Web3 Galaxy Brain 🌌🧠

Nicholas: Welcome to Web3 Galaxy Brain. My name is Nicholas. Each week, I sit down with some of the brightest people building Web3 to talk about what they're working on right now. On today's episode, I'm joined by Julien Genestu, founder of Unlock Protocol. Unlock is a protocol for memberships as time-bound NFTs. Creators can monetize their work with Unlock by deploying sovereign membership contracts that allow for trials, partial refunds, and membership trading, in addition to the various superpowers granted by the ERC-721 ecosystem, such as airdrops, token gating, and governance. On this episode, Julien and I discuss the attention economics at the heart of advertising-driven Web2 services and social media. Julien weighs in on Nostr, BlueSky, Farcaster, and Mastodon, to which he contributed in the early days. We talk about the lessons he learned about open communication protocols and monetization in his prior startup, Superfeeder, which generates RSS feeds at scale and was acquired by Medium. Finally, we discuss how Unlock Protocol synthesizes Julien's career experience into a sophisticated paid membership system that enables creators of all stripes to layer crypto subscription memberships into their revenue models. It was great having this wide-ranging conversation with Julien and to learn about the Internet-native economic theory underlying Unlock Protocol. As always, this show is provided for entertainment and education purposes only and does not constitute financial advice or any form of endorsement or suggestion. Crypto is risky, and you alone are responsible for doing your research and making your own decisions. This episode of Web3 Galaxy Brain is brought to you by RainbowKit, the best way to connect a wallet in your Web3 project. RainbowKit is built for developers and designed for everyone. Try it today at RainbowKit.com. If you'd like to sponsor the show, visit Web3GalaxyBrain.com for details. I hope you enjoy the show. Hey, I think I heard you. I think it's working.

Julien Genestoux: Yes, I'm in. What a mess.

Nicholas: All right. I guess it was my phone. I really don't understand, honestly.

Julien Genestoux: I mean, it's Twitter, you know, these days.

Nicholas: Yeah, I'm happy it's working. Did you see the thing the other day? he was streaming on Periscope? Like, it looked terrible. I just caught it the other day. Yeah, Musk. He's like, I guess they disabled Periscope. I don't know. I didn't even know that they had killed Periscope fully.

Julien Genestoux: Oh yeah, they killed it, I think, two years ago. But it wasn't, I mean, I guess if it was streaming through Periscope, it was not fully, fully killed.

Nicholas: Yeah, I think, you know, he got some of the engineers still stuck working there on H1Bs to figure out how to get the Periscope shit working.

Julien Genestoux: Yeah, yeah. Oh, well.

Nicholas: It's a mess. It's a mess. Do you mess around with other, like, Twitter alternatives at all? Is there anything that you're interested in?

Julien Genestoux: Yeah, definitely. I think I've tried them all, actually. I mean, I've been on Musk for, you know, not quite 10 years, but like since very early days. I was actually part of W3C, the World Wide Web Consortium. They had a social web effort, and I co-authored a few specs. And I was in a group that actually wrote ActivityPub, which is the core protocol behind Mastodon. So I did play on Mastodon for a while, then left it. That's awesome. And played again with it recently. Honestly, I'm underwhelmed by Mastodon at this point. I did play with Nostr, which I find interesting, even though very Bitcoin maniac, kind of crypto maniac. And recently, honestly, I've been playing with BlueSky. And I'm actually pretty excited about BlueSky. I think they're doing a lot of the right things, the right choices. I did play with Farcaster, which I think is cool as well, but very niche. It's like, it's, you know, about 10,000 people, not more.

Nicholas: What is it that makes you think BlueSky is more relevant than Farcaster? I'm curious. I've really only played with Farcaster of all the ones you mentioned.

Julien Genestoux: Yeah. So Farcaster is cool. I actually don't know much about the technical details about how it works. I didn't read the protocol of the WARP. Actually, Farcaster is a protocol and Warpcast is the app. But my understanding was like, it's still a fairly basic messaging system with a single host, basically, that you go to.

Nicholas: I think the way it works is like, there's nodes. I think there's two things that are interesting about it. I'll explain it quickly, but then I'm curious what makes you think BlueSky is more relevant. But I think it's like node software that keeps about 10,000 messages. I think everyone currently duplicates the entire state, like this buffer of the last 10k messages. And the history is just lost unless someone builds an archive node. But the other piece is that it's somehow like they're creating a new wallet for you, but the identity that you're logging in with is Ethereum-based, like EVM wallet on main net. Yes.

Julien Genestoux: So that's what I like about it. It does use private keys, which is cool. That's it. So I was actually invited early on by Dan and I don't mean to dismiss it. I was invited in January of last year, I think. And then I dropped because there was nobody there. It's very small amount of people, like 10 people that I knew and not a lot of messages. And so I moved on. And then when I started to reactivate it a couple of weeks ago, months ago, it was like, oh, well, no, your account is lost. Which is like, fuck. I mean, if my account is lost, even though I did have the seed phrase, so I was still able to take a But I felt like, well, I mean, so I was disappointed. But then the bigger thing is like, there's not a lot of people there that I know. A few people from the community, the Ethereum community, so that's cool.

Nicholas: I feel like that's its core demographic right now. Ethereum devs and adjacent.

Julien Genestoux: Yes, absolutely. And I mean, that's Dan's background, which makes sense. So maybe that's the thing that I'm not worried about, but like, feel frustrated. What I like about it, it does integrate NFTs as Postgres, as you said, it's like wallets-based, which I think is cool. I hope it moves beyond the Ethereum crowd. But if anything, it stays there. It's not a bad thing. About BlueSky, so I actually like very much the decoupling between identity. I mean, I use my domain name, for example, and the servers that use it. Even though it is a federated model, and I don't think Federation is a good model anymore. I used to think that basically 10 years ago, it was like, no, Federation is the way. And I think now, no, Federation is not the way to go for many different reasons. BlueSky's model is interesting in a way that it decouples the data and the identity from the servers on which you publish. And yet it is not kind of a purely kind of FML-like model that Project Forecaster does. And it's also not as peer-to-peer as Nuster. Because for example, Nuster, I don't know if you've used that, but...

Nicholas: No, tell me about it.

Julien Genestoux: It's very strange. So first, it's all private keys and public keys. So it's hard to even find people unless they give you their public key. How do I find you there? I don't know your public key. And then it's basically peer-to-peer. So you basically connect to other networks. When you launch the app, it connects to a bunch of peers, and then it fetches data that might be stored on there for you. So there's kind of this always kind of eventual consistency thing, where you launch the thing and you don't see nothing. And then after a couple to appear in a random order, which is always going to be distressing. Like, wait, what am I supposed to read here? Is this the fresh stuff? So it's the Nuster. I think it's a lot more peer-to-peer. This approach is, even though there are servers, but they're never storing anything. So they go online to get stuff and messages bounce around. Again, at this point, I've played with them all. What I like about BlueSky, I guess, the most is it feels very much like Twitter. So no need to learn weird things. Again, you would do on Nuster. It works on the web, it works on mobile phones. I'm a big web guy. So I'm always looking for stuff that works on the web. But at the same time, it is or it seems to have some bunch of sound technical decisions around how it's built with the decoupling of identities and the servers on which you post. But this is all theoretical at this point. The identity is decoupled, but the federation doesn't exist yet. So I hope they create this.

Nicholas: And in all of these cases, is BlueSky like a private key model? Or how does it manage access?

Julien Genestoux: No. So right now, it doesn't rely on private keys, but it does have the IDs. So you could imagine that at some point, you might create other authentication methods based on what server you're going to. You don't connect with your what if that's what you're asking. Right now, you connect with a password, and you can create multiple app passwords. But the servers then use your repo and your identity, your DAD basically as a way to identify you beyond that.

Nicholas: Okay, so there's still like a centralization around the accounts in the current BlueSky?

Julien Genestoux: Yes, but on a given node, basically, right? So think of it like as right now, there's just one node, which is why it's this. But I can definitely imagine a world in which there would be at some point, another node on which they decide that authentication is not happening through passwords, but through private keys.

Nicholas: Okay, but you said that the nodes and the identity and the data on the nodes are all decoupled.

Julien Genestoux: Yes, you would still authenticate with your node. And so basically, that's how you would access your repo. So each user has, it's called a repo. It's almost like a Git repo. I mean, I think it's not Git, but it's a Merkle tree. So basically, you add messages, it rebalances the tree every time. And the only way you can write to the tree is because you have the matching DAD for this. So I assume, and again, I haven't actually implemented things, that the way you prove that you own a DAD is different based on different servers. So you could imagine that at some point, I would, and you can attach multiple identities to a DAD. So I could definitely imagine a world in which I first log in with the password, I add authentication by a private key, or I say, hey, this other identifier is able to modify the repo for that user, because I'm logged in with the password. So I can say, I can indicate that I'm okay with this other DAD modifying repo. So that then when I log in from that second server, this time with my wallet, I can modify the repo and take over the account. Does that make sense?

Nicholas: Yeah, yeah, you can add accounts to the ownership of the DAD. I don't actually know how DAD works, but is it some secret data that like is the...

Julien Genestoux: So DAD is just the identifier. I think it's more in the logic of the node itself. that would basically say, okay, I accept this new piece of content, because instead of the repo is marked as somebody who can write to that repo. Does that make sense?

Nicholas: It does. There's still something a little bit missing for me between how if two nodes have different sensibility of different parameters for what counts as authentication for themselves. Yeah. Are they looking at each other's history in order to decide whether or not it's permissible to add new forms of authentication? Yeah.

Julien Genestoux: So I think it's inside of the repo itself. So it's not the server. It's like the repo for the user includes the identifier that is allowed to modify itself. Right?

Nicholas: So it's not tightly coupled to a specific kind of authentication because...

Julien Genestoux: That's my...

Nicholas: The repo is where the data... And the repo has to be replicated to a new node.

Julien Genestoux: Yes. So basically you basically in pull with the, I think, main architect. behind this, I think there's something interesting that I never thought about. But yes, basically the repo is where your data is stored. It's all public data. Right. By the way. So there's no encryption or anything. Oh, I see. And so when you connect, you're basically fetching your repo in that way and then updating it up there. And what I found really interesting in his approach is like, they expect at the beginning, people to just create their account from, again, the main BlueSky or the first BlueSky app and then to bring it somewhere else rather than start from somewhere else. Which I think is interesting because it makes discovery then much easier in that context.

Nicholas: Can you take me through why it makes discovery easier?

Julien Genestoux: Because then that server knows... I mean, you can find multiple logs from that server, basically.

Nicholas: I see. Okay. Okay. I think what gets me about Farke... It feels like maybe if there is... Like most people don't give a crap about decentralization. That's right. As much as people complain about it, I don't think they care enough about the free speech, scare quotes, issue to leave. Like, it's not more... They care more about the network than they do about that issue. Yes. For sure. Because they care about exposure. Yes. So it feels to me like part of the... If there is going to be some kind of new social media, of course, there will be new social media that are like AI-centric, equivalent to what TikTok was for short form video, but for these new modalities of creation. And that maybe will be upsetting to existing social networks. And then on the other side, it feels like Farkesters bet... I don't know if they made the exact right technical choices, but like MetaMask is pretty popular. And it does feel like one substantial difference from existing networks and new potential networks is ownership of the keys or in whatever respect. Yes. Not specifically the EVM model, but it does kind of make sense that... Like what's crazy to me about EVM and MetaMask, I mean, plus others, but at least from my perspective, the popularity of this wallet format, pretty crazy. I mean, people talk about it relative to PGP and stuff.

Julien Genestoux: Yeah. I mean, it's funny because like...

Nicholas: It's very popular. Yeah.

Julien Genestoux: Go ahead. Sorry.

Nicholas: No, no. That's always good.

Julien Genestoux: I mean, a good friend of mine, when I started to look at... I mean, creating on lock-in, we talked a lot about crypto and blockchain. He told me like, you know, if this thing is... It was 2018. So kind of a while back now. He told me like, well, it looks like if... Bitcoin at the time and Ethereum and all that. The one thing that he would have done is that it would have become the biggest use of probably private key cryptography and would have made it really obvious for lots of people that password are not the way to go to authenticate yourself on the web. I think we're definitely moving away from passwords as the default on many things. I do think that, yes, there's going to be some, you know, private key cryptography. But I also think we're starting to see the emergence of other systems like, you know, MPC and threshold signing, which are also, I think, in my mind, very interesting. And I hope that during that we move away from the model of like. your identity is tied to a specific server, kind of my Google account or my Twitter account, but more something that is decoupled from any kind of application. And I can use these kind of things to authenticate your application, whether it is private key cryptography or even, you know, kind of MPC signing, basically wallets to authenticate onto applications.

Nicholas: Is Shamir a type of MPC?

Julien Genestoux: I don't think Shamir is a type of MPC. Okay.

Nicholas: MPC is, but it's a similar principle or not at all?

Julien Genestoux: I don't think it's the same. And again, I'm not a cryptographer, so I would say things. but Shamir, if I'm correct, is basically you share a secret between multiple entities, but you don't necessarily have the ability to sign from that secret. So basically, in the context of if I'm correct, I'm definitely not a cryptographer. What's interesting about MPC is basically, if you think of it like it's splitting the computation of something between different nodes, so that if you have enough of these nodes to provide you their piece of the computation, you can reconstruct the thing. So basically signing a message, you never actually access the private key. You just have all of these different nodes sign the message in their version of the signature. And it's the combined signature itself that actually gives the full transaction in the context of a blockchain. Which means that you don't need to reconstruct the secret ever in that context.

Nicholas: Right. So it's more like the way NoSysSafe functions in terms of user experience.

Julien Genestoux: Yes. But it's off-chain NoSysSafe in that way.

Nicholas: Off-chain. I mean, I guess NoSysSafe is kind of off-chain in a way. But you're saying because it's to verify the authenticity of something without necessarily putting it on a blockchain.

Julien Genestoux: Exactly. Exactly.

Nicholas: Interesting. Yeah. And also, I don't, I frankly have not fully understood it, but account abstraction seems to be kind of a maybe even counter trend in this respect. Like against more towards maybe sort of MPC, but at least against having like individuals with private keys.

Julien Genestoux: Yes. Yes. But at the same time, so the problem with account abstraction is like it puts the account inside of the chain, even though the account can be a smart contract or a private key or any of that. It puts basically, it removes the idea that accounts are, you know, it can be anything basically. It's abstracted away, but it still puts them on chain. And what I like about MPC, even though it is at this point, I think still fairly theoretical how decentralized it could be, is that it actually puts it above the chain. A simple example, like account abstraction, you know, that's the kind of thing that Arjun has been doing for a couple of years now. We all have seen, or even like NoSysSafe, we all have seen people send funds to an address that they think is the right one, but in the wrong chain. And these funds are lost or worse, they can be stolen. What happened to, I think it's the winter mute hack from last year. Yeah.

Nicholas: They didn't have an initialized safe or something on the chain at which they received the OP.

Julien Genestoux: Exactly. And then basically somebody deployed, I think, a couple of hundred safes because, you know, the addresses are kind of sequential, deterministically sequential, so you can get there. Deployed a bunch of them to a point where they're like, cool, okay, now we have the right one and we're going to take them, take the funds. Which was very smart, but shows, highlights to me the challenge of putting accounts inside of blockchain rather than accounts above blockchain. It's kind of hard to describe, but like outside of that system.

Nicholas: Off-chain. But I mean, at the same time, like the constituent private key cryptography at the core of EVM is kind of, I mean, not on top of, but underneath, if anything, like it's outside, the private keys exist outside of the chain. Yes. You have to play inside of the chain.

Julien Genestoux: From my perspective, private keys are, I don't say better, it's not the right word, but like are like, so do you think private keys is basically one single thing off-chain that allows me to execute things on-chain and that is independent of the chain that I want to use? so I can write to Polygon or to Optimism or to Mainnet? It's the same. I could even sign a message and send it by email to you. It's fine. What's a kind of section?

Nicholas: You can even have password. You can have separate passphrases associated with separate seeds. And I feel like the wallets don't at all, at least the Ethereum ecosystem wallets, the UI based ones that I play with most of the time, don't emphasize at all the like derivation possibilities, like having trees of derivations. I haven't seen any UI for EVM that really flexes that. Maybe more of the CLI tools are better for that. I actually don't know about that. It seems like a whole huge, huge exciting part of what private keys can do is like totally unavailable in all the wallets that I use.

Julien Genestoux: Yeah, definitely. But the challenge with these private keys is they get lost all the time. I mean, they get leaked all the time. It's a limitation. And so for me, smart wallets, I'd say, I can have structured wallets like Argent and GoSafe, provide an interesting mechanism to have basically, you know, more guardrails and recovery and multiple signers for security and then two multiple factor auth, which is basically a multi-sig at that point, which are interesting. The challenge though is like they are on-chain, which means that you lose the ability to do that across multiple ecosystems or multiple environments. in a way that would work. And again, in theory, MPC from that perspective, for me, represents the best of both worlds because it is off-chain, but it would in theory be programmable. So imagine a scenario where, you know, you have your, instead of having a C-trace, even just your main desktop and your phone both have a share. It's not a share, but like our signers on an MPC secrets, you could, you know, and you would need to remember password or any of this, you could just, you know, send a transaction from your desktop and then your wallet client would say, hey, you need to first approve it on the phone. The phone vibrates, you approve it, and now the transaction goes through and you don't have to remember C-traces or any of that. And this cannot leak because there's actually no such thing as a private key here.

Nicholas: Okay, that's the part that I don't understand. So you're saying it's not functioning like each device has its own private key and there's a threshold that you have to reach. It's not the same. It's not like that.

Julien Genestoux: No, I mean, they all have their own private key, but it's not your private key. It's basically they're all doing a part of the job of signing the transaction.

Nicholas: But is that not like signing a NOSIS safe transaction and then only one person is actually signing the transaction?

Julien Genestoux: So, the signature is, I mean, either the, I guess, on-chain. So you basically have, you know, a smart contract that has, okay, X, Y, and Z to both, to all sign this. And this doesn't get executed unless X, Y, and Z have signed. Or maybe it's a signature, but that's a signature that is stored on the safe servers, right? The safe, I mean, the company behind the safe provides that. But here it's different in the sense that you don't actually need to store any of these things on a centralized piece of content or piece of infrastructure. It's like, it's multiple parts compute or run a function to compute the signature. And it's the combination of these things all together that is the actual transaction to be sent.

Nicholas: Right. Okay. Versus in the safe land, it's doing, it's an execute transaction on NOSIS. safe is easy recovering the signed messages from each private key. But is that not? Yes. Is it not tantamount?

Julien Genestoux: No, it's not. It's not how it works. No, no, no, it doesn't. There's no, because there's no private key, so you don't actually have that mechanism. I mean, it's the combination of these things that represents the message that is easy recoverable by the chain into a user, but the private key itself is never exposed.

Nicholas: Okay. But is it? All right. I don't mean to belabor this too much. I'm just trying to understand, but the, in the safe world, you know, the signers of a safe don't reveal their private keys. They do have private keys, but in this, it sounds like you have private keys, but in the multi-party computation, they do have some kind of secret that they're using to know. they don't have an avenue, but how do you know? Why can't anyone run the computation?

Julien Genestoux: Yeah. So basically the, the MPCs each have their own logic to verify, you know, that the user is the right one. So for example, you can have something where it's all off space. And that's actually, you know, there's this company called Taurus or what's called Taurus, now it's called Web3Auth. They use Google. So basically each of the MPC nodes will just verify that the signature that you provide is a good OAuth signature. And if they do, they will run their share of the signature generation if you want, in a way that if you combine all of these things together, that's how you end up with, you know, a transaction signed by a user.

Nicholas: So you could sign with Gmail, essentially.

Julien Genestoux: You sign with Gmail or basically, yeah.

Nicholas: I could have an MPC setup that has expectations of three of my friends with their Gmail all signed. Yes.

Julien Genestoux: Exactly.

Nicholas: So it's more agnostic to the implementation of the...

Julien Genestoux: Yes, exactly.

Nicholas: DID, I guess, or I don't know what you would say.

Julien Genestoux: Yeah, the identity itself. Exactly. It's like, and so going back to the phone, you know, you put in WebAuthn or PASKYs now, which is getting more and more popular on mobile devices or, you know, on, I mean, even like your, you know, on iOS, they can, the biometrics thing. These are stuff that would basically be used as a verification that, yes, I trust this is Nicholas. I will do my share of the computation to generate the signed transaction that Nicholas wants to send.

Nicholas: So, okay. But so this is all coming from this idea that, like, it seems to me like, it has the same problem, but it seems to me like people having a new relationship to login as a... Maybe as a, hey, I already am an Ethereum degen or have a BSC wallet or whatever. Maybe this is sort of the first step towards owning my relationship to the open protocols that I interact with or signing, just signing stuff more generally. But it does have the same problem that the other, that just changing up the decentralization story of a network has, which is for most people, they don't really care about that as a value proposition in and of itself. Yeah, I agree.

Julien Genestoux: I mean, for me, there's always two aspects. There's the censorship resistance, which is the thing that we say, oh, you know, like, I don't want my stuff to be captured. And I think most people don't care. And honestly, I don't care about this because I theoretically care. But I don't think it's a problem right now for most people. Where I think it's more challenging is around the permissionless innovation. Like decentralization allows people to build things that they don't need to ask permission for. And when you think about stuff like, you know, the web, it is permissionless in that way. Like you don't need to ask anyone's permission to create a website. You can just go buy a domain and then put that on some hosts and there's no permission to be asked. And that is, for me, more powerful than the app store model where every time I do an app, I have to ask Apple or Google's permission to release this to my own customer. And so it's not about the censorship like, oh, Apple will not let me publish something that says bad things about, you know, Steve Jobs. I don't think they would actually block you for releasing an app that says bad things about Steve Jobs. But I do think they're limiting what can be done because by definition, they only approve the stuff that they have anticipated could be done.

Nicholas: It is paradoxical though, because they do seem to have a monopoly over what's possible in native software in the world. And that sucks.

Julien Genestoux: Exactly. And that sucks because it sucks not because they will just block your app that does something that might compete with. It sucks because they think they've defined whatever is possible. And I don't think it's the case. I think it's impossible for anyone to come up with all of the rules. So basically, it's kind of forbidden by default in the ecosystem, right? Like it's only the stuff that they've allowed. Anything else, well, cannot exist because they don't have, they haven't taken into account that it could exist. But the web is different than this. So you can, if you can create a new website with some random new things in there and then push it out there, you do not need to ask someone's permission about how it works or to make it behave in ways that was not expected.

Nicholas: I don't remember who said it first, but I think they did, Apple in particular, did strike a pretty genius. Like in this second era of Apple, they figured out that they could have the dangerous experimental open web inside of a single app and limit what that app can do. And they can benefit from having access to all of what's on the web while limiting. I mean, now apparently they've shipped notifications to Safari on iOS. I haven't seen it in person.

Julien Genestoux: I haven't seen it in person either.

Nicholas: But you know, like it's been impossible to do anything. Obviously you can't create something equivalent to a native app in the web on mobile devices for years. And even back to like original iMac, Bondi Blue, whatever, Internet Explorer or whatever. I think it was Internet Explorer at the time.

Julien Genestoux: The first, yeah, the first browser for Mac OS X was in the Explorer, which is kind of funny.

Nicholas: I remember. And then they killed it with Safari at some point. But I remember that era.

Julien Genestoux: So the funny thing is, you know, Explorer was always done by Microsoft anyway. And Microsoft invested in Apple in like 96 or 97 and kind of saved it. I think they put $200 million in there and that was kind of the thing that saved Apple altogether. What's funny, I mean, not funny, but like the Safari, basically the team that built it was actually coming from Camino, which was a fork of Gecko, which was a fork of, I was going to say Firefox, but it was before Firefox. But then, ah, it's not communicator, it's the other one. But anyway, so basically there was this Gecko engine that was forked by... Navigator? Navigator, yeah, sorry. That was forked by these two guys. One of them was Mark Pinkinson, Parkinson, I don't remember the name. And then basically, it's funny because I was actually working at the time on creating a web browser for my internship. And I was creating a web browser, one of these kiosks web browsers. So like, imagine like a station or some place where they would have a computer and... And so I did use the fork that they had from Camino, which was the web browser at the time. And I had an issue compiling some stuff. So I messaged them and they were responding very quickly. And then all of a sudden, they did not respond at all. And I was like, what's going on? And then six months after, I see Apple announced Safari. And I know that these two guys were there after this. And the Camino project, I think, kind of died after this, which is kind of funny. So, but yeah, no, it's... So Apple, basically, they do control some of the really important things, right? They control the payments. And that's actually the problem with Twitter right now. They will take their 30% tax no matter what. And will not allow you to bypass this easily. Control this. They control a bunch of APIs on the phone as well. Like in the end, for privacy reasons, right? But like a bunch of authentication mechanisms. So the web is actually a worse citizen on iOS than the mobile apps. Like than native apps, like for example, Facebook can do a lot more tracking inside of the app than it can inside of the web. Because Apple blocks all of these third parties and all of that mess, making it much harder for things to monetize on the web. And my big belief is that that's the way you kill the web without actually telling you to kill the web. You make it impossible for people to monetize these things by saying, Hey, you want to make money? You need to create an app. You need to create a native app. And I think that's been the strategy there. Same, they did a bunch of other things like background running. So even though on the web, you could actually do some stuff like, you know, service workers and things like this to have background jobs running, they make it impossible on the web on mobile, which makes it really hard. But you can do these things with native apps. So I think they basically, you know, I don't want to say crippling, and I don't think it's the right word to say this, but they're making it much harder when you create apps to have things that are eventually really good to make your app successful and push you down their native app flow, which is permissioned. So it's, yeah, you can do all of these great things. But in the end, we'll have to put the timestamp, the approval stamp on it every time you need to do something, which is not ideal.

Nicholas: Right. But so, which brings us to crypto, because crypto is one of those things that gets people to install new apps and use new websites and spend a lot more time in their browser and do all kinds of, I mean, there are some others, you know, YouTube remains, but crypto is one that really gets people onto new kinds of websites and installing new extensions and buying hardware and all this stuff. So it seems, but I think the threshold there is that there's this speculative opportunity, perceived speculative opportunity in getting on board on Bitcoin or whatever, EVM or others. And it's hard for, you know, like for Twitter blue, it's like, you know, you're not going to sideload F-Droid on your iOS just to pay Elon $3 less for some shit you didn't want. It's tough. But OK, hold on. I want to get to unlock protocol before. Please give me a second. I need 30 seconds for today's ad. This episode of Web3 Galaxy Brain is brought to you by RainbowKit, the best way to connect a wallet in your Web3 project. RainbowKit is built for developers and designed for everyone. Try it today at rainbowkit.com. My thanks to RainbowKit for supporting this episode. If you'd like to support an episode, go to juicebox.money slash at Web3 Galaxy Brain. You can buy an NFT and get an ad like this and talk to this audience. OK, so I want to talk about unlock protocol. And it feels related because the challenge to for creators or communities, all the people who is that the problem that unlock is solving? It's kind of it feels to me like the neighborhood of Patreon and Gumroad has something to say about crypto. OK, absolutely.

Julien Genestoux: So basically it comes from a realization. So let me give you a bit of background. So big open Web pack forever struggled over the years with the idea that if we want the Web to be strong, we need to have money on the Web. There is no world in which the Web can survive. It doesn't have a business model that is actually scalable. And right now, the only business model that we built on the Web is ads. And it's scalable for a very small number of parties like Google, Apple. I mean, actually not so much Apple, but Google, Facebook of the world. Everybody else, Amazon actually emerged last year as something that could be plaguing. Everybody else has been struggling with smaller and smaller crumbs. At the same time, I have seen first hand. So I sold my previous company to Medium, a blockchain platform, and I worked for Medium for a couple of years. And I've seen...

Nicholas: Yeah, can you just explain briefly what was Super or is?

Julien Genestoux: Superfeeder, yes, absolutely. So it is because it still exists. And it's pretty cool. It's an API for RSS feeds, either for publishing RSS feeds or consuming RSS feeds. And usually people ask what are RSS feeds. So RSS stands for really simple syndication. It is basically a standard format that can be used to represent many, many different kinds of data from many, many different websites that allow easier consumption. So think a website is great. It's actually very hard for a machine to consume it and make sense of it. It's easy for a machine to display it, but it's hard for a machine to make sense of it. Back when there was no AI, again, 20 years ago, 15 years ago, one of the good ways was like, okay, let's create feeds of data that always have a title, a description, a bunch of other attributes like a published date and author. And that would apply to news, obviously, but that could apply to social networks, that would apply to ecommerce as well. We could put all of these products in. By normalizing the feeds, you made it really easy for reader applications. Feed readers was a thing at the time. If you're listening to this as a podcast, I wish I think it's going to get eventually this. It is also RSS feed powered. The podcasts are all the same format, a description, an attachment, which is the piece of audio that we're listening to, a title, and maybe some images, an author, obviously you, etc. So RSS has been a thing for a while. Superfeeder built APIs to both publish in real time. So basically, if you had a news story that you wanted to publish as fast as possible, you could use something like Superfeeder. And then consuming on the other side wasn't really painful because the way you would normally consume RSS feed was to do some polling, which is basically the kid in the backseat saying, are we there yet? Are we there yet? Are we there yet? Every five minutes to check or every two seconds to check whether the feed has been updated. So if I wanted to know whether there's a new breaking news, well actually way for the breaking news, I would query CNN every two seconds. Hey, is there something new? Is there something new? Is there something new? Which doesn't scale very well, as people probably guess, and is incredibly wasteful. So we built these APIs that would allow people to either publish in real time, but also consume in real time and say, hey, stay where you are, we'll tell you when there's a breaking news. And we would do the tough job of actually polling or finding other mechanisms to identify when there was something new. And then as soon as there's anything new, we would send the information down the line to the subscribers.

Nicholas: RSS push.

Julien Genestoux: Oh, go ahead.

Nicholas: RSS push. RSS push. On that side for the... Exactly. That's very cool. Okay, my big question before we move on from Superfeeder is just, what is the primary use case for RSS today? Obviously, podcasts are very successful, but are there in doubt? Like there was a minute where Safari for Mac OS had RSS reading built into it, and you could get Twitter feeds and everything was an RSS feed. And then eventually they realized that for proprietary networks, it wasn't profitable to provide everything as RSS feeds. Maybe you can still get YouTube or something, but most...

Julien Genestoux: So at this point, it is very limited, to be honest. It is still used in some context. So actually search...

Nicholas: In back end stuff, maybe?

Julien Genestoux: Some back end, but search actually indexes a ton of data. So if you're a search engine at Google, and I haven't spoken with them in a while about this, but when I was a Superfeeder, I was doing Superfeeder, we spoke with a large search engine pretty regularly. They're kind of all struggling with the same thing. Like, we need to index data. So somebody actually... The first time I was in New York, it was just before the US Open, the tennis tournament. And they told me the thing was like, okay, we're basically a week before the US Open. Everybody's searching US Open stuff right now. That's basically very spiky, right? People look for this the week before, they're doing the two weeks of the tournament, and then maybe a couple days after this. And the average indexing time was about two weeks. So basically, right now, people would search, they would find stuff from last year, and they would find stuff from last year, pretty much the whole tournament. And then when the end of the tournament happens, well, now they start having fresh content, but nobody cares about this anymore. Because it's actually a pain to index stuff. Like if you're a search engine, you have to go crawl a lot of websites all the time to find new stuff. And relying on RSS actually made it easier for them to identify much faster when there was something new in this.

Nicholas: So Google News is powered by, and even other, but essentially, like the new stuff is powered by Google indexing RSS.

Julien Genestoux: But not just the new stuff, not just the new stuff, actually the actual search stuff. And we did some tests where we were... Oh, because it's easier to index. Yes, because it's easier to index, or easier to detect whenever something has been updated and you want to crawl it, which...

Nicholas: So it's like low-key semantic web.

Julien Genestoux: Exactly, low-key semantic web with real-time key communities built in to allow for faster detection of updates. I don't know that they're still doing this, but when I was at Medium, they were definitely getting a lot of the Medium content through that system as well. So it can have been real-time. So you would do a search and you'd find something that was published like two minutes ago. And you're like, how is that possible? Well, because they were told. Otherwise, they can't really index all of Medium every two minutes.

Nicholas: Okay, so it's interesting. So it's SEO advantageous to have RSS today. And then obviously podcasts. Although I don't have a good sense of how many podcasts there are, but like how much download... I guess there's quite a bit. It's a little bit weird too, though, because at a point like Spotify is not RSS.

Julien Genestoux: So it is actually. So it's interesting. So it is. I mean, it's not just, but it is. So you can, if you create your content somewhere else, you can actually ingest or feed it literally into Spotify via an RSS feed. And what they don't have is like outgoing RSS feed. So you can't really subscribe to...

Nicholas: But also the user pages are not populated by RSS. No, no. They maybe have one Spotify agent.

Julien Genestoux: Oh, I should clarify. In many, many cases, the pages are actually not served from RSS feed. It's very rare that the reader itself is... So today the popular feed reader, I think called Feedly, for example, they're basically aggregating stuff on the backend and then just serving this on their apps or on their content from their backend. It's not directly from the RSS feeds.

Nicholas: Right. This is probably going to come up in all of the decentralized networks. It's just not. It doesn't turn out that it's worth it. Okay, so wait, but maybe we can bring this back to Unlock.

Julien Genestoux: So the challenge that I found is basically this open web that I love, it's going to struggle if it only monetizes with ads because there is no way you can actually build a scalable ad system in a decentralized way. I mean, the basic example, right? Like if I have a billion dollars spending ads, it's so much easier to go to Facebook and say, okay, here's a billion dollars, give me all of the eyeballs than to go to a thousand or 10,000 small sites and say, hey, here's your share of that billion dollars, give me the eyeballs. So basically, as long as we're going to have attention as the only way to monetize anything online, it's going to be a struggle for smaller, indie, small platforms to monetize because they're always going to be at a disadvantage versus the big ones that are able to sell to advertisers, you know, economies of scale, basically. Right?

Nicholas: Yes, but one question on this. Like 10 years ago, it came out that all of Facebook's analytics were lies and they were significantly exaggerating the exposure that they were getting to people who were purchasing ads.

Julien Genestoux: Yes, yes.

Nicholas: And I feel like we just sort of let that new cycle go.

Julien Genestoux: Yeah, but it's still the same, right?

Nicholas: They didn't fix it, did they?

Julien Genestoux: It's still the same. I mean, I don't know if you've been to Facebook recently. It's a nightmare. It's just ads. I think it's also one of the reasons why it was easier because it's easier to lie a thing that is centralized. Like, you know, decentralized network can't lie. Like when you say, oh, these crazy numbers of transactions, well, I can go inspect, oh, I see it's not true, or I see it's just fake transactions. And so when we see, you know, I don't want to name names, but like blockchains announcing crazy numbers, it's actually very easy to see, oh, that's not true. When it's a closed environment like a Facebook thing, it's so much easier to say, oh, yes, there's going to be 10 billion people seeing your ads. It's like, sure, okay, well, I guess I'll trust you because there's no way to verify this. So again, the...

Nicholas: On a philosophical level, what's interesting, though, is that it seems like people want to believe the lies because they could easily be measuring the conversion rates for their own websites.

Julien Genestoux: But that's not how it works. Whoever sold the ad campaign doesn't want to tell their brand, hey, we got screwed. They're going to say, you know what, let's do it again. You know, it's, again, it's... I think the challenge about these attention economies are all over the place, and that's definitely one of these. But when it was at Medium, so Medium tried to do ads, results were abysmal. There was no way Medium could actually make money with ads unless you would completely degrade the reading experience. And it was always kind of the founder's vision to have a very pure and clean interface. And so that's when we started working on that paywall. And at the time, I don't think it's still very popular. People are upset at this. But I do think that was the right move for Medium. What we looked at was stuff like New York Times. New York Times was crossing the 1 billion subscriber. Now they're at 10, but they were crossing the 1 billion subscriber threshold. We're looking at stuff like Patreon, like Kickstarter, and say, okay, actually, it's not that people don't want to pay. It's that you need to make it easy to pay for stuff and you need to actually create value for them. And so Medium did this membership. And honestly, I was very surprised that in the first day, the first 24 hours, we had almost 10,000 people pay $5 a month, which in the end, it adds up a lot. Like it's $50,000 a day if that stayed on that same curve.

Nicholas: And those were people who had never experienced the token gate.

Julien Genestoux: No, no.

Nicholas: Exactly. Like the subscription. Exactly.

Julien Genestoux: And so that was the best thing about this. They paid and then we looked at the analytics. And it's like none of them was actually paying after they were facing the token gating. Basically, there was so few stories behind the payroll that there was no way any of them would actually pay because of that. And so when we asked them, like, hey, why do you pay? Thanks, keep doing it. But like why? The answer was consistently the same. It's like, oh, I care about the mission. I care about the vision that Medium has. I care about the logo. I care about this. I spend my days doing this. It makes sense for me to pay for this. And also, I want the green halo around my avatar. I don't know if you know, but on Medium, paying members have a little green thing. And so basically, people were like, this is a status symbol. And to me, that's when it clicked. It's like, okay, the web is actually just about the status symbols. And we're on Twitter space right now. It's just about, you know, who has it.

Nicholas: In the comments specifically, is where they surface for experience?

Julien Genestoux: Every time there's your avatar, so your picture, there would be a green halo around it. And you know that it's a Medium member.

Nicholas: But most of the subscribers are, well, on Medium, are they authors also? Or there was a notion at the beginning that all comments are also kind of posts or something?

Julien Genestoux: So yes, but it was in the beginning. I don't know what it is right now. I don't think it was actually, no, I don't think there was actually a correlation between what kind of person that was. It wasn't as much readers than writers than commenters. That just lurkers people. Oh, interesting.

Nicholas: Okay.

Julien Genestoux: I mean, that's what I remember. It's now five years, so it's kind of far. But what was really interesting is like the idea of like status. People pay to be able to be considered differently and treated differently inside of the app. And so when we message them, that's what they will tell us consistently. And I realized like, well, actually the web is just about these statuses. Like it's even more true in crypto where we use, you know, our PFP, our NFT is just PFP. Like, look, I'm rich, I've got a punk or, you know. And so I realized like, oh, okay, that thing, that membership is like, I'm a member of the cool group of people that can do XYZ. That is the business model. That's what we should have been paying for. And again, if you go back 20 plus years, first time you connected to the web, bulletin board systems, you know, there was these tags or these people that were like the fancy guys that could access some channels that you wouldn't be able to access. You realize like, oh, if it was easier at the time to pay, you know, $5 to access some channels, maybe you would have done it. And maybe it would have been a very different web today versus what it is.

Nicholas: I remember hearing stuff, I think via Rui Ma or maybe A16Z, but about how in China for the better part of a decade, subscription content has been very popular there because they have apps that let you like WeChat pay.

Julien Genestoux: Yeah, exactly. In China, it is a lot more.

Nicholas: Very popular like private subscription podcasts. It's not like it's a lot.

Julien Genestoux: Exactly. And so one of the things that I think has been hard in the West is the idea that the web arrives. So actually two things, two anecdotes. There is an HTTP protocol, HTTP status 402. So everybody knows 401, you know, a bunch of status codes, 44, that's the national account. There's an HTTP status called payment required from 93. So you can say, oh, we have this. Except that if you look at the spec for that payment required status, it still says reserved for later use. And I do think that's the original scene. Like it was too complicated at the time to build payment natively inside of the protocol. So people then pushed it to later. It still hasn't been done in a way that forced monetization at the application layer. And that's the banner ads. We're born in 97. That's what happened. It's basically saying, okay, the monetization cannot happen at the protocol because we don't have this. So it's going to happen at the application layer. And what you said about in China, a bit of a trip with the places is like, so then we got used to basically this free web with ads everywhere, right? But there's a bunch of places where the web arrived later or became more prevalent much later. And in these areas, there was no preconception that everything had to be free on the web because everything was free for us on the web in the late 90s and early 2000s. And so for them, it's actually much easier to pay. I do think that we're also moving into a point where it's different. And that's again, maybe thanks to the apps. At some point, king.com, the candy crush maker, they are approved. The average revenue per user was more than 20 bucks per month, which is kind of insane. Nobody's doing this. And so I think when you put these numbers, you realize that, well, actually, people are fine paying for stuff. It just needs to be smooth. It needs to be for something that they care about. It needs to be done in a way that doesn't feel like they're going to get mugged in the streets because they've paid for something once online, right? Honestly, I remember the first time I bought stuff online, I felt dirty. There was like, no, don't put your credit card number on the website. I don't know if you're the same age that I am, but it was for the longest time, something that people were very worried about.

Nicholas: Yeah, I remember when I was very young, people being worried.

Julien Genestoux: Exactly. And in practice, there was weird worries because first, it actually didn't happen that often. And honestly, it's riskier to pay at a shitty gas station in a town that you've never been than it is to pay for anything online in practice.

Nicholas: Yeah, because all the information required to steal your credit card is on the physical.

Julien Genestoux: Exactly right. So it's like, well, but anyway, so...

Nicholas: Hold on, but there is one little thing there, which is that I think we spend far too much time, like in the reality of credit life, modern consumer behavior is that we offload all KYC and all risk...

Julien Genestoux: To the banks, yeah.

Nicholas: Actuarial calculations...

Julien Genestoux: And we all pay 2% for that.

Nicholas: Financial functioning, it's just Visa, credit card, PayPal, whatever they're all doing. They're just able to borrow money cheaper than the amount of theft, and then they have the ban hammer. that's somewhat...

Julien Genestoux: Oh, actually, I think that... That's it. Maybe there's that, but I actually do think we all pay 3% tax on everything because of that. Like, if you think about the striped charges, like 2.9% plus 30 cents, the real cost is the 30 cents. The 2.9% doesn't make any sense. Like, there's no reason why transferring $10, and we're talking about like, you know, digital dollars, $10 is more expensive than transferring $100. And so the 2.9% is just the insurance. It's just like, basically, the cost that we all bear for transferring, you know, for having these risky transfers of money. The fixed cost is the real cost.

Nicholas: Absolutely. You talked about how the significance of the kind of flex social game of internet, at least in the West, at least thus far, and the amount of money that people were making on... What was the app that you mentioned that was making $20?

Julien Genestoux: It's Candy Crush. It's King.com. It's the company that owned Candy Crush.

Nicholas: Yeah, right, right, right. Which is just a mind-numbing game, in a good way, I guess. It reminds me of, you know, Nikita Beer's company, that Gas sold to Dissords?

Julien Genestoux: Yes, I do see that.

Nicholas: I doubt you've seen it, because I think probably almost nobody has seen it.

Julien Genestoux: Talk about it.

Nicholas: It's a YouTube video. Have you seen the app? The Gas app? There's like one YouTube video of a middle-aged dude who somehow got access, because you need to be a high schooler to use Gas. Or I guess it's probably defunct at this point. But basically, I think it's interesting for people who care about creating new social networks and opportunities like this. It's basically questions about your classmates. You have to register. They launched it at a couple high schools in Georgia, and very, very super focused on specific schools. They had launched... I mean, this is a little bit the Kremlinology of the thing, so maybe some of this is exaggerated, but they would launch at a specific high school and station people outside the high school in the days in advance of the launch. They would follow kids from the high school on Instagram with a private Gas account. And then on the day of the launch, they accept all of the friend requests for the account, launch at the high school, and time the launch by having people sitting in a car outside the high school watching for when kids get out of class, which I guess seems like a little bit apocryphal, because I don't know, don't kids all get out of class at the same time? But anyway, really hyping up the popularity of the thing at this like kind of 10x Facebook pincer move thing. And they were able to convince kids... The actual app is students filling out quizzes of like almost like face mash, but like four options. And it's like, who is the most whatever word? And then you have to pick amongst four options. And then it notifies the person that they've been voted for as the most applicable to some sliding scale of horniness. Some of them are very innocuous, and some of them are a little more flirty. And there's like a feed of, you know, someone in 10th grade voted for you. And you can pay, I think it's $10 a week, or even maybe $30 a week. There's a few levels, but incredibly expensive subscription to be able to reveal the first name and first letter initial of the people who voted for you. And also to have some other like clues and hints, a limited number per time interval. So it is actually possible to get people to pay a ton of money because I think they were quite popular if what they want is relevant.

Julien Genestoux: You said a ton of money, and it feels like a ton of money. But one thing that I keep thinking about is like, you know, the time spent versus money, right? So, you know, the ratio of like how much money you're spending versus time. And I think a lot of stuff that we do on the web is the worst. It's like, basically, we're spending our lives there and we're paying cents of cents. But you go see a movie, and you're paying like literally $15 for an hour and a half. There's nobody who would expect to pay an hour and a half for $15 on Facebook or any of that. So we have basically something that is incredibly important. And after a year and a half of lockdowns in the US, it's pretty obvious that without internet, we'd be killing each other. And yet we're paying nothing for this. I get the point. I'd say it's a lot of money. But in practice, if you bring that back to the amount of stuff that we pay for other things that we don't care about, it is outrageous how little we pay for.

Nicholas: Stripe takes 3%, but Facebook takes 97% of your time. It's free, but what you get is a kind of quality that's like, I mean, TikTok, if you just look at it, I love it. But it's just the most evolved in terms of manipulating your behavioral psychology for keeping your engagement by packing jolts into less than 10-second clips.

Julien Genestoux: Not only it takes your time, but I think it's even stealing more of your time. They optimize for the amount of time that you spend on this. And it's not even like, I want to spend an hour and I would spend this much. You spend an hour and they actually give you an hour and a half or take an hour and a half from you because you don't realize this thing that you're watching is actually not really that you're adding anyway.

Nicholas: Yeah, it's trouble because you want to focus on the ones that are valuable amongst the thing. But in all of these experiences, it does feel like the majority of the time is not actually well spent. I would rather that they be manipulating my psychology into doing something I actually want to do.

Julien Genestoux: Yes, useful. Yes. And it's hard. Some of the most research that I've seen on this is about YouTube and how YouTube optimized algorithms to get you to spend literally seconds more per day, but that would actually turn into tens of millions of dollars, which is kind of insane. And it's like these small dummy things eventually are what I think not only make us dumb, but also starve the thing that are pretty smarter, more interesting, and not just being built to steal your attention.

Nicholas: Well, I think they steal all of the productivity gains of the last however many years.

Julien Genestoux: That's a theory, yes.

Nicholas: It seems pretty obvious that the entertainment has just gotten more attractive and we don't make more use of the superpowers that we have. So, okay, unlock protocol.

Julien Genestoux: And the worst thing is that we don't actually enjoy this. We don't seem to be happier than we were 20 years ago or 10 years ago. And yet we spend lots more time on leisure activities. So it's kind of a challenge. It's kind of a challenge that... And so my realization to think about unlock is basically that so instead of doing attention as the business model, maybe there's a world in which we use memberships. Like I want to be part of a group of cool people. I'm going to pay to be part of that group. And this is what I'm going to get in return. And that group of people could be fans of a brand, fans of a musician, or even just my group of friends. And we're going to have... We'll let it out. We'll group online. Each page, spend over a month, and we can do certain things together. That's the idea behind unlock. And think about all of your, obviously, social web, but beyond this. It's all about this kind of, oh, I pay to be a bit different. Like if you think about an app, right? Like a game, you pay to unlock a level. So you're part of the group of people that can play level two. Or you play... Or if it's a CD of the Blue Checkmark, I am definitely a payer for Twitter. I spend my days there. It just makes sense for me to pay for this. I'm not necessarily a huge fan of Musk these days, to be honest. But I do think I'm getting value from this. So it makes sense for me to pay for that. And I think by paying for this, I am also... And I don't know if it's true, but like buying for myself the opportunity to just move on from this, because I know that's what I'm paying for. And if I don't pay for this, then I don't need to spend more time there. Does that make sense?

Nicholas: Yeah. More or less. More or less. I mean, I agree with you. It's also kind of interesting how paying for Twitter... I think paying for Twitter in particular is... You know, he made some promises. Half the things weren't there. Half the things don't seem to still be there. For sure.

Julien Genestoux: The details are awful. And I think it could actually be a lot smarter than actually... So instead of paying... I mean, again, I have to pay for Twitter. But instead of paying for Twitter, maybe what I would love to do is actually pay for the people that I care about. So let's say I really think you're... doing some great stuff here. Maybe Twitter should offer you the opportunity to kind of... I say, token date some tweets and say, hey, these tweets are not going to be for everyone. They're just for my biggest fans. And then every time somebody becomes one of your biggest fans... To be honest, it's something that Twitter is working on, has worked on, shut down, and is bringing back. The idea of super followers, then I would pay $2 a month to actually read your content. And Twitter could take a cut of this. I think it would be fair in that world to do that. So as long as... That's the big difference between what Twitter is doing and what we're doing. As long as the membership status is not stuck on the Twitter platform. So right now, if I pay for you on Twitter, well, it's only on Twitter. Which means that if tomorrow you start publishing content on Tumblr or on Instagram, well, I cannot pay for you there because you don't have... The membership is coupled with the platform. But in an ideal world, and we should talk about this next week, there is decoupling between these memberships and the platforms. And so you could definitely build a world in which, hey, I bring my smart contract to Twitter and say, Hey, Twitter, I am Julian, here's my contract. Don't show some of my content. And I will tell you which content you need to show and not show to people based on whether they have one of my... If they don't, don't show it to them, show them a purchase button. If they do, great, show them the content. When you show them a purchase button, I am actually cool with you taking a cut because you are selling memberships on my behalf.

Nicholas: Right, you're the retailer.

Julien Genestoux: Exactly. And you're kind of the absolute leader of all. that is not... You're still providing distribution, but you're not locking me there, which means that if I'm unhappy with posting content on Twitter, I can go on WordPress, post my content there, and then have my same membership contract so people that are paying for me on Twitter can come read these stories. And same way, if they do a better job, if WordPress does a better job distributing my content, I will do it more fun there because I'm getting more subscribers through them.

Nicholas: In the world of Unlock, are the contracts that a person deploys in order to create membership tokens sovereign contracts or is it a platform?

Julien Genestoux: Yes, they're completely sovereign contracts. So it's your contract. We have no control over this. Once you deploy, you deploy from a factory contract, but then it's yours, fully yours. Nobody has any control over this. You can do a bunch of things. So Patrick on my team loves to say they're updatable. So you can change terms, so you can change the duration, you can change the price, you can change the currency. You can even revoke someone. Like, hey, I don't like you to be my member, I will kick you out. It's possible because that's what a membership is in practice. They're upgradable. So it means that whenever a new version of the protocol exists, you can opt in to support that new version. And they even are programmable with the concept of hooks that allow people to modify the behavior of the contract on certain things. So you can say, hey, when somebody buys, give them another NFT or refund them 10% because they've done X, Y, Z or any of that.

Nicholas: I've even heard you talk about being able to resell a subscription and deduct a certain percentage of the duration as a fee for transferring.

Julien Genestoux: Yes. So it's actually a very good thing. Like, I could have a discussion about royalties, which I don't think is a real thing. But we do think that there is actually value in saying, hey, if you transfer your membership to somebody else, there should be a cost to this. And the cost that we associate is actually time. So it's like if I have a month-long subscription to Netflix, when I transfer my Netflix subscription to you, maybe Netflix has said, hey, you can transfer things. But every time you do this, we burn 10% of the time that's left on the membership. So if I transfer 30 days to you, you only receive 27 days. And that's actually monetizing the security market because that means that you're going to renew earlier than if I had not transferred.

Nicholas: Am I right that you also have like a notion of trials in the contract?

Julien Genestoux: Yes. That is a very touchy one because that allows for a lot of abuse if you don't have the right things in place. But yes, we built... So think of the OnBlock contract, it's a membership-based contract. It is actually in ESG71 with a ton of features on top of it. One of the features is the ability to do trials. So you can say you can get a refund. So actually, before trials, we have a mechanism for refunds. So you can say, hey, I allow my users to take a refund when they want to cancel their NFT. So the way that you can cancel your G membership, you could get a refund. I don't know if that's actually Gems to use, but sometimes they do. For Rada, exactly. So basically, if you cancel halfway through the month, you will get half the money that you spent on that month unless the creator has set different rules. And they can set two types of rules. They can set a penalty, which is basically the minimum amount that you'll lose no matter what. So if you purchase and cancel right away, you might lose 10%, which would be a normal penalty. That's actually the default that we have. And then they can also do the opposite of a penalty, which is a free trial, say, hey, during for some time, maybe for a day, anybody who's purchased and canceled, they get the full refund. So basically, you have the trial period, the penalty, which are both offsets, basically, and then the pro Rada.

Nicholas: Very cool. Two things. First of all, do you think that the... I mean, for instance, what we were talking about just earlier, I mean, it's very unlikely that Twitter is going to integrate token gated tweets using Unlock Protocol, of course.

Julien Genestoux: Why not, though? I mean, why not, though? Yes, I can imagine that it's not likely today. I don't know. But down the line...

Nicholas: Sorry, what? Incentives, I said.

Julien Genestoux: Well, what if they realize, which I think is the case, that eventually they could actually sell more by having... It's the same discussion between the web or the native apps. It's like, of course, you could make more... You have more control when you're building an app, but you also lose a lot of distribution. And so maybe by giving power to users, they could, in the end, capture more from more people. I think it's really kind of a... I actually don't think... Sorry, I actually do think, I'm convinced that... but that we will see big social platforms eventually adopt some kind of token gating mechanism, because that's actually maybe a better business opportunity for them than ads are right now. Especially in a context where ads are harder and harder to do, where there's a bunch of legal issues around this, around privacy and stuff, and where they are actually less and less viable, like, CPMs have been dropping for pretty much 20 years, with a few exceptions during COVID. So, frankly, I do think that eventually... And that was kind of the case. Instagram was about to launch some token gated feature. So I do think that we'll see that. Yeah, so you could connect your wallet to Instagram. I still think it's in the app. That team has been let go. And I know that one of the next things that we're trying to do is say, okay, what if there is stories on Instagram that are only visible to people that have a good Internet?

Nicholas: Okay, they were going to use like 721 as the basis of the authentication of the token gating?

Julien Genestoux: Yes, 721 and balance off. And you don't try to do the same balance off function as the...

Nicholas: Fascinating. I was listening to a show from last year. I think it was Web 3 Deep Dive, something like that. Okay. They were doing a special on unlock protocol. And they mentioned this thing. I don't know if it came from unlock or came from them, but status, access, power and stuff.

Julien Genestoux: Yep.

Nicholas: Is that something from unlock or that they developed?

Julien Genestoux: I don't know of that specific one. But yeah, we can do that kind of thing with unlock. So let's take an example on my blog. And I'm not going to share the blog here. It's not advertising, but basically, there's a membership. And it's an NFT based membership, obviously. And you can only access if you have that NFT. What's interesting, I could do that on Substack, right? Here, I'm actually doing it on my own site, which reduces the reliance on third parties like platforms. But what's more interesting is that that membership exists in the Ether. It is not tied to a specific application or backend. And we're not going to be doing a call here, but I use a virtual webcam. And I should say hello to Catherine here from a company called Galactic. And Galactic has this cool virtual webcam that allows you to overlay stuff from a programmatic view. And what I do is basically I pull the list of subscribers on my blog and overlay them on any of my posts. And so that's going to be a benefit that I offer to my members that is not tied to the blog itself. And so it's really kind of status. Now, hey, you can show around. I don't think it makes sense for people to walk in the street and show that they're members of my blog. But it's really kind of the idea like your status is recognized as different from if you didn't have that NFT.

Nicholas: Sure, sure. The people at the end of a YouTube, you know, the Patreon supporters, at the end of a YouTube, etc.

Julien Genestoux: The Patreon supporters, exactly.

Nicholas: So, okay, so sorry, I mentioned these words, but basically in this show, they're saying that these are like four ways that you can make use of Unlocked Protocol. So status, you mentioned. Access, we talked about like token gated content or communities, discords, etc. Powers, which I think powers like an ability that you unlock if you are led into the token gate. You could do something special.

Julien Genestoux: Yeah, I guess it's the same as access, but it's access to content and access to features. I guess the way to frame this.

Nicholas: Yeah, yeah. And then stuff, which is getting, I don't know, a t-shirt or some physical experience.

Julien Genestoux: So that's also one thing that I tell people, like whenever if you're a member of my blog, and we ever meet in person, and I guess that would apply to you if you get a membership, I pay for coffee. That's the stuff that we get to people when I meet in person.

Nicholas: Does that make sense? Yeah, that's cool. Yeah. So do you think, I guess the big question I have is like, do the users, like, I have a sinking suspicion that crypto is more for merchants than for customers. Like, especially I traveled through Japan. I was in East Tokyo in Japan, and they have really highly developed, powerful RFID payment rails. And the same paths that you use for the subway system, you can also use it convenience stores and sort of fast food restaurants and some other stores. But basically, like, things that you want to access quickly, you can generally use this prepaid debit card that is primarily oriented towards the subway, but has been expanded into all these other. And it functions, you know, you sort of load it with cash. Maybe you can load it digitally with a card also, but primarily they load it with cash, in my experience. And it just made me recognize experiencing that in Japan, and then a similar system in Hong Kong. It just made me feel that crypto is not, it's going to take a long time for crypto to feel as fast, anywhere near as fast as this stuff.

Julien Genestoux: And just, I mean, I don't see why that's, I mean, maybe, but I don't think so. It's funny because at lunch today, we had a discussion with one of my teammates exactly about this. Basically, how fast, you know, it became easy to pay with your phone on pretty much any merchant now, because you just tap the phone. I think that's actually a good thing for crypto. Because that means that now instead of having to, you know, pull cash or card, which crypto would never work with cash or card, a crypto app could be the one thing that you tap on when you buy something. Instead of paying for your Apple or Google Pay card, you would actually pay from your crypto wallet. So to me, it's actually, I don't think it's going to take that much time.

Nicholas: It just can't require block confirmations, really at all.

Julien Genestoux: Yeah, but that's, but yes, but that's different as well. Like, so think of, you know, LTVs, I guess that's the thing, but like, I do think that in most cases, you could have signatures. So imagine a scenario where instead of, it's kind of the same with credit cards. If you think about this, when you swipe your card somewhere, it doesn't actually take the money from your bank account right now, right? There is a ton of stuff that happens and sometimes it takes like two days to actually settle the thing. It's just a matter of how you build these systems. So for example, imagine a scenario where when I type my phone, what it does is sign a message that is basically a transaction to spend, send it to the other part. So the other part do this and they can instantly verify my balance. They could also verify some other signals like how fresh or, you know, how often is this can protect you to decide to allow this or not. And to reduce the risk of fraud massively, because by the time I've sent a signature, it's going to be really hard for anyone, for me, in the meantime, send a double-spent transaction to withdraw from my US balance to prevent them from actually finally fixing the transaction.

Nicholas: Especially if you require that they have a minimum of 50 lifetime transactions and activity in the last month and stuff.

Julien Genestoux: Exactly. Which is kind of how cards do work in practice. Like you, you know, you, you're, you're, whenever you swipe your credit card, what it verifies is like how far you are on your balance on the credit card. And it will not work if you're above your balance of credit card spent.

Nicholas: So from the user's perspective, they don't really care, right? Like they don't care about decentralization until, until things like Twitter starts shutting down all the APIs and then Farcaster or BlueSky have more fun applications because they have an open API, which still, I don't, I, I'm not really convinced that.

Julien Genestoux: Yeah. I mean, I agree. Like the censorship is not the thing that people are worried about. And I think there's mostly right in the West and some, you know, I'm not worried about being, maybe I'm wrong, but I'm not worried about being censored right now in Williamsburg at this moment. What I'm worried about though, is the innovation being, you know, limited by the platforms that are basically setting the rules of what's possible and what's not possible. For sure.

Nicholas: Except that if we look at the web, like people still spend the majority of their time in apps as far as I know, right? So even though the web is more fertile ground, it's been kneecapped sufficiently by the centralized platform, hardware platform providers.

Julien Genestoux: That's exactly right. It has been kneecapped. And to me, that's the, you know, when we talk about like regulation of tech, I think that should be the thing that, you know, regulators focus on. Like, can we make sure that there is real competition and kind of making it impossible or punishing the platforms that are kneecapping the open ecosystem, including the web, but also including crypto. Like when I'm really sad and worried when I see, you know, Elizabeth Warren building the entire crypto crowd, because if she's after big tech, and I think that's what she said, then she should realize like crypto, for better or worse, is the way to go after big tech. It's definitely not by doing regulatory capture and say, nope.

Nicholas: For sure. But they're too out of touch. They don't understand technology. They're 50 years too old to understand the new.

Julien Genestoux: So that, but actually, I think it's even worse than this, because, you know, I think it's true for anything. I think it's, we have done a terrible job in the crypto space at making sure that people love us. Like right now, when you look from the outside, it looks like a bunch of grifters. And it doesn't look like, oh, this is the way to go after big tech. So I think she's even more, don't want to pick on her, but I think she's even more opportunistic in a way that she's just reading what her supporters want to say about crypto, which is, and arguably, it does look like a bunch of grifters. And so I think our job would be to start building delightful differences that allow people to do things that were not possible to do. And can assure, hey, you know what, you want these things to be real. And then that's actually happening because of crypto, not, and crypto is not just about this campaign and that. So, you know, yes, we can educate the politicians. I think more than this, we should focus on building cool, delightful applications and show that crypto is actually the way around a more egalitarian world, I guess, or a world in which there is less capture by big tech and big companies altogether, because I think big bank is not that different.

Nicholas: For sure. I don't disagree. But again, from the consumer perspective, the average consumer doesn't seem to care really about very much, except for the next three seconds of entertainment they're going to experience.

Julien Genestoux: Exactly. They don't care about it very much. But if you build stuff that, if we start building applications that are not possible to do, instead of crypto, that are both delightful and interesting and fun and, you know, don't feel scary, we, you know, crazy, crazy, doing that, then I think people will use these things. because, again, it's, you know, if it's fun and cool and something that wouldn't be possible to do without crypto, I think people will use these things. As a matter of fact, you know, people do use TikTok because it's new and different and any of that. So I do think there's an opportunity, but we just have to build these things. To me, that's one of the things that I'm worried about ecosystem. It's like we tend to build things just for ourselves more than for, you know, others. And if we don't actually start to grow by bringing more people in, then it's going to be really hard to actually change the perception that crypto is indeed full of scams and not that impactful.

Nicholas: I agree. Although I do think that maybe, you know, like it took Steve Jobs 40 years to deliver something like the computer that he wanted from the start. Maybe he didn't know that he wanted it exactly like that from the start. But essentially shipping the personal dynamic media device, it took about 40 years, maybe even 50. And I feel like crypto also...

Julien Genestoux: I don't know what is the...

Nicholas: Early 70s to 2007 iPhone, you know?

Julien Genestoux: Apple is late 70s, right? It's more late 70s than early 70s. Yeah, but I mean, anyway, 30 years, but it's still a long time.

Nicholas: Or if you go from like transistor to personal computer, it's like, you know, 50, 60 years. I just mean that it might be a niche. For what? I just think like the majority of Web 2 has operated by the logic of like the Facebook India, Facebook basics. You know, like it's just like permeate as much as possible, deliver lowest common denominator stuff, exploit behavioral psychology of the entire population, serve things that cavemen would understand as like instinctive desires. And versus like crypto, it seems to me like it's more in a phase of being a niche product for people who actually do care about what it does, rather than just like seek sugar, fat, sex. And like, it's a little bit more intellectual than that at this point. Even in terms of like seeking speculative opportunities, it's a little more intellectual than watching someone get hit by the football or whatever.

Julien Genestoux: Yeah, I'm not dismissing what we do right now. And I think this is cool and fun. And I'm definitely one of these DJs. So I'm not dismissing it. But I do think that if we want to grow beyond, we will naturally need to start thinking about...

Nicholas: It just might take a lot longer to reach the kinds of audiences that Web 2 products reach because we're not as easy to get as that. There's a lot more friction.

Julien Genestoux: Yes, maybe. Maybe. I mean, yes, I mean, it does take longer than what I thought it would, right? So I don't want to dismiss that either. But I also think it only is a function of what we end up building. Like you say, I mean, in the end, I think it's if we focus on our attention and building stuff, then I think we can do a lot more than if we just wait for people to come. Like I feel like one of the cultures that I've been always kind of a bit worried about is the HODL culture, right? Like it's like, okay, don't do anything. You sit on your point. And I can see why, okay, this creates kind of artificial scarcity and then tokens go up. But in practice, that also means you're not doing anything your point, which means that you're actually not doing creating any value for the ecosystem. And I don't think the value of HODLing anything is actually that universal. Like, I mean, sitting on a bunch of rocks doesn't make these rocks more useful. It's like when you start carving the rocks and start extracting oil from the rocks, and you start doing things with that, that you create value for the ecosystem. So I think a lot of the value in a lot of the cultural references in the space are actually built around the idea of immutability, of not changing things and just waiting. And I don't think that is how it becomes impactful. I mean, I can see in some context, but yes, okay, this is a key characteristic. But I think what I've seen from tech specifically, but like, you know, rapid iteration, almost kind of, I don't say wasteful iteration, but almost kind of to a point where you end up creating things that nobody cares about, but you do that a hundred times, you will end up creating something that is impactful and useful. More than if you just sit on your hands and wait for things to happen outside of this. So I think it's more a cultural thing that we have to change on our side. I'm not saying nobody's doing this. I know a bunch of people that are trying to do that. But I do think a lot of the culture around crypto has been around, you know, waiting for institutions, waiting for the masses, waiting for X, Y, and Z. When I think it's probably not the right approach, in my mind, a more interesting approach is the one where we build things. Maybe we're wrong. It's okay to be wrong. But we're trying different things. And that's how we make progress by accumulating knowledge over time.

Nicholas: That's a beautiful way to put it. I totally agree. I totally agree. This has been really interesting. Thank you so much for sharing all this information about Unlock and all this wide ranging conversation we've had. This has been really great.

Julien Genestoux: This was a pleasure. Thank you very much for everything.

Nicholas: Yeah, this was great. I hope to talk again soon. Really great talking to you. So thanks for coming through. And thanks everybody for coming to listen. Another episode next week, Friday at 5pm Eastern Time. Thanks again, Junya. And thanks to RainbowKit for sponsoring this episode.

Julien Genestoux: Thank you very much, Nicholas and Chet Su. Thanks. Bye-bye.

Nicholas: All right. See you. Thank you.