✨ DEBRIEF | Unpacking the episode: https://shows.banklesshq.com/p/debrief-dan-romero
------ Dan Romero was an early employee at Coinbase and is now the Co-Founder of Farcaster, a new decentralized social network that has captured the attention of many in Web3.
Aside from sharing a secret passcode to access Farcaster during the episode (Bankless Premium subscribers only)…Dan also shares his thoughts on why he doesn’t think Web2 social media is broken, why people should care about decentralized social media, how decentralized does it have to be in order to work, and how all of this will fundamentally change the internet and the trajectory of humanity.
------ 📣 Crypto Tax Calculator | Free Crypto Tax Calculator https://bankless.cc/CTCpodcast
------ 🚀 JOIN BANKLESS PREMIUM: https://newsletter.banklesshq.com/subscribe
------ BANKLESS SPONSOR TOOLS:
🐙KRAKEN | MOST-TRUSTED CRYPTO EXCHANGE https://bankless.cc/kraken
🦄UNISWAP | ON-CHAIN MARKETPLACE https://bankless.cc/uniswap
⚖️ ARBITRUM | SCALING ETHEREUM https://bankless.cc/Arbitrum
🚁 EARNIFI | CLAIM YOUR UNCLAIMED AIRDROPS https://bankless.cc/earnifi
------ Topics Covered
0:00 Intro 6:08 Is Web2 Broken? 10:25 Web3 Improvements 19:25 Unique Web3 Properties 24:42 Developer & User Sovereignty 31:57 Farcaster’s Feature & Protocol Thesis 45:25 Unique Properties of Decentralized Social Platforms 58:29 Growing Farcaster & Client Competition 1:05:39 Incentive Mechanisms 1:11:30 Is a Token Needed? 1:15:55 Protocols 1:19:48 Advertising & Farcaster 1:25:48 Farcaster On-Chain Footprint 1:27:47 Spam Prevention 1:31:31 Discoverability & Regulation 1:33:45 Identity 1:36:00 Closing & Disclaimers
------ Resources:
DM Dan Secret Passcode https://twitter.com/dwr https://danromero.org/
----- Not financial or tax advice. This channel is strictly educational and is not investment advice or a solicitation to buy or sell any assets or to make any financial decisions. This video is not tax advice. Talk to your accountant. Do your own research.
Disclosure. From time-to-time I may add links in this newsletter to products I use. I may receive commission if you make a purchase through one of these links. Additionally, the Bankless writers hold crypto assets. See our investment disclosures here: https://www.bankless.com/disclosures
No summary for this episode yet.
I've been in crypto for 10 years. I think Ethereum is the right balance between decentralization and pragmatism, and I think building things that people can use are an underrated thing in crypto. I think a lot of people tend to focus on the ideology, which is important. But what I showed up to work at Coinbase, and now what I'm doing with Forecasters, I want to build a protocol that has a billion plus people using it. But I want to do it in a way that doesn't compromise on the decentralization and the sovereignty at the user account level. Maybe it's a fool's errand, but that's what we're trying to do.
Welcome to Bankless, where we explore the front-tier of internet money and internet finance. This is how to get started, how to keep better, how to front-run the opportunity. This is Ryan Sean Adams. I'm here with David Hoffman, and we're here to help you become more bankless. Do we really need decentralized social media? Our guest today, Dan Romero says, yes, he explains why. We talk about the rise of decentralized social networks. A few takeaways for you in today's episode. Number one, why Dan thinks social media isn't broken, uses a different term, and we discuss what he means by that and why.
And number two, we talk about what's new with decentralized social media. Why should people actually care about this? Not the hardcore crypto people who love decentralization, but the normal people, the masses. Number three, we talk about how decentralized a social media network actually has to be in order to work. What should be on-chain? And what doesn't need to be on-chain? Number four, we talk about why a protocol, not a company, is really the way forward. And finally, we conclude the conversation about how all of this will fundamentally change the internet and the trajectory of humanity.
David, what should listeners pay attention to this episode? Dan Romero is of course building farcasters, but I think he also really just lays out the blueprint for any decentralized social media network, any protocol. And so while this is under the context of farcaster, I think whatever decentralized social network does emerge out of the Web3 space, it's largely going to be synonymous with the blueprint, the map that Dan has laid out in this episode. And so I think this episode I think is actually going to be relatively timeless. Even if farcaster for some reason doesn't see success, some other social media decentralized protocol will eventually, and it will largely follow the same blueprint that Dan laid out in this episode.
So I think that's really the thing to pay attention to is as you are listening to this, understand what primitives will be true no matter what flavor of social media protocol comes out of this space. Because I think if you're bullish on this space, you are certainly bullish that eventually one will come out for maybe it's farcaster, maybe it's a few others, maybe it's others we haven't heard about. But they're largely going to follow some of the same primitives and same first principles that were communicated on this episode. In the debrief, David and I are going to discuss what we think about our experience with a current decentralized social networks as well.
So stay tuned. If you're a premium member, you can click right into that debrief before we get to the episode, we want to thank the sponsors that made this possible, including our friends at Douglas Nation, talking about decentralized social networks today with Dan Romero. Here's the co-founder of farcaster, which is a new decentralized social network that's captured the attention of many in Web 3. There are a few platforms competing for our attention these days and farcasters certainly one of them. We're going to talk all about that today and Dan, it's great to have you, welcome to bankless.
Thanks for having me. Okay, let's start here with a big question. What is broken about Web 2? Why do we need Web 3 social networks? Why can't we just be satisfied with the Web 2 versions of these things and how do we even improve on those versions in a crypto version of a social network? Well, maybe I'll start with a controller and answer is, I don't necessarily think Web 2 is broken. I think it can be approved upon. And so if you kind of just take the history of the internet, Web 1, you have decentralized protocols, HTTP, SMTP, DNS, and if you wanted to put information online, you basically went registered to domain, figured out a web host, put up a website, and then hope people would find you.
Google made it a little bit easier. And then you kind of had with Web 2 this explosion of user generated content. And with that user generated content, you got to a place where distribution became a lot easier, right? You could start a podcast, put it on the iTunes store, and then eventually on YouTube, use Twitter for that distribution, have a substack. And so you kind of have all these great publishing tools, where you can reach a massive audience, right? Like the whole creator economy is a result of Web 2 getting these platforms to scale, and then allowing people who might be a little less tech savvy, but really want to go be entrepreneurial and build an audience like yourselves to get to a really large scale.
But as that scale happened, you know, we went from kind of this, the classic nag on Twitter on the early days was, you know, it's just a bunch of people saying what they ate for breakfast, and then kind of call it around, I don't know, 2017, 2018, it was a threat to democracy. And so it's kind of a pretty big jump in a 10-year period to go from kind of a bunch of nerds talking about what they had for breakfast to, you know, this is going to conditionally change the way our society has run. And I think with that kind of growing importance in society, and just by functioning of having billions of people using all of these Web 2 services every day, we're starting to see some of the limitations of kind of having a single platform be, for example, the public square on the internet, if you wanted to describe Twitter like that.
And it makes it really difficult when you have a global audience and a global user base to have a group of people in San Francisco, whether it's the public version of Twitter or Twitter owned by Elon Musk, making decisions that impact how discourse speech communication happens globally. And I think where we sit now is, I think people from across the political spectrum, I think for the last few years it was almost kind of like hardcore crypto people were interested into centralization, and then there was this kind of branding that the only people interested in new social networks were kind of like right-wing people who had been de-platformed from a Twitter.
But I think quickly we've seen with Elon taking over, there's a whole new group of people who are all of a sudden being like, hey, I should be able to post a link to a Macedon post and not have it be censored. And so I think people are reevaluating some of this stuff from first principles and realizing, hey, there are other protocols out there on the internet today that we use that are decentralized, permissionless, the web, and email being, you know, two really important ones. I mean, Sub-Saxon is a great example of a web-2 company built on top of a decentralized protocol email. And so I think what we're hoping to do with Forecaster is marry the kind of decentralized and permissionless protocols of web-1 with the usability, convenience, ease of use that web-2 has ushered in that billions of people around the world are now able to, you know, create user-generated content online and kind of participate.
And so I think if you can marry the two of those things together, then you have a superior system. So I think what you're saying is the distinction that you're making when I use the term broken to maybe characterize web-2, you're sort of toning that down. You're saying, well, not broken in that we need to do a wholesale, you know, rip and replace. But there are fundamental areas of improvement. So not broken, but improvement. And I think maybe this is kind of a semantics between the idea of something broken, the idea of something needing to be improved. But I think what you're saying is don't throw out the baby with the bath water here.
There are some fundamental things that web-2 did correctly, namely getting the use case, right. And what we actually need to do is figure out a way to improve on all of the flaws. If we were to look at some of the flaws web-2 right now, I think you sort of hinted at a few of them. This notion of web-2 going to a place where Twitter going to a place where it was basically doing nothing to it was destabilizing our democracy. Wow, that happened very quickly, right. And so that's one thing that people point to. Another thing is bots are everywhere. There's no concept of no notion of good identity or human based identity.
Another of course is you kind of alluded to this as it's very controlled. That means it's very sensible. I don't have the ability to kind of own my own Twitter profile in my social graph. It's basically the property of Twitter. And I feel somewhat like a, you know, a peasant on someone else's feudal land. So maybe these are the things. Is that list what you would point to as the improvement areas? And I guess my question is how does web-3 help any of that and does it actually help any of that? Yeah, so I think all of those points that you made are accurate and things that web-3 can improve on.
But the way I described Forecaster is there two foundational principles that we are trying to achieve. And they solve two different problems. So the first is more for someone who has an audience, right? You have an audience with this podcast. You rely on web-2 platforms, which give you, again, massive distribution, ability to get your word out. But you don't own your audience on, I'd say most of those platforms. Substack being a rare example where if you're using platform like YouTube, you could have built up 10 million followers over five years. And on an arbitrary whim, no matter what the reason, they can turn your account off.
And you have no ability to get one last export of those subscribers. They have, by the way, Dan. I don't know if you're aware of this. But on Mother's Day of last year, YouTube shut down bank lists. I thought you were going to say that they let you export the subscribers. And they're like, wow, that's a pretty fine hearing thing. No, they shut us down. No, the worst one. Yeah, okay. And so what do we do? We go into no word, no word of warning, just an email, a friendly email from some bot somewhere that says, your accounts are down. And there's no kind of escalation process. And the only thing we could do, Dan, at that time, was to get real out on other social media platforms and say, look what YouTube did to us.
Which was only a luxury that we really had. Because other accounts were shut down that same day that didn't have the same sort of like feed following that we were able to invigorate to get the attention of YouTube. Yeah. So apologize for interrupting you. But this this happens. It happens in crypto. It happens in other communities as well. So go on. Well, yeah, I guess I don't need to sell you on this. But the idea that you could have built up this audience, generate a lot of revenue for for whatever platform. And with the snap of a finger, it's gone. There is no arbitration. There is no getting in front of the judge and make your case.
Instead, you get an automated support email treating you no different than one of the kind of like CZ spam bots on Twitter. I think that is unacceptable in a world where creators are real real businesses. And this is your livelihood. And so contrast that to email where if for whatever reason you decided to not work with SubSack or SubSack didn't want to work with you, it would be ludicrous if they they held your emails hostage. Right. And part of that is just a norms thing. Right. Email pre-existed a lot of these webtoop platforms. The idea that if you have an email newsletter you own your newsletter subscriber list is just kind of that that's what the market forces.
Whereas I think with with these webtoop platforms, because they've been so attractive from providing distribution and scale, they've been able to get away with not allowing a kind of publisher or creator to own their audience directly. And so I think that foundational to to Forecaster is you own your relationship with your audience. And actually, it's not even quite that because it's your audience owns the relationship with you. Right. So someone follows you on Forecaster. It's not like somehow you you now can kind of go ping them. They've actually opted into to following you. But the important thing is no third party can come in and say sorry.
These two adults cannot decide to connect with each other over the internet. So that's on the creator publisher side. I think that the second thing that's really important and maybe I'm dating myself here. But I remember when Twitter was a very open API and had a flourishing ecosystem of apps. And a lot of people don't even know a lot of the core features of Twitter were invented A by Twitter users and B first adopted in third party clients. So app replies, hashtags, quote tweets, retweets. Those were all conventions within the 140 character tweet box. And then you had clients like tweet bot, Twitter Rific, Tweety, inventing, okay, this is how you natively handle it.
And for the first few years, Twitter didn't even have a mobile client. They actually acquired the app called Tweety, which invented pull the refresh. So all of the apps that use pull the refresh today you actually use a patent that Twitter owns and has decided to not generously not prosecute against. But anytime you use pull the refresh in an app that was invented by a third party developer on Twitter's API. And then roughly around 2014 they basically shut that down. They needed to make money on advertising and it's very difficult to do that in a kind of third party app ecosystem. And actually more timely, as of I think like last week, they just turned off third party apps, right?
They kind of existed in this like weekend state for the last 10 years. But but now like I don't think tweet bot works. Like they just shut it off overnight. And again, no communication. These are people building businesses, even in this kind of like we're Twitter never really like their API. And another prominent example from the previous era was Facebook had an open platform, pretty, pretty mobile. It was kind of like out of all these games and all these things that are built on top of it. Zenga was a company that basically built this entire business on Facebook. And then in a very short period of time, they kind of said, hey, this is deep prioritized, tough luck, like, you know, go find go find a new way to to build out these games.
And so that that those two examples from that era changed. I think a lot of developers opinions on I'm not going to go build a business or or make a serious investment in a platform knowing that you can have the rug pull that from under them. And so I think with Forecaster very much inspired by Ethereum and my time and crypto, it's you need to build a platform that hits a bar that is truly permissionless both on the access to the data and the APIs to the system working. So that as a developer, if if you want to go build a client, I as you know, one of the early people building the actual protocol shouldn't have a say what what client can be built, just like Ethereum, right?
I want to go build a competitor to Metamask. There's not some approval process that said, well, Metamask was the first wallet or, you know, one of the first wallets. So we're going to actually decide whether wallets can come in a rainbow Coinbase wallet, you know, and anyone can go build a wallet on top of a network like Ethereum. And I think that makes Ethereum stronger for it. So that's the other bar we want to hit with Forecaster. So you own your own audience or that relationship belongs to you and as a developer, your relationship with your users and the data completely permissions. I think that that was a really great exploration into some of the shortcomings of the current Web2 world.
And Dan, I know that you say there's not much broken about Web2. There's only things that we can like learn and build upon. But I think in your answer just now, to the answer to A, answer to the question of what's broken about Web2 is actually kind of like the shareholder equity incentive structure, because that's what created a bunch of monopolistic tendencies that you just listed out from Twitter and Facebook where Twitter began as a very open ecosystem with APIs to enable third-party creators. But due to what Twitter is, a Web2 company, any sort of thing of value that got built by somebody else was simply absorbed into Twitter.
And eventually that aggregation and centralization and monopolization created the Twitter, the closed form of Twitter that we see today. And the same thing was Zanga and Facebook. So maybe to answer what's broken about Web2 is actually like the traditional equity structure that a lot of these normal companies are kind of built into. And so I want to ask a question, we can ask the question, what's broken about Web2 and what does Web3 do to fix this? I want to ask this in a slightly different way. There's a very early implementation of Ethereum was Peepeth, which was Twitter on Ethereum, where we could just say, oh, we can make Twitter, but Web3.
We'll take Twitter, we'll copy and paste it, and we'll put it on Ethereum, and then boom, we have decentralized Twitter. That really never took off, but maybe under a Farcaster kind of structure that might be able to work. But there's one aspect of this conversation, which is how can we just take systems that already exist Twitter, Facebook, and then make them decentralized in a more open environment. But also what does Web3 native social networks enable that Web2 social networks will never really be able to enable? So there's like some conversations, which are like we can a decentralized Web3 version of Twitter can help with making sure that people don't get de-platformed, but that's only about fixing Web2's faults.
What about these more decentralized social networks? What can they really enable that Web2 will never really be able to enable? What's unique properties, what unique properties about Web3 social will Web2 never really be able to have? So I want to start with first the going back to the study of developers having permissionless access to the data in APIs. I think that in and of itself is underrated in that if you allow developers complete freedom to do whatever they want, they're unbounded in terms of if they have a creative idea, they can go they can go do it. And then they have to convince other people to use it as a separate thing.
But there are no API terms of service that they necessarily need to worry about. And I think that alone is actually an important thing and whether you call that out, you could build that in Web2 or Web3, put it to the side is like that just existing is a primary motivation for building something like Farkaster. As it relates to more specific features around cryptocurrencies, we can talk a couple, we have one today. So we actually allow you to connect another wallet within Farkaster. And so you can kind of have a native newsfeed of what the people you're following, you know, you're following them for the Twitter like content, like small text posts, but also what's happening on chain.
And so you're kind of getting this blended newsfeed of the kind of stuff that they're talking about, but then also what people are actually doing on the blockchain, which it natively starts to create conversational points around anything that you are doing on chain, right? You meant to then FT now that natively shows up in the feed, which now NFTs instead of being this kind of single player thing can now be more multiplayer. Composibility of feeds. Exactly. So that. And then I think that the other thing, Farkaster's name space account, right? Like the way you kind of define yourself is also an NFT.
And so that's going to be an Ethereum one next year. And so then you get all the composability that can happen with an NFT. So you could have a account that's owned by a group of people, right? And you could have an ERC 20 token associated with that. And I don't know, use a Nouns style governance like it's just like the possibilities are endless in terms of ownership of a given account. So there's going to be a lot of composable stuff that will happen there. And then I think the other important thing is every user on Farkaster by definition has a working public and private key happens to be in a Ethereum address.
And if you think about what that all of a sudden starts to enable, you can out of the box have end to end encrypted direct messaging that is as good as sick and in some ways better because you don't need to have a phone number associated with it. Another example that you can start to kind of think about because there's a lot of development that's happening in zero knowledge proofs is you kind of have all these academic potential uses of zero knowledge proofs that now can be developed permissionlessly by any developer because you can make the strong assumption that every user already has a live working public and private key.
And so you could potentially make new types of speech that is kind of like based on some amount of on chain credential. But at the same time, you don't necessarily have to reveal who you are. So a basic example would be I want to cryptopunk. I want to actually be able to say something anonymously, but have the credibility to say I want to cryptopunk. If that's something that matters to you, but you could imagine other types of on train credentials that could be I don't know around politics or around medicine. And now I could be a doctor who is well regarded, right? Maybe maybe I've won some prestigious medal in the field and I want to speak up about something in medicine, but I feel like if I was to do that under my real name, I would get a lot of pushback from the community versus I could say something have the credential that is provable on chain and actually do that in a zero knowledge way.
So I think those primitives are going to allow developers to have just kind of a lot of creativity in what they built. And so we're really focused on building the core level set of primitives. And then really trying to make it a developing developer platform that you build uniswap you get uniswap or sorry you build Ethereum then you eventually get uniswap or or open sea and you start to get people's creativity is actually what drives drives the growth in the use cases. Okay, Dan, we're definitely appearing at the top of this very large rabbit hole which is this whole entire conversation of more decentralized social networks.
And we've touched on a bunch of different possible like past that we can go down. I want to go back to one that we've already started to go down. I really want to drive this point home because I think it is at the basement of what is a very large stack of cool awesome expressivity of developers and users on top of it. And it goes back to what you said about assurances around like the open API structure or like the assurances that developers can develop on this platform and knock it rubbed. Because that's that's a big talking point about like even even like Facebook's like changing or closing down their APIs.
This conversation goes back to like why does Ethereum exist to build unstoppable applications. And this also is a very core property a very core theme of what we often frequently talk about at bankless which is settlement assurances. We use the term settlement assurances to describe property rights like why is Ethereum in Bitcoin different than like Litecoin or insert your faster smaller blockchain here. It's because of settlement assurances. You are assured that your property is yours because of the of how strong the security is on Ethereum and all and smart contracts right. And so these settlement assurances allow people to think into the future.
And we've related this conversation to like things like laws with the Supreme Court. An executive order is a very fast law out of the president. You can get overturned by Congress. But if Congress writes a law that can get overturned by the Supreme Court and when the Supreme Court says it's final it is final. And what we've created with smart contracts is a new level of finality that gives people settlement assurances to think into the long term. And that's the same conversation I'm seeing what you're talking about with creating a developer platform that developers can't get rugged. And I want to ask you about what is the difference in the assurances that developers have in this new world and what are the downstream net effects of producing a developer platform where the developers have the assurances that the structures that they build will be around forever.
And we're like what kind of creativity is unlocked at that point? Yeah, so I want to be clear up front. One of the reasons Farkass were still invite only is we aren't at that bar internally of what we'd say is like an unregable platform. And we've put a lot of effort in over the last year and we're hopefully getting pretty close. We're hoping to have it kind of midway through this year at a place where Farkaster from a developer standpoint is actually permissionless and you know call it unregable or the idea is you're sovereign. As a developer you will have full access to the data and APIs of the protocol.
And that's built with something called Farkaster hubs, right? And I actually think it's important to delineate that we only do a minimal amount of stuff on chain. And part of that is actually we're trying to keep cost fees and usability usability high and cost and fees down. And so what we do is when you sign up for an account we actually create that on chain. Today it's on testnet it will be on Ethereum, a one hopefully you know by July. And what's great is as a default every single user on Farkaster owns their actual account, their identity. And to use another example a system like email. So if you use a Gmail address you actually don't own that Gmail address.
Gmail can actually at any time just revoke that Gmail address. There is no arbitration. You can't port it to another service. Now if you're a little bit more sophisticated and you own your own domain you can use G Suite, pay a little bit more and then now you have the ability to port your email. So email is a technically it does work in that you can port but the practical reality is most people use a free email address. And actually they don't have to think about it most of the time. But if you do run into an edge case where you do get suspended by Google and you lose access to your email that's a really really tough thing.
With Farkaster we're making the decision that every single user no matter how sophisticated they are actually gets a sovereign identity that's going to be on Ethereum. And we actually use something called a like a gas list transaction, a meta transaction, and we just seamlessly onboard people on through the Farkaster client. And we actually think that that's great because it allows both a sophisticated person and a non-sophisticated person just onboard just like they would web do. But behind the scenes they do have the actual ability to move that identity anywhere they want and within any client they want.
And so that's actually a foundational component to what the user is on Farkaster. But more importantly that also gives the developer the assurance that if they convince a user to use their app, there is no third party that can come in and say, you know, like this Twitter example, oh sorry, we're the one that actually owns the identity, we're going to we're going to revoke that. Contrast that to actually a federated system, something like Macedon, you're getting your identity provided to you by a server. So if that server wants to revoke your identity, unless you're running your own server, like it's going to be very similar from an email standpoint in that you just lose access to that identity.
So having the pointer be sovereign and actually on Ethereum, one we think is worth the cost and the complexity to do that. In addition to the composability and I think the general benefit that you can get from from having it sovereignly owned on Ethereum, the getting the other data in APIs to a permissionless state once once these kind of hubs exist, which you can think of a hub is like the Ethereum node but for the Farkaster network. The other decision we've made is we are actually going to limit the amount of data per user on the protocol in an effort to keep the requirements of running one of these hubs reasonable for an average developer.
And so you can think of it like this, we're going to cap you at the equivalent of 10,000 tweets we call them casts. So on the 10,000 and first, you know, cast, we bump your oldest one out at the protocol level and then that's the implementation for the hub. And so if some people might be like, oh, well, you know, I wish we could have all of it, but the nice thing, one, all the data is permissionless so you could back it up to our Weaver IPFS. You can do that yourself. But the benefit that you do get in kind of making this bounded data per user is we really think we can get up to 100 independent hubs that are running, which are all going to have a full global state of the network.
In the same way that if you're running in a full Ethereum node, you're going to have full global state and that ability for a developer forces the network from a kind of like, we can't be a bad actor as the kind of initial developers and having the biggest client, we can't say, oh, we're going to gate this stuff because we've actually intentionally built the protocol in a way that I can be, you know, two years from now as an independent developer, I can spin up a hub and I can actually have the full global state of what the Farcaster protocol is in a given point. And we think it's worth to parallels with this design with how the Ethereum protocol is designed, which makes me think that you're onto something here.
I've been in crypto for 10 years. I think Ethereum is the right balance between decentralization and pragmatism and I think building things that people can use are an underrated thing in crypto. I think a lot of people tend to focus on the ideology, which is important. But what I showed up to work at Coinbase and now what I'm doing with Farcasters, I want to build a protocol that has a billion plus people using it. But I want to do it in a way that doesn't compromise on the decentralization and the sovereignty at the user account level. Maybe it's a fool's errand, but that's what we're trying to do.
That's super cool. This idea of building a social platform that is rugless, not bankless, but rugless, which maybe that's another podcast we spin up at some point in time, David. But I think devs will appreciate that, particularly devs who kind of look at sort of the long term. I think that the bet here you're making is that, I mean, when Twitter destroyed its APIs, it essentially destroyed all of its permissionless innovation, didn't it? And the only innovation that could happen is innovation that could happen with its internal employees. By the way, they just reduced their head count by 50%.
So do we think we're going to get any new big features from Twitter anytime soon? We'll see. Probably not. But this creates kind of a permissionless ecosystem. And I think that's the bet that permissionless ecosystems win out because you have everybody in the world sort of hacking on this and the best employees and the best developers don't actually work for your company. Do they? They're outside in places in India or all parts around the world. So I get that. I think from somebody who's never not a developer, somebody who's never been de-platformed before, they're still not quite clear on this value proposition entirely.
And I want to get back to David 's question of what's new about Web 3? Like, why is it a step improvement from a Web 2 social media tool if I haven't been de-platformed yet? And I think that's the majority of people listening to this have not been de-platformed from Twitter or from Facebook or from their YouTube account. They can imagine it might happen, but things would have to get bad, just to be in order to do that. And so they don't really see the value proposition of being rugless. That's why I want to go back to kind of the on-chain features that you're adding with Farcaster. So Dan, before we entered this episode, and started recording, David and I had an opportunity to go download Farcaster on our iOS device, right, the app type version.
And for listeners right now, I believe Dan, this is all kind of private data. So it's invite only. And so we asked you, like, hey, Dan, you're coming on. Can we get the invite link? And you sent us that. Thank you. What's interesting about this is I kind of see things that social media apps are not providing today, right? Like, like the Twitters of the World and the Instagrams. You have this whole section of events that you were talking about, where I just plugged one of my eith addresses in. And I can see that I minted the bankless ultrasound money podcast episode. It's right in my event's queue.
And I could go look at David's profile and I could see all of the things that he did with his on-chain data and his wallet. And that's kind of interesting to me, right? Maybe I set an alert for every time David buys an NFT so I can ape in because he's got the alpha. And then, and then I, there's this tab called NFTs where I could see kind of the NFTs added to my address. And then you're making this comment about identity. Well, I was actually able to set up an RSA at RSA is my handle on Farcaster right now. And you're saying that in the future will be kind of its smart contract driven and is fully composable with the rest of the on-chain DeFi, NFT, NFT, Web3 ecosystem as well.
And so we have these kind of primitives. And I guess what I'm saying is I think this is a thesis now that he get into it. We're still very early. But I think that every single social media app is going to need to be redesigned to incorporate on-chain Web3 data sets tokens, NFTs, identity, all the innovation that's happening with in Web3. And I think Dan, they're going to be super freaking slow to do this and respond. And they're going to be leapfrogged by social media applications and platforms that actually do. And maybe even they'll find that it's counterproductive to their business model to expand into Web3.
I reminded of like Instagram right now. They just rolled out an NFT feature. Guess what? There's a 15 to 30% tax if you issue an NFT on their platform that they take. That's a pretty high take rate for Web3. Like we don't do that in crypto, do we? That's kind of again. Yeah, we don't do that here. And so I guess what I'm saying is this is early stages, but I sort of see when I download something like Forecaster, I get like the Twitter experience. But now I have some on-chain data that's added to this. And it's starting to become valuable because the big question for me going to a conversation like this Dan is like, I know people aren't just going to come for the censorship resistance.
It's decentralized. Okay, cool. I don't care. I'm not being de-platformed. They will come for the features and to the extent that you are able to innovate faster on those Web3 features and make that valuable, then I think we'll start to kind of draw the audience. Anyway, that's how I'm thinking about it. What's your take on that thesis? We completely agree. It's actually one of the reasons we're doing both the protocol and the initial client at the same time because our review was one, we could go build the world's best protocol, but then we're going to have to spend a whole bunch of time trying to convince other people to build on top of it.
And that in of itself is extremely hard. And so where we started was, hey, we're going to build the protocol and the initial client. And we're going to do a whole bunch of hard work to try to figure out features and a set of use cases that are going to get at least some initial group of daily active users. Because I don't think any other developer is going to want to go build on the protocol if it doesn't already have active users. Users use software, they use clients. And so thus why we started with building our initial client, built those initial features. And our thesis is if you build the best network for people who are interested in crypto and Web3, that is an amazing initial starting community, especially if you believe that we're going to continue to have more stuff that ends up on chain, whether that's visible or kind of like behind the scenes in the background, that's our bet.
So we're building the initial app is a social network for people interested in crypto and Web3. And thus the set of features that we've built, right? If you say GM on Farkaster, we change the like button to a GM. You have the NFT features. If you actually go to any given NFT, a crypto punk, you can actually see everyone else on Farkaster that provably owns that NFT, right? So if you find that there's some community, whether it's not crypto, coven, crypto punks that you are interested in finding those other people, you can actually just go and follow every single person. So you could go take your bankless NFT, click that as a category, and then actually find everyone else on Farkaster who minted that NFT.
And so those types of features are for our initial user base, highly engaging. And we're going to continue to add more of those. And then actually we have a kind of bet that over the course of the next few years, you're going to have more and more just things. Web3 objects is what we call them, just natively represented on chain. A mirror post is another great example. From the surface to the average user, that looks like a medium post, right? Like no difference. It's a bunch of text rendered on a webpage. But what's cool about a mirror post is you can have the the instantiation of it on our weave.
We can permissionally pull that into our client. In addition, there is interactivity with the mirror post because you can mint it and you can trade it and collect it. And so all of that functionality, if we're kind of taking Web3 as the bet and understanding natively what's happening on chain inside of a social experience, you almost start to get all this innovation that's happening on chain for free, right? So if you're natively servicing whatever someone's doing on chain in a social way, there's some new use case that pops up. You're now starting to kind of have an initial conversation about it, but then you can actually add and make the features even better.
And so that's for sure how we're approaching it. And I think a lot of criticism that I've heard about Farcaster's futile to be working on this stuff because it's too crypto-centric. Our point of view is we'd rather actually focus on what might seem like a niche, but a niche that is the future. And for us, if you can actually get the smartest people in crypto engaging in high quality way on this new social network, I think you'll naturally start to bring other people who are curious. And because they're going to say, oh, this is high quality discourse, right? Vitalik is talking about the need for new institutions on Farcaster and he gets 40 replies that are actually thoughtful and not a lot of spam.
That, you know, if itself is also kind of interesting. And so I think for us, it's very much about we have to build a compelling product that people can use as a way of bootstrapping the actual protocol. Because if you can get to some semblance, whether it's 10,000 or 100,000 active users on the protocol, then I actually start to think that you're going to get critical mass for developers interested in building other new applications, which may actually increase the aperture, right? Someone goes builds a very specific app that's dedicated to sports fans. We're never going to be the best at that.
But if someone else has a great idea for that and you already have 100,000 people who are actively using the protocol, that's your initial seed community. And then you bring in people who have no idea that this is all powered by group that they're just coming to use that app. So the initial use case here for Farcaster and the focus for it from a user persona, like the type of user that you want is like the full NFT crypto native type person. Yep. And they're going to want to migrate here because Twitter doesn't have the features, doesn't have the on-chain features that they actually want to need for their communities.
Not the censorship resistance. Because I completely agree with you, the reveal preference is most people don't care, right? Like they only care when they eventually get booted. But for most people, that doesn't happen. So Twitter actually works for them right now. So you have to have better features. You don't care, Dan, that right now the criticism people say, yeah, but the crypto native population is so small. I mean, what are we talking about? Maybe millions of people. How many daily active users does Metamascad? That's kind of your target market, basically. Your response to that is, I don't care because that number is going up, and it's going up real fast.
Basically, I think that and people who are building the future are inevitably, that's where people are going to want to spend time. And I think if you were to look at early Twitter in 2007, 2008, it was a bunch of tech nerds in San Francisco who were going to South by Southwest. And that was the criticism. It's like, okay, this is just a bunch of nerds who would want to like share what they're having for breakfast. These people are narcissists. And maybe there's some truth to people who use Twitter. I used Twitter a lot that they're narcissists. But the reality is, as it turns out, there are a lot of people in the world who find that interesting.
Even if they're not a detect, it just took 10 years to get there. We're about 10 years out from threatening democracies as what you're saying with our people. My goal is to not have it throughout the democracy, but yes, I think things that are... That kind of impact, massive impact. Things that are big look small and are very easy to ridicule. And I mean, I saw the Secoine Bays. Like, I was there in 2014, 2015, 2016. And so many people were like, why are you wasting your time on Bitcoin? Like, oh, like, did Bitcoin die? Like, all all that stuff, right? And actually Ethereum was what really kickstarted, I think, the second chapter for Coinbase.
And then after the kind of like 2017 hype cycle, it was clear that like, crypto wasn't going to go away, at least to smart people. And I think you've had multiple cycles. But I think generally, I think that that's the same thing is going to happen for decentralized social networks is people are going to say that, oh, they're toy, they're never going to hit scale. Like, you know, it's impossible to beat Facebook and all this kind of stuff. But look, I mean, if people say that you can't like compete with Facebook, one TikTok went from zero to whatever it is today in a very short period of time.
And and Facebook really hasn't been able to compete with that. So a that's a that's a counter example that you can compete. And then two, we're seeing this how even now, like people said last year, Google is you can't compete with Google. And then all the sudden, chat GPT shows up and then people are like, oh, my goodness, like, this is going to replace Google. And I think that's probably overblown. But but if you're working in technology and you're working in an area where there's just constant innovation and things are getting pushed forward, to have the point of view that that we're just going to be in the stasis where it's like impossible to beat the big guys, people used to say that about Microsoft and IBM.
And so it's like, it's just like a lack of history for for most of these folks. And so I'm completely fine being ignored for a while. All I want to do is make incremental progress every week, grow, grow the user base 5% every week, find people who are interested in the features that we're building, right? These crypto native features that aren't being built on other social networks. And this is the home for them. And over time, I think that the market gets big enough. And unfortunately, I think we have a long-term perspective on that. It worked for Coinbase and hopefully it works with our concern.
This Ryan reminds me of Chris Dixon's take that he had when we we're talking to him way back when about how a 16 Z used to be called an internet fund. And then it just became a fund because the internet took over. And his take was that that's going to be the same thing for crypto. So like first a 16 Z's got like a crypto fund. But that's just going to take over what venture capital is because everything's going to become on crypto. And I think that's the same take that Dan is leading with is like there's a population of highly engaged crypto people that crypto populations up only. And in the future, fast forward five, 10 years, the crypto people are just the internet goers that that we know of today.
Dan, there's a there's a I want to drill down into this client centric design of a decentralized social network. Because my take coming into this podcast is that there's actually no way to produce a decentralized social network without it being a client centric ecosystem. And also something you said earlier is about like we can we can create like a specific thing for sports fans that's built on farcaster and it's just optimized for them. They don't even know that it's crypto. And this also goes back to what we were talking about earlier with like the open model for Twitter or Facebook where there was like an app ecosystem and anyone can build on top of Twitter.
The take the belief I have is that that kind of open design allows for these systems to adapt and flex to the wants of their users. Like when many, many developers can build on top of something they will build something that's fundamentally useful or or or it's not and they stop building it or it is and they continue building it and they continue refining it. But first and foremost, it takes that open model that open API model that open design model to actually allow for developers to kind of believe that there's something useful there that they want to build and then go build it. And in start contrast to the current web to model the work current web to world where like if you open up Twitter on the browser and then you open up Twitter on the phone, it's the same Twitter same thing with Facebook like if you open up Facebook on the internet Facebook.
com or the Facebook the app it's the same Facebook. But like on my intuition tells me that something with a decentralized social network that's a very client heavy ecosystem is that it can change shape. It can change forum based off of how you want it to be or who you are or the clients that you're using because you want to engage with it in a particular way. Can you elaborate on this like unique property of a decentralized social network. And I'm actually having trouble understanding what you mean by a client centric. What does that even mean? Maybe the question for David and Dan too. Yeah, so the way I think about it is we want to build an internet scale protocol.
Internet scale protocol is going to have billions of people using it and they're going to be many apps and services built on top of it. In order to get the many apps and services built on top of it to have those billions of people, you need developers interested in building. Developers are going to build on the thing that makes it easy for them to find users, right? The reason people build on mobile app stores despite the 30% tax and all the ring and roll that they make you go through is when you have a phone it's attached to your body when you're a daily active user of a phone. So it's this amazing platform to be able to build, get someone on, you know, to install your app, then you have push notifications.
And so Apple has been able to bootstrap a huge ecosystem by virtue of actually just getting people to buy their phones and then hold them on their body all the time. So we wanted to take the same approach where the way to kickstart the Forecaster ecosystem is build the initial client, build some of these features that get people into the door. A lot of people didn't find in that interesting, but there were enough people that found them interesting that we were able to kind of create an initial kernel of users, highly engaged users, that once you hit a certain amount of conversational liquidity, so if you think of like DeFi, you need like TVL or whatever, that for social networks is especially a public social network.
If I open up the app, is there something new to see, right? And if it's too stale, then you're just going to have people turn. Whereas if you can kind of beg borrow and steal whatever you need to do to actually bootstrap that initial community, then what you can do is use the initial client to continue to onboard people, continue to grow the total number of addressable users on the protocol, but you're going to have developers start to tinker around because now there's all this data and there are all these users. And so we've seen a whole bunch of different apps built on top of the initial set of Forecaster APIs, which I think are really, really bad, but it just shows you that there's a lot of demand.
People love playing around with this stuff because they didn't have to ask for anything. They can just once they want to build it, they can just go and do it. And it'll only get better later this year when we have hubs and we hit the bar around permission lessness. But I think to go back to this kind of like why we think the client ecosystem is so important to get right, it goes back to your point. Or rather, if you think about what the hard thing is when building a social network, yes, it's coming up with some novel new mechanic and idea. But there are plenty of examples of people having built social networks with the mechanic, maybe five years too early or whatever.
And it's not on the merit of the idea, it's all about distribution and getting users. And so what we're trying to do is dramatically reduce the difficulty in having an app that has social features, people are already going to have an identity, they don't need to be onboarded again. They're going to potentially have all this public data because it's a network like Twitter that you could potentially use machine learning to figure out what their preference is up, right? So I could, if I built that sports client, that's like if you're really into football. And you had already talked about that you're, I don't know, Vikings or a giant's in one of your casts before, I programmatically can kind of go look back.
And then when you open up the app, it's like, you know, ready to auto suggest what you're interested in. And so this access to the like foundational identity, the actual data of like, you know, what you're posting, what you're liking, who you're following, all that information that is available. Another developer can actually use to bootstrap their new app. And so making it easier, dramatically easier for new social apps to be built, because as a user, I don't need to re-sign up for like the whole system. And then similarly, like they, they can actually do a lot of intelligent stuff based on all the publicly available data, gets you to a place where it can be a shared layer of identity of the data.
And then actually the other thing about it is you can now use it for distribution, right? And so we actually just had an app come out this weekend, still early stage called absorb. And what they're doing is they're using the Farcaster protocol to post seven second videos in an old divine. And the way it works in our client, it's going to show up where it says I just posted a video to absorb click here. And we don't even have video in our client yet. We just haven't gotten around to putting that in there. But if you were to open up absorb, it's actually going to show you all the other posts that are on the the Farcaster protocol that follow that format in a very vine or TikTok like way.
And so it's using a subset of the protocol for all the kind of like primitives that it is available to it. But then then it can actually just build a completely new experience. And let's say that app starts to get bigger. Maybe there are plenty of people who use that app who don't even know that it's using Farcaster or crypto. It's just like, oh, cool. This is like some app where you have a bunch of people doing what they're doing. And maybe they come up with some new novel mechanic that's slightly different from vine or TikTok that does really well with a certain sub segment of users. Great.
That's all benefiting the Farcaster protocol. And they're benefiting from the Farcaster protocol as well. So it's pretty simple. I mean, I get all this. And that's why I was kind of questioning David, like why why you call this a client centric design? Because to me, that's not that's it's protocol centric. Really what Dan is saying is we're doing the base protocol so that anyone can build a client on top of it. We're just bootstrapping the very first client in order to get people that's like our bootloader for the client. But the value is really is kind of the protocol. It's like as if Twitter released the Twitter website and the app in the early days and then also a protocol that anyone could build on top of that was completely rugless or it's as if RSS, you know, RSS didn't pick pick up in many ways when RSS was first introduced.
Well, maybe that was because RSS there was never kind of a company developing the first client for RSS and kind of like acting as a bootloader. So that's why I was asking David why you called it a client centric design? Don't you mean like protocol centric design? Yes, yeah. Protocol centric as in there's one farcaster protocol. I guess what I meant by client centric is that whereas Twitter and Facebook are highly monopolistic and don't encourage a expressivity in having many, many, many clients. Farcaster does. And so it's putting clients as like, yo, we need clients. We have clients need to be a first class citizen because there's many different ways to access the farcaster protocol and each client can pick and choose their own particular way of accessing the farcaster protocol and those different choices will be determined by the users of that particular client.
But it's really like farcaster winning like a thumbware in like protocol design. Whereas like in theory, there could be a farcaster protocol and then there could be like the Facebook of farcaster and the Twitter of farcaster. These things can become apps. And the fact that farcaster is a protocol is like, you could have the Twitter app that is built on top of the farcaster protocol. So yes, Ryan, you're right when it's there's protocol centric. But the idea of a Cambrian explosion of specific clients that are built for specific purposes for specific audiences is what's missing from the web to world that makes it so like stagnant and stale and monopolistic.
Dan, does all of this stuff check out? Yeah, and maybe to marry the two together, I would say it's the protocol is only good as the as good as the clients that are available for it, right? Because you could have technically brilliant protocol. But if there's no way for people to use it and use it like in a really highly usable way that they can install on their home screen, right? Because it is actually a battle for your thumbs. Are you going to hit Instagram or Twitter or you're going to hit farcaster or some other app that's built on top of it? That's zero sum. Social like time is a zero sum thing and that's what you're competing for in social.
And so in order to make the protocol useful, you need at least one client, right? And one client that's hopefully driven a whole bunch of users. Then I actually think the protocol is it's kind of a flywheel, right? So it's now you've actually bootstrapped it with some users. Now the protocol starts to be more attractive for other people to build clients. And then hopefully kind of ping pongs back and forth there where is the more clients and more diversity of clients and different experiences you have, the more growth you're going to have for the protocol, the more growth for the protocol, the more people are interested in building clients.
And so I think that is where we have been really really focused and what might differentiate us from others is we plan to work on our client like this initial bootstrapping client for a very long time if not forever. Because I think for us, it forces us to dog food the protocol that we're expecting other people to build on. And so it's inevitably we are going to know the pain points in the frustrations as a developer. It's like, wow, this is really hard. How could we, if we can't do it ourselves or this is a pain they ask for us to do, why would we expect anyone else to go through this? Like they're just probably going to go build somewhere else.
And so by being kind of like putting ourselves as a developer on the protocol, it makes the protocol better because we're developing the protocol. But it also going back to what makes the protocol successful. It's getting users to use the protocol who become addressable users for other apps on the protocol. And so that that's where I think the client and protocol aspects of the designer are critical and they they're very intertwined. I think Web2 users, which I know is everyone can really resonate with this with this with this like natural, this like fixed state of the Web2 world. Like why hasn't there been like a new Web2 app that's come out in the last like decade?
We almost had what was that the the audio clubhouse? We almost had clubhouse clubhouse took us by storm in the middle of COVID. Like we almost had a new one. But then what did what happened? Like Twitter just absorbed that feature into their monopoly of Twitter and took away that like Dan you've been calling these like this cool new mechanism that it justifies like a new like social media app. But then all of these like Web2 companies just like absorb those features into like their like larger centralized stack. And I think one of the things that you alluded to is that as the farcaster protocol grows in users and utility when somebody trigger like thinks up of a new mechanism like clubhouse did that can be appended to the farcaster protocol via a client.
But not actually have to compete with the other clients on farcaster because everyone is contributing to the growth of this protocol. And so it's a little bit like how I view the Ethereum layer two ecosystem. Like when optimism wins and grows users in TVL like their net overflow of that also benefits all the other layer twos on Ethereum. So like that's net beneficial to arbitrum because when our when optimism finds a new user like they're just one hop away from arbitrum. And that's just not true in the Web2 world when Facebook finds a new user it comes out of the detriment of Twitter. But I see this like festival of the commons of farcaster is when somebody triggers events like some cool new mechanism that's brand new that can become an app on top of the farcaster protocol and actually grow the utility of the protocol at large.
And hopefully that is what unlocks a brand new like Cambrian explosion of cool new social features that exists on the internet. Is that where you see? Yeah. And so going back to I would give TikTok is probably the one counter example to the last 10 years. But other than that, I and maybe snap if you want to go back that far, but very few, right? I think the way to think about client competition on something like farcaster. So let's someone comes up at the clubhouse equivalent on farcaster. And we have this initial client and has a lot of users. And so if we try to do the same thing that Twitter did with spaces, right?
The the the managed Twitter had with spaces is not spaces was a better product spaces had more distribution, right? So if you do a spaces for bankless versus a clubhouse for bankless, you already have a huge audience on Twitter, right? And so I can guarantee you if YouTube did a spaces product or I guess do they have it? It's like YouTube live, but something equivalent it would also be really popular because there's a lot of existing users and distribution. And I think what's key with farcaster is, so if someone goes in clones, you know, one of the bigger apps on the ecosystem clones, the kind of up and coming, the one who innovated, they have to continue to keep pace because the distribution is actually put at the protocol level.
So now as a user, if I actually want to listen to the equivalent of like, you know, audio room that is at the farcaster protocol level, I can actually choose the best client. Rather than have to say, Oh, do I want to choose the distribution or the feature set? And so you can actually get both where the client that you opt into has the feature set, but you can still have access to the entire social graph and distribution. And so going back to this idea of monopolistic tendencies, it forces clients to actually have to be extremely responsive to users because otherwise someone's going to get their attention, right?
And so from our standpoint, look, we have 99% market share today because we're the initial client. There are a whole bunch of other clients being built right now. And as we build the ecosystem out and make it easy for users to actually move between clients, I consider it a huge success if we only end up having 10 or 20% share, assuming we've grown the farcaster protocol because that to me is the success of the protocol. And then we're going to end up with a better ecosystem because these apps are going to be competing on feature set, not distribution and like proprietary lock it where I think that that's where we've gotten a lot of the reason things are so stagnant, right?
Instagram had this whole controversy recently where they're adding more videos to compete with TikTok. And you had a whole bunch of users, a lot of high profilers was even being like, we don't like this. Like we want to have it be more like the traditional Instagram where it's photos or stories. And the beauty of a system like farcaster is if the equivalent of Instagram is built on farcaster and has a massive audience assuming it's on the protocol, you can actually go build the traditional Instagram client, which actually someone did. And what did Instagram do? They sued them for access to the API and then got them kicked out of the app store.
Whereas in this system, it'll be like, oh, this is a better client. I'm going to go shift over there. And two examples like this exist in the web one world. So the first is browsers. Netscape starts at Internet Explorer crashes them through a variety of kind of underhanded tactics through bundling. And then what happens is Firefox comes out and Firefox without any huge budget or anything wins a massive amount of market share on the merit of just being a better browser. And the beauty of the web is I can go, you know, there was no lock-in of i.e. for the most part. And then Google comes along and says, well, we're going to build a better browser than then Firefox and got a lot of share.
Whereas now with these mobile app stores, essentially Apple and Google can actually force and actually more Apple than Google because I think Android's a little bit more open here. Apple basically can say we're the only the browser on iOS. Like Chrome is like basically this wrapper on top of some APIs, but they don't get like full system access. And then the second one is superhuman can come along and just be like, we are literally giving you the same effective set of data that exists on Gmail. But we're going to build a product that's so productive for the people who are excited about that. And we're going to charge $30 a month or whatever they charge.
And they can go do that because email by definition is kind of this permissionless thing where I can actually move my my data between different clients. And so that's that's what we want to get to with Fargaster. We're the best clients that the biggest clients are winning on the merits of their features, not some lock-in that they then can kind of extract, right? Like the classic example I always use with Web 2 if it's not even social is Airbnb's cleaning fees. They're able to do that because it's like they built this huge marketplace in brand. And most consumers aren't going to go try to go find the other things that exist out there.
And so then they can kind of just keep turning the dial and not having to really respond to the market versus if you imagine that as a protocol, right? Like the ability to have the inventory as a protocol and then the trust and safety, which is actually the hard thing to go do. But if you can actually get that to a protocol, I mean, I thought about like how you'd actually go do that. But imagine if that was more of a protocol, then you'd have the best clients all competing on it, right? And that's actually in some ways if you think about flights, a much better example where you used to have HitMonk and Expedia and Kayak and Google flights, like people can actually compete on that data set.
And that's not even a protocol, but just like having open access to data, you win on the merits of how functional the software you're building is, not some proprietary lock-in. So this all makes sense. But here's the question I'm trying to understand in my mind, Dan. So why? Let's go back to incentives, right? So Twitter certainly could have in the early days of Twitter not shut off its API, okay? It could have been a bit more standard-centric and protocol-centric in its design. Sure, it didn't have crypto, but it could have adopted for some kind of open standards. It could have made itself more rugless.
But the reason they didn't is because they had back to David's point earlier, they had venture capitalists, investors to please. And the model, the monetization model for social is ads. And so once they established their moat, it was a good business decision to then pick up the ladder and pull it up so that no one could kind of come behind them. They had no incentive to create this protocol behind them to let competitors, potential competitors out-innovate them because they were encended towards an advertising model in order to maximize the ROI for their investors. We understand this. This is kind of capitalism.
And I can imagine, farcasters not using sort of a Wikipedia model where it's purely grants funded. That's another example of how we've created public goods on the internet. So can you tell me about the incentive mechanism? I mean, you've been talking about how you're very excited to propagate the farcaster protocol and not just the farcaster client. But then what further incentive do you have to continue building that protocol, I suppose, or like, don't you want to be the most dominant client in the space? Are you planning to monetize ties this with ads? This is the part I don't understand is how do investors in something like farcaster maintain alignment?
How do you yourself, your team maintain alignment? You were talking about I would love to develop our client forever for as long as I can. Why? What is the monetization? What is the incentive mechanism here that keeps you aligned with the public good and the good of the protocol? Does my question even make sense? Yeah, it makes a total amount of sense. And I think that the answer is one, I can give you the most glib or whatever crafted answer. And this is something that is only earned through actions over a sustained period of time that people actually believe that I do care about the protocol or Varun and I care about making the protocol successful and not just our company's called Merkel Manufactury.
So making Merkel successful. And so the way we approached it is one, Varun and I both worked at Coinbase. So we had some economic success coming out of Coinbase. And when we sat down to kind of like work on potential ideas, we were both aligned on wanting to work on something for 10 plus years and actually have it be meaningful and great to the point where we could look back when we were 80 and say, I'm really proud that I worked on that. There were resilient other companies we could have gone worked on and let her kind of like very nuts and bolts SaaS companies in crypto and probably could have grown it faster and gotten to revenue faster or whatever.
We just didn't opt for that. We wanted to work on something for 10 plus years that we wanted people to actually kind of say that's stupid that you're working on that. Like it's never going to work because I think when you have a long time horizons, you can actually accomplish great things. The second thing is we bootstrap the company for the first year and a half. Not because of lack of investor interest is that we actually wanted to build the initial version of Farkaster and how we wanted to build it the way we wanted to because again, we're aligned on the long term here. When we did raise money last year and we raised money because one, we thought being able to scale the team a little bit more than we had originally planned would benefit the protocol.
But two, I got a sense having been through multiple crypto winters at Coinbase that we were headed into a down market. So we started fundraising end of April early May of last year before things imploded. So we got lucky on the time. But one of the things is we said to all the investors that we brought on board is we're optimizing for the protocol long term and Verrune and I have kind of voting control of the company. So basically we're in the driver's seat around we can build the company the way we want to do it. And our view going back to that kind of analogy is today we have 99% market share or 95, whatever version of that high number you want to pick.
But we have 8,000 people in a beta. The goal here is to get to 100 million a billion people on Forecaster and if we have 10% in those two cases, you have 10 million or 100 million, which is like that, that, that is going to be great. And so I think that you have to earn the trust of the developer community in within the Forecaster ecosystem. That I think will be majorly earned when we release hubs because then people will actually start to believe like wait a second, like I have the same playing field access to the data and APIs that Merkel does with the amount of money that they raise, the fact that they were the first in the ecosystem they have the most users.
And it seems like Merkel is taking their API or their their limited resources in terms of time and staff and building out features like make it easy for an app developer to install. You know, you can easily install a new Forecaster app and actually letting people export to a new client. And so upholding that standard, but I don't think that there's any deterministic way for anyone to believe that other than if we've been working on it for three, four years and and we've consistently made not perfect, but consistently made the right long term decisions around optimizing for the protocol. And so that's that's the best answer I can give you is like I believe it and I want to show the world that I do.
But until until we hit a sufficient scale and hubs exist, I think it's completely reasonable for anyone to be skeptical. Dan, does it? Is it required that a protocol have a token for this, you know, to succeed? So I just, you know, we've used some analogs of the Ethereum Foundation, for example, right? And basically if I were to kind of sum up the Ethereum Foundation and what's done is it's bootstrapped. It's R&D development based on the value of ETH the token. At some point, the idea is that ETH may be ossifies in some way and actually ossifies not the right term. David, what was the other term we were batting around?
It goes into homeostasis. Homeostatic. Where there's less development, ongoing development needed. And so kind of the Ethereum Foundation is almost no longer needed as sort of a shepherd and kind of fades into the background. But they've used the token price to kind of stay aligned with the rest of the ecosystem. So they're very aligned with, you know, Arbitram and Optimism layer two is anyone in D5 because they have kind of that alignment in the token. Is it token needed in order to line parties for something like a decentralized social network? Or what do you think of that? So the way I think about it is the protocol going back to your idea of like how do you bootstrap R&D and continue to have a strong protocol that's independent, protocol on these sources revenue.
And so the way we are going to actually test this is once we move to main net, we are actually going to have two types of forecaster accounts. There's an account that you can kind of register. Think of it as like you get a numerical ID and you own that forever. It's gas only one time. So you could do that once and then then you have a forecaster account forever. But we think that most people who are engaged on forecaster are going to actually be willing to pay some amount of money. We're going to test that at roughly $10 a year worth of Ethereum directly to a contract that is kind of like the protocol treasury contract.
And if we're right about that hypothesis could be wrong. But if we're right, then the protocol is actually going to have a sustainable yearly revenue source into perpetuity. That's something that email doesn't have or DNS or any of these other things that were kind of more academic and then captured by a lot of the large internet companies. And so by having a revenue source of the protocol, now the protocol can actually start to hire its own team and actually get outside of like whether Merkel manufacturer or other big companies in the space are actively contributing the protocol. Like we still want to but like I want to get to a place where the protocol can actually really be independent.
And then at that point, I think it's worth figuring out depending on how many users you have and how engaged people are is how do you actually decentralize the core governance of that strategy away from the initial team. And so I have like kind of this very pragmatic or give myself credit on the pragmatism. I'm rooted in pragmatism most of the time. And my view is if we focus on the decentralization of the governance too early, we're going to spend way more time talking about like what the systems and processes should be versus actually making progress. And the thing that matters is like we need to have a million or 10 million people who've signed up for this thing improving people are willing to pay for the protocol and all this other stuff before we can worry about getting to those other things.
Again, people might say you don't want to never want to lead with the algorithms. Right, just going to move slower. And so for like there's a term in open source software development which is not what we want to do, but benevolent dictator for life. That's what Linus Torval says. Right. I think ours is benevolent dictator until product market fit and then actually turn it in over to the protocol. And look, some people will say, oh, well, what's your incentive? You're going to get there and you're going to want to hold on to control yourself. Yeah, that's what would be the big corner response.
But the reality is, if I wanted to do control, why would I be spending all this time on the protocol stuff? Like I could be moving way faster. 50 80% of my engineering versus on the team right now are spent on the protocol. And so like this is the thing. I don't have to convince everyone this. All I have to convince are an increasing number small, but but steady group of people who are like buying into the vibe on Forecaster. And that's fine. Because if we can continue to show that we can grow it, we will get more and more permissionless over time. And I'm fine with that. And so I'm not walking around saying Forecaster's permissionless and all this if that is aspirational at this point.
But when we hit that bar, I'll be very happy to kind of pound my chest and say, Hey, we've had we've had these these milestones and we do have actually a lot of decentralization and it is rugless. But until then, it's more aspirational. And it's like we're actively investing resources in it. But I don't think it's intellectual and honest for anyone just to be claiming that if you haven't actually hit the threshold. No, that's cool. That's pretty grounded. Yeah. That was kind of the root of my question, right? Because like what you're looking for if you're if you're doing a protocol is sort of a public sector and a private sector, right?
And both need to be funded. And if we think about kind of you know, governments, everyone's familiar with obviously their their nation state. They're familiar with taxes and then they're familiar with businesses. And so what you're saying is basically the forecasted protocol could be configured in such a way that it has some sort of lightweight tax type system to pay for the public goods, which is maybe when you register your name or something was, you know, $10 in each per year. And this goes into a fund. And if you multiply that by large number of users, then you actually have a pretty good tax base in order to fund the public goods, the, you know, the roads and the hospitals.
Of course, not those things literally in this case. But the the public goods that everyone will share that what you haven't figured out, which to be fair, no one in crypto has figured this out is how to actually govern that treasury and make sure that it's managed effectively. Of course, like in the nation state, quite famously have large bureaucracies that sort of kind of do this work. Maybe crypto will develop something around that though. That's kind of the hope. I mean, we are working on these get coined type governance quadratic voting types of things and something might come along. But that was the root of my question.
So in in in a fully realized forecast or you have this public sector, but then you have this private sector of all of these various clients that are creating businesses based on the public infrastructure and are competing with one another in a way. But they're also contributing towards the the network of fact of the protocol and even their contributing tax revenue to the protocol to the extent they're able to onboard users. That seems to me like a pretty grounded approach and does not necessarily require a token at the base layer, although maybe that's a design strategy as well when you think about governance.
Very helpful. Thank you. Yeah, and I think in some ways it's an improvement on web one, which you always have to be careful about because like obviously web one is incredibly successful. But I do think if you think about whether it's, I don't know, HTTP, DNS, SMTP, the fact that those protocols don't have a built-in revenue source despite billions of people using them. You have to run them as non-profits that don't have any funding whatsoever, basically. Or get funding from the big internet companies who then get to decide what the standards are and what we want to want. Things like that. Right.
And so like a good example of like web off end, which is like this new like touchless way of signing into things. They don't allow like arbitrary signing of like ECDSA curfers or whatever. And if you want to actually get that done, it's this incredibly bureaucratic process run by kind of middle management like the biggest browser manufacturers. And in a world where governance is a little bit more decentralized and then like there is actually a revenue source directly to the protocol. One, the protocol hopefully is independent so that it can actually make its own decisions both best for the protocol.
But more importantly, if you actually get to a good solution for decentralized governance, if you want to have more influence there, you're going to figure out, okay, this is the path actually doing that versus this kind of opaque big company like the fact that Microsoft, Apple and Google can kind of get together and be like, maybe we won't add this feature to the web browsers like. And people will be like, oh, that's this conspiracy theory. And then it turned out like when Steve Jobs was around, they weren't recruiting engineers from each other because like they big companies do do that. So I don't know, like I think call me naive, but I do think having these public goods have a funding source is is a improvement over the status quo.
Well, actually, I agree. It's so long as we can govern them effectively in an uncruptible way. I guess just last piece here, then you back to kind of the business model for a second. So remember that public private kind of relationship we just described. And the clients themselves would be the private sector here. And I suppose you would be open to all sorts of new models on the client side that web three enables. So back to my early example of Twitter, the only model, the only way for them to appease their shareholders and make money and return value to the investors, which is necessary, by the way, let's not live in a naive world or investors invest in things without expectation of maximizing their upside.
But you could have things that aren't maybe ad revenue funded, for example, in the clients. So NFTs like a client, you can imagine a forecaster, maybe it gets some kind of transaction fee type revenue for the amount of posts with the amount of NFTs that are sold inside of the inside of the client or you know, maybe there's some other token funding apparatus that is more web three native and less eyeball centric. We, you know, make money based on the number of eyeballs we get in this system. So I guess that private sector is even more diverse than the web three social, which just like depends on attention as the currency depends on advertising as the sponsor of everything where all users become the product.
Yeah, so I fully expect there to be clients in the VARCaster ecosystem that use advertising. It's the reveal preference of billions of consumers they like free, and they're willing to have their attention taken by an ad. And I don't think all ads are about either. But I think what we're planning to do with our client, so I can speak to that, is we're going to have a freemium model. So we're going to actually have subscriptions. And our strategy is to keep our team extremely small like from a core team. So our general expenses are going to be pretty low. And we think that within a reasonable number of years in terms of growth, we continue to grow 5% every week, we, we will get to a place where we're break even or profitable based on some percentage of users willing to subscribe for the kind of like pro subscription, right?
And I think subscriptions in 2022 are much more normalized for consumers to say, hey, I'm actually spending a lot of time in the SAP. I'm willing to subscribe a few dollars a month to the more kind of like premium version of that. And so that's something we're going to test. I think you're spot on in that if your feed has a bunch of stuff that's happening on chain, there's a natural economic model with that. And one example I always bring up, a lot of people don't realize Apple gets paid $20 billion a year by Google for the default search placement in Safari on iOS. So that's probably their highest billion 2020.
And it's got to be pretty profitable. It's their highest profit margin. It's no cost. It's literally, it's just, you know, you can either be duck duck go being or Google and Google pays 20 billion. Because you can change it. It's not like it's like the only option. And so that that's an amazing model when you have some amount of scale, right? Like, you know, obviously Apple has a lot of scale. But Firefox's entire budget basically is just the fact that they have the user base that they do and Google pays to be the default little search bar in Firefox. And so and now you realize why Chrome is so valuable because they don't have to pay themselves that.
But in our case, we have the NFTs. You could imagine our client saying, okay, what's the default? If you click on a link in an NFT, where is it going? And and now you potentially have different marketplaces being willing to bid on that. Or if you mentioned a cryptocurrency, it's like, where are you going to trade it? And so I think the the the idea space for this kind of like transaction based commerce is something that was frankly, and maybe Elon fixes this with Twitter. I don't want to count them out. But Twitter is this kind of like feed of information and the ability for you to like tap in two things and then buy I recommend did, you know, someone tweets a link to the wire cutter like why isn't there a deal where you're surfacing that product review right in the feed?
Not not oppressively. It's like I'm interested in this click. It opens up and then I can actually do the commerce right there. I think that there's just a whole idea space that can happen there. And then the beauty is people are inventing new things in crypto all the time that are going to have natural built in economic models. And so what I would like to get to is a place where basically we never have to have ads because we figured out other ways to monetize and also kept our cost structure really lean. But even at that case, like you can imagine maybe when we have 10 million or 100 million people, there's a tier of user that's like, Hey, I actually just want the bare bones completely free.
And then maybe you're showing them some some simple simple app. But like no by no means are we interested in doing that in the near term. And here's the beauty of the protocol. If that's the wrong decision, some other app is going to come along offer all the same features they're going to offer an ad based product ad based client. And the reveal preference of users will that they'll use it. And then we'll have the confront and potentially change that. So I think that that is a world that I want to live in where we can run these experiments and let the market decide rather than, you know, you make a decision 14 years ago.
And actually, I think with Twitter now that they're not public and they're owned by one person, they're actually having an opportunity to rethink that business model. And if you were to ask me what is a Twitter that is not on by Elon today, it's they made the bold decision early on that they weren't going to monetize like Facebook. And then instead, they were going to say, Hey, we're going to monetize better than any online newspaper, right? Because we know these people who are really into Twitter probably willing to pay $20, $30 a month. And maybe they didn't have 6,000 people. They had 1200 people or whatever.
But I think from first principles, I think you can get to a reasonable profitable business with a subscription model that you don't even assume most people subscribe to. Like assuming you just are really, really efficient in terms of keeping your head count lean. Dan, this has been such a great exploration into the into the world of the future of social media networks. And I'm pretty sure listeners or imaginations are probably running wild right now. There's a few more conversations. I just want to quickly check the box up just to as we wrap this up. One is like, what is the actual on-chain footprint of this thing?
Like it's not it's on blockchain, correct? And I want to know I want to make sure that that gets communicated. But what are the actual crypto things that are about? What is its footprint in the world of crypto? Yeah, so you have two parts on-chain. Today it's on Ethereum testnet. It's on-girly. And it's moving to may net hopefully halfway through this year. And you sign up, you get an FID, a Farcaster ID. And it's just an incrementing number. So if you're the 10,000 user, that's your number. That's largely abstract away from people. But that's actually critical for the whole system to work. Then there's an optional $10 you're kind of like user name, like kind of premium account.
But but effectively that those are the two primitives. That is also an NFT. So that can actually be permissions they put into anything in ERC721 works with. OpenC, you can put it on, I don't know, pseudo swap, you can use a party bed. Completely composable. So wait, my account is an NFT. Yes, I hope you got to get user name there, David. Well, I hope I don't lose it. No, no, you won't lose it. We're any any active user during the test. And we're going to migrate to main net on on our done. Okay. But then but then it's $10 a year, just like it would be if you own a domain to pay and you paid directly to the protocol.
But yeah, like that, that is the core primitive on Ethereum, relatively simple set of smart contracts. But the core of the whole system is all mapped to that. So if I post something from DWR, which is my user name, every time that happens, there are servers that are effectively checking Ethereum to make sure that the public key, which is an Ethereum address is that, you know, matches the private key that signs that content and it all just magically works. Cool. And then the other one question I want to talk about is the world of content moderation and spam prevention as someone who has been absolutely hounded by Twitter bots lately, how does spam prevention work in a decentralized network?
Like does it is aren't you just opening yourself up to spam attacks? So one of the things again, small team and we're just trying to focus on what we think are the most important priorities first is it's it's gated to sign up. And we actually took inspiration from Facebook on this. Facebook is still the largest social network in the world. And for the first two, two and a half years, it was gated to.edu email addresses. And our our approach is we need to get hubs out. We need to get to a theory. I mean, and then the second half of this year, it's figure out how do you get to completely permissionless on pretty.
And then then I think you start to hit a bar where okay, the protocol is actually permissionless. But as a result of that, we haven't had any spam issues for the most part because everyone right now has to come through my Twitter DM. And that's not sustainable long term. We're moving away from that. But I think that that's allowed us to a keep the community small and incrementally grow rather than have it grow too fast. People kind of churn and then it feels like a ghost town. And I think the long term solution and this is not fully big. So there's a little bit of a punt here. But I think the the high level answer is something like PageRank, which you know, powers Google is as the social networks develops and you've kind of had time for kind of the active users, you can actually start to map any given user on the network, how they interact with all the other users.
And it gets very difficult for a spammy account to you know, penetrate the core of the network in a way that if you basically get good on your kind of machine learning data science, you could potentially even publish that information so that any client out of the box can get actually decent spam filtering with the caveat that that client can choose, hey, actually I don't think this account is spammy or whatever. So they can they can modulate how they want to approach content moderation. I also think that what's important is similar to email, which everyone uses content moderation and email, we just call it spam filtering, it's a client level decision versus the network level decision, which is it relatively straightforward.
The only thing you can store on Farcaster is text like 320 character little tweet like things. And so the amount of stuff that you can actually put on there that is really bad from like an image standpoint and video content is relatively limited. That's that you can still put some nasty stuff in 320 characters. But inevitably what you're going to have is you're going to have clients are going to do the same thing they would do with emails is they're going to say, okay, we're going to not allow hate speech or abuse or all this other stuff, but that will again be a client level feature, which is which is how the the internet works, right?
It's like how the web works like Google doesn't have to index your web page, but that doesn't mean your web page went offline. So even if it's full of deplorable stuff, Google just takes it out of the index, you just get no distribution, right? And I think that that is the content moderation gets wrapped up in the fact that today most of the platforms we're talking about from content moderation are doing the hosting. They have the identity for their given platform and the distribution. And so thus they have to do content moderation because then people are like, well, why are you hosting this video or why is that tweet still up?
Where the reality is if you have a marketplace of clients, first of all, there's no single on the algorithm, right? Like each client's going to have a different way that they're implementing that. And B, I think the average user does want some amount of content moderation. But like when we have a marketplace of ways of solving it, we call it spam filtering, right? It's like it's focused on low quality versus this like, is this fundamentally something that should be posted or not? I don't think most people are that worried about like the fact that there are some really kooky websites out there on the internet.
But if they don't have any distribution, it doesn't really matter. Yeah. And that kind of just answers my next question, which was like discovery ability in the algorithm, right? The reason why we bankless post videos to YouTube is because the YouTube algorithm allows us to be discoverable. But I'm going to go ahead and guess that the future Web 3 protocol version of this is the same answer that it was for spam prevention. It's just like that is a vector upon which the client's compete upon and discoverability and whose algorithm is best, whose algorithm is most enjoyable by people and makes them feel good is probably something that the client layer competes upon.
And it's also probably the way that farcaster also becomes like compliant with the nation state regulations is that clients are going to have to like make sure that no illicit activities happening on them. But that doesn't this no longer responsibility of the protocol like the protocol is permissionless. But clients still got to make sure that the activity inside of the specific client is like not going to trigger the iron of regulators. If it's like, you know, illicit trades or you know, all the bad things that happen on the internet that is still something that the clients have to optimize for and work on that the protocol doesn't leaving the protocol still like peer and permissionless.
Yeah, it's very similar to Coinbase, right? Like Coinbase has to keep things clean on their end and they're subject to the laws in the countries they operate on. But the blockchain is a protocol and what happens on there, they have to take care of what interacts with them. And I think that is the right approach because it works really well for the web and email. And I think we have the possibility if we can achieve what we want to achieve with farcaster for actually making the same for social. And I do think on the algorithm side, if you view that like let's say that there are five or ten major apps and they all have a slightly different algorithm, instead of hacking the algorithm, which is a lot of like what happens on these like monopolistic networks, you actually get to a place where the best content across a variety of different algorithms is probably actually what succeeds, right?
Because it's hard to fake quality versus any one algorithm probably can be gained, but it's harder to fake ten at a time. Then this could probably be this question about to ask you could probably be an entire podcast. So we don't want to do that, but it's one thing I just wanted to cover before we close here and we're getting ready to. So we've been having on bank list this conversation about identity and a quote from kind of bank list, the bank list thesis is phase one, we take back our money, phase two, we take back our identity. And social certainly has a role in web three in phase two, I think.
How does this fit with the story of identity? How does farcaster kind of feed into that? We just had the folks from signing with Ethereum on the podcast. People can look at that in our archive, bank list listeners in like maybe 45 seconds or so. Could you could you help us answer that question? Does this play a role in decentralized identity in that story? I think it does. I think look, you can today and betalog is a great example of this. You can have your own website and that can kind of be the primary pointer to your identity and internet. But the challenge is your website is going to have a very hard time unless in a very rare cases getting distribution.
And I think more people will be willing to actually start with a sovereign identity, something that they really do control and can kind of instantiate their identity or display their identity in anybody they want if they can associate it with distribution. And so I think with farcaster or other kind of decentralized identity approaches, if you basically can have that be the pivot point, assuming that you can then get distribution for whatever you're trying to say or achieve in the world, I think that is the most promising path because distribution is the thing that people care about. It's the reason as many of the people who are interested in decentralized social media and all the things that farcaster represents that they still continue to use Twitter is because they already have large audiences there.
And I can't fault them for it. That's actually a hard thing to do is build an audience and distribution. And so if we can do our part in making it easier to have a decentralized way of connecting with a large group of people, then I think you'll start to get sovereign identity. This is great. Bankless listeners hope you enjoyed this conversation with Dan. He's building a decentralized social network protocol called Farcaster. Of course, it's one of a couple that we will be a few. I should say that we're exploring on bankless and we certainly wish you well on your endeavors. Dan, this has been a fun conversation.
Thank you for introing us to decentralized social media and also to Farcaster. Thanks for having me. I'm happy to offer a bankless listener as a Farcaster account. Are you sure about that, sir? Because that's a lot of people. You're about to get a lot of hearing about this. We'll keep it to the the bankless nation discord. The first 100 people who send me a message with the word Arizona in it. I'll put a little filter in my Twitter DM inbox. I'll know they've listened to the end of the podcast. Dan, is it okay if we beep that word out and only have that go out on the premium feed? Oh, there you go.
That's the benefit of the premium subscription. So we've just beeped the word out. So if you want to unlock that beep, then go become a bankless premium member and we'll save Dan's DM. So he doesn't hear from two minutes. The full bankless community. That's real proof of work. I guess that's probably not the right word. That's awesome. Well, we appreciate it, Dan. This has been a lot of fun. Thanks so much. Thanks for having me. Risk and Disclaimers, guys, gotta end with this. Of course, crypto is risky. You could lose what you put in, but we're headed west. This is the frontier. It's not for everyone, but we're glad you're with us on the bankless journey.
First, we take back our money. Then we take back our identity. Thanks a lot. Hey, we hope you enjoyed the video. If you did, head over to bankless HQ right now to develop your crypto investing skills and learn how to free yourself from banks and gain your financial independence. We recommend joining our daily newsletter, podcast and community as a bankless premium subscriber to get the most out of your bankless experience. You'll get access to our market analysis, our alpha leaks and exclusive content and even the bankless token for airdrops, raffles and unlocks. If you're interested in crypto, the bankless community is where you want to be.
Click the link in the description to become a bankless premium subscriber today. Also, don't forget to subscribe to the channel for in-depth interviews with industry leaders, ask me any things and weekly rollups where we summarize the week in crypto and other fantastic content. Thanks everyone for watching and being on the journey as we build out the bankless nation.