to, uh, the Google mobile services. I don’t know if it’s set up because Google doesn’t want people to fork it out because now they feel like there are, there are applications that have parity or because they are afraid of antitrust. But for either reason, I think it makes sense for them to be allow for [01:13:00] somebody like a small startup, take their entire service set and build up like add ons.
[01:13:05] Toly: It like, because the alternative is that like you start getting enough shots on goal that We would replace all their services with like competitive products. And eventually one of those will hit, right? Like if not salon, then like out of the, you know, every year has got to be like 10 or 20 people trying to, trying to do this.
[01:13:24] Toly: So,
[01:13:25] Balaji: yeah. So, so skiff is actually, um, I’m not, I’m not sure if they have salon support, they have a theorem support, but they may add salon support at some point. Skiff is worth looking at if you haven’t seen it, which is like doing a Google docs and Gmail. With crypto login, right? And maybe they could add Solana login and, uh, They’re doing basically provable intent encryption and so on, right?
[01:13:44] Balaji: So, uh, something like that is a good example of what you’re talking about. Where you basically have drop ins for each thing. And, uh, one of the reasons that people might want this is, You know, Google is starting to do things where, um, You know, it’s, it’s, uh, this was, this was just [01:14:00] a demo, But it was posted on Twitter, you may have seen this, Where a landlord was being auto completed as being offensive to somebody.
[01:14:06] Balaji: Um, and, you know, just like, I don’t know, there’s stuff like that where you just might not agree with what Google is doing server side, they’re, they’re now just sort of showing, it’s almost like, you know, Big Brother can see you kind of thing, right? Have you ever heard of T Bow sorting? Do you know what that is?
[01:14:22] Balaji: Charles T Bow? Okay, so all the way back in 2013, um, this is actually part of a bunch of tweets, I had nuked a bunch of tweets a long time ago, but I tweeted something like, um, The long term consequence of the smartphone will be to enable T Bow sorting. And there’s this economist named Charles T Bow, if you google T I E B O U T, right?
[01:14:41] Balaji: There’s something called the T Bow model, okay? This is a purely theoretical model in, like, the 1950s. And he had seven assumptions in the TVAO model. It was something like, everybody has perfect information, and they can, um, you know, we now say remote work. They can work from anywhere, [01:15:00] uh, and they have mobility, and services in one location don’t spill over to the other.
[01:15:05] Balaji: So for example, if a group in location A is funding services, then only those people who paid for the service can have access to it, and group B can’t, and so on. So it like, it limits the free rider problem. And so it’s over. But he had a bunch of assumptions, which are all very unrealistic in the world of the 1950s.
[01:15:19] Balaji: What’s happening is the internet is making all of those happen and, uh, the Solana phone or the crypto phone more generally is making them happen even further still, you know? So it’s a, that’s something, you know, just like, you know, you, um, the verifiable delay function was something that you’d sort of built something and you weren’t aware that it kind of intersected with some piece of literature.
[01:15:40] Balaji: If you haven’t seen that, so we’re taking a look at the T Bow model, um, and seeing how the crypto phone really statuses a bunch of those assumptions. Go ahead. Do you
[01:15:48] Toly: think that, like, because the, the phone has been built kind of like by these big tech companies, do you think it’s going to become so, uh, like, the [01:16:00] consumer differentiator between any device is going to get so small that, like, it’s a commodity product and then crypto can be that differentiator that, like, has a viral flip?
[01:16:10] Toly: Like, do you
[01:16:13] Balaji: I, I think so, right? Because, I mean, smartphones Uh, they don’t get that much better each year, you know, like, uh, it is true that your iOS, iOS will get slower. People don’t know if that’s a real thing or not. Right. But assuming, I mean, there’s a, there’s a, and you, you know what I’m talking about, right?
[01:16:30] Balaji: Like if you don’t, if you don’t upgrade your iOS, um, or you don’t upgrade your
[01:16:34] Toly: phone, I don’t think it’s malicious. I think it’s just the devs are building on the NextGen device. And they’re using all the hardware of the NextGen device for their like testing. And then when you ported it, when you install that code on an old gen device, you’re like, Hey, why is it sluggish?
[01:16:53] Balaji: Right, exactly. So, so it’s, it’s like, you know, there’s that saying like, uh, whatever in the, in the nineties people said, whatever Intel creates, [01:17:00] Microsoft destroys, like meaning Microsoft eat all the RAM, we’re going to eat all the CPU or whatever. Right. But actually, you know, what’s interesting about it is, uh, you know, you’re a low level guy.
[01:17:10] Balaji: Do you know this app a long time ago called WordLens? It came out in the early 2000s, uh, early 2010s, late 2000s. Okay, this is a Brazilian guy, um, you know, a friendly acquaintance of mine, who coded this. And it was basically able to do real time, uh, OCR and translation of words. So you could hold it up and a sign in Spanish would, in an augmented reality way, translate into English, okay?
[01:17:34] Balaji: And this was in like the early 2010s with very limited cameras. I think it was even pre deep learning, okay? And, uh, it was just incredibly impressive low level engineering where he just coded everything in Objective C at a very low level. And it kind of showed, I mean, he was, he was like one guy basically and he was ahead of like all of Google’s ML people or whatever, right?
[01:17:59] Balaji: I eventually got acquired by [01:18:00] Google, well deserved acquisition. And, uh, and that kind of showed what could happen if you took your kind of skills or his kind of skills and actually did push the metal, you know, which hasn’t been done. And the reason I think about that is in the event, you know, people are talking about some conflict over Taiwan or what have you.
[01:18:19] Balaji: And hopefully that does not happen. Right. But in the event that there is some sort of sustained supply chain disruption, we might be stuck with the current model of phones for a long time, you know, or at least years, right?
[01:18:31] Toly: Basically 10X improvements in software across any application you can think of.
[01:18:36] Toly: Like you can, like the way things are built right now with react and all of these, like. Very fast to market stacks, uh, it’s not optimized for performance. It’s really optimized for speed of getting to market. And that makes sense when you have like a, a very rich environment, like richer resources, if we needed to push the envelope on like performance, uh, there’s just [01:19:00] like, you know, 10 to a hundred X of optimizations people can start doing in software.
[01:19:06] Toly: I’m very confident of that. And you can see that in gaming consoles, like especially the Switch or PlayStation, those have a very long life cycle, and it’s the same chip for like 10 years, you know, on the PS2 or whatever, and you saw the quality of games get better and better as engineers really Figured out the nooks and crannies of the hardware and then the libraries and then the skulls scaled out, but that takes a very large Concentrated effort of continuously working on the same thing
[01:19:37] Balaji: It is actually amazing when you go back and play some of the old NES games or something like that how?
[01:19:43] Balaji: Constrained those engineers were in their debugging and shipping and yet how good many of those games are I you know people have talked about obviously starting new companies new communities new, uh, new currencies for a while. Uh, and I’ve got this whole book on starting new countries, right? [01:20:00] And I’d love to hear your thoughts.
[01:20:01] Balaji: Obviously, you know, crypto provides a lot of the material for that. It provides currency, it provides identity. It actually provides something underestimated, which is community of a bunch of people around the world. Um, you know, what, what does, uh, Tholiland look like? Tholi, Tholia. You know, um, what, what, what is, if you can start a new country, uh, you start a new currency.
[01:20:21] Balaji: So if you start a new country, what, what does, what does that look like? What, what would you do with that? The state
[01:20:25] Toly: is the, is the oppressive part. So I would start a non country, like free association. Anyone can participate in it if they want to. And whatever they bring in, bring into it is, is like, is up to them.
[01:20:39] Toly: Right. Like, and I think. That’s a very natural state of living because the absence when in like communities, like, I mean, that was like the, the state of living in prehistoric times, like everyone lived in some community and like everyone contracted, you had ideas of private property, but you also have this shared.[01:21:00]
[01:21:00] Toly: thing that you are working on. Um, and you kind of see that naturally happening with the internet, open source communities, people build licenses and stuff like that to create some rules of enforcement, but they’re not. I don’t have to use a GPL, though it’s forcing me to be a GPL only person. I can then go code Apache licensing or whatever.
[01:21:23] Toly: And you kind of free associate with communities based on whatever common social rules and standards they’ve created. And that to me is like a very natural state, and this is what like, Totally Land would be like. We all have our PKI, if we want to coordinate on ideas. Uh, it’d be open source idea as we would like, if we want to pull funds together to go do a certain thing, we could use a DAO, you know, and like, you know, set up a group and dialect where we like have these like smart messages to tell everyone, Hey, go sign this.
[01:21:53] Toly: Transaction and like tell the, and like activate the DAO to go like pay for like, you know, market [01:22:00] marketing campaign for this open source product or whatever that we built, whatever, whatever it is, the thing that, that we worked on together, and that’s like a very kind of like transparent free flow of like, uh, association environment.
[01:22:13] Toly: Um, can that thing scale? I believe that can actually scale to a very large number of people. Cause this is the natural state of people. Can it be like competitive to a nation state is like I think a question, like there’s one set of anarchists, like back in the French revolution that their whole theory was that, like, we just ignore the state and we do this like small commune association thing.
[01:22:41] Toly: And eventually, if enough people do it, the state just goes away because we just, we just ignore it, like, and then it like disappears. And that’s a very kind of utopian, nonpractical thing, point of view, but it feels like very. Very attractive to believe that that’s possible. [01:23:00]
[01:23:01] Balaji: If you can get enough people from the state to believe it doesn’t exist, then it doesn’t exist.
[01:23:06] Balaji: Like, the Soviet Union doesn’t exist, right? So, there’s an aspect of that which sounds utopian, but there’s also an aspect of that which is It’s, it’s actually getting, I mean, that’s the thing I think a lot of people don’t get about crypto. Crypto is a software install, but it’s also a, like, uh, a mental install.
[01:23:23] Balaji: It’s a wetware install. Because, um, it’s not just the bytes on a hard drive or a phone that matter, but it’s the concepts in people’s heads that value those bytes and, and treat them in a certain way, right? And, um, so, you know, there’s, there’s the dumb version of it, which is, oh, let’s all just, you know, close our eyes and pretend the government doesn’t exist.
[01:23:45] Balaji: But there’s the smart version of it, which is governments have been convinced to not exist, right? East Germany does not exist, you know, like the Soviet Union does not exist. These oppressive governments were basically convinced in a sense to not exist. Now, of course, the counter counter [01:24:00] argument is they were convinced in part at gunpoint, you know, there was the USA and so on.
[01:24:04] Balaji: And I’m not a utopian about that. Um, certainly that was a very important component of the whole thing. But ultimately, at least a big part of that revolution was a peaceful one internally where people did change their minds, right? So at least part of it, I’m not saying all of it, but at least part of it is the thing you’re talking about where you, you build essentially a parallel society.
[01:24:26] Balaji: It shows a better way than the legacy society. And people decide to opt out enough and then they just kind of withdraw support for the old thing and maybe the old thing doesn’t have the legitimacy to even enforce on the new one. Maybe that’s
[01:24:36] Toly: some reality that like This kind of already exists, and in Western, like, liberal democracies, it mostly exists too, like, people naturally, like, have, so far, like, we have more or less the freedom to kind of limit the actions of the state, and like, slowly but surely, we kind of, like, do that, and for the most part, like, you know, um, [01:25:00] In our kid’s school, people get together and pool, ask money together and like, go build something right for the school.
[01:25:05] Toly: And there’s no, like, the state is not involved, right? And like, yeah, it’s a network, right? We pool, it’s a network. We work, people work on open source software together randomly, right? Like random people, they put in work, they get releases out, they coordinate. And the state is more or less not involved there outside of like the, like, you know, uh, small copyright things, right?
[01:25:29] Toly: Like I finish what you’re saying and then I’ll, then I’ll put more stuff. Crypto as like, you know, tooling land, right? Like is, is like that happening at a faster pace and accelerating at a bigger scale? And like once. This layer of services that’s effectively provided through these like networks starts eating away at other services, like some are commercial, right?
[01:25:49] Toly: It like now open source software is competing with Microsoft built operating systems, right? Are not as interesting anymore as Linux. So there’s like places where this stuff is clearly one. [01:26:00] As the system happens, I think the, the other parts will naturally recede and kind of how you described and eventually may only exist in name only.
[01:26:09] Toly: And that is like, I think what crypto and like this idea of an open database with PKI like coordinate, not just on ideas anymore, but now on like financial assets. I think that’s the really, really exciting part.
[01:26:22] Balaji: Some people up, watching up to this, this segment will think, Oh my God, these guys are so impractical, blah, blah.
[01:26:27] Balaji: But to put, to put some flesh on the bones of this and just to give it a little more practical, or not, not to give more practical, but an angle from which it’s maybe perceived as more practical. The concept of the network state, the network is the digital part. The state is like the physical part. That’s one way of cutting it.
[01:26:41] Balaji: And I think people aren’t really thinking through the extent to which the network can actually act as quote enforcer, but nonviolent enforcer of good behavior. So for example, with crypto, the network can enforce property rights and identity and do birth and marriage and death certificates, and it can, um, you [01:27:00] know, have registries of who owns what, and then how does it enforce?
[01:27:03] Balaji: Well, uh, We already know that wealth can be represented on chain, but it’s a little, you know, maybe less obvious is, uh, for example, how did you get into your, um, office today? Right? Did you use a key card? Right? Okay. It’s a key card to get into the office. For many people, they have, um, like a Tesla style fob to get into their car.
[01:27:24] Balaji: Um, many pieces of, you know, obviously you need like some sort of, uh, Um, cryptography to get into your phone, whether it’s face ID or, you know, passcode or something like that. Uh, and many pieces of equipment, so, so buildings and pieces of capital equipment will basically all be gated by your phone or a fob or a card or some consolidation of all those that looks like a phone.
[01:27:47] Balaji: So that means that all financial assets, because all stocks, bonds, all that type of stuff, all property, digital property, and all capital goods, right, all the way in electronics and stuff, Which is maybe [01:28:00] if you add all that up, right? Because all real estate can be gated by a lock on the door, right? All electronics, all vehicles, every crane, every ship, every plane, every car.
[01:28:12] Balaji: All capital goods and all, um, property can be put on chain and guarded by that. And in theory, maybe you could say, oh, I’m going to go and hotwire a Tesla. Well, maybe, but that’s pretty hard to do. I’m going to hotwire and get into this, you know, 30 story building. I’m going to climb up the elevator because I don’t have the crypto key.
[01:28:30] Balaji: Maybe, but it’s pretty hard to do, right? As a practical barrier, um, if you have ENS or SNS based authentication to those systems, you can protect buildings and cars and phones and electronics, as well as all the on chain property. That’s probably like 90 percent or more of the value of property in the world.
[01:28:50] Balaji: It’s not going to protect the clothes on your back. It’s not going to protect the literal, you know, Uh, a glass of water that you’re holding, but modulo those things that are [01:29:00] too de minimis in value to have electronic locks on, everything else can be protected in this way. If it’s done with due process, see, deplatforming today is done in this arbitrary corporate way.
[01:29:11] Balaji: But you can imagine a world where deplatforming is more humane than violence. It’s something where, okay, you’re suspended, you’re this, you’re that, you’re locked out of this, locked out of that, that’s your punishment. You can still, you know, like, go to work or whatever, but, you know, the actual punishment is you can’t go to the club or something like that.
[01:29:32] Balaji: I don’t know. Um, and, uh, and right now it’s being wheeled in this totally arbitrary fashion, but there’s a forum in which, with due process and with societal approval around it, A digital punishment could be more humane than a physical punishment, potentially. And I’m not saying it totally replaces it. I’m not saying, Oh, you know, some people say, Oh my God, do you think there’s no need for violence ever?
[01:29:54] Balaji: No, I’m a, I’m a realist on that. I’m trying to think of a good example of this. You know, if, [01:30:00] um, if you have an issue where there’s, there’s somebody who’s, uh, the people do this in buildings already, right? To get up to the 30th floor, you, you might need a key card to get in there. That’s an aspect of security there.
[01:30:11] Balaji: We’re just going to. Discourages folks from mentoring. It makes the space more secure. It’s digitally secure. You don’t, you need a guard because the key card is, is, is stopping it. The guard would actually have to be intimidating at the front door, but this would actually block it into four you different slus and gates.
[01:30:24] Balaji: So anyway, uh, just some thoughts, but you know, actually it’s not totally land. You know what we should call it on it? An all Anatolia isn’t that good, right? So Anatolia, the national currency, Solana, everybody’s a volunteeristic. Uh, I’m glad you’re thinking about this because I actually think that all the Solana phone holders could maybe crowdfund some territory, maybe do something interesting there.
[01:30:48] Balaji: I, I do think that the crypto phone is an important thing for civil rights and property rights, and so I’m going to try it out simply on the basis I don’t have any stake in it or anything like that. Um, I, you know, I hold [01:31:00] Solana, I hold Ethereum, I hold Bitcoin, I hold all this other stuff. The, the Cryptophone is something which I, I believed in even before you guys announced and I’ve talked about it for a year or so.
[01:31:09] Balaji: I’m glad, really glad you’re doing it. So thank you for that. All right. And so with that, thank you very much, Anatoly.
[01:31:14] Toly: Thank you for having me.