3: Dan Romero - Why Information Should Flow on Protocols

[00:00] Hey, I'm excited for you to listen to this episode with Dan Romero, co-founder of Farcaster. [00:05] We covered a wide range of topics that I think you'll enjoy regardless of your familiarity with crypto or Twitter-like social networks. [00:13] That said, the middle section of the episode goes fairly deep into the weeds of Farcaster and crypto. [00:19] If you're less interested in that depth. [00:21] I've included markers in the show notes to help you navigate. [00:24] Enjoy. [00:25] You've had a... [00:26] I guess eventful week, eventful month, [00:29] Bull markets are good for forecaster. [00:32] You also are pretty... [00:33] Level set most of the time. Anyway, how are you feeling? [00:36] I'm feeling pretty good. It's been a [00:38] bit before... [00:40] I guess like if I think back, we had a good early part of the year and then it was pretty slow. [00:45] I think the crypto market's been pretty slow for the last two years. So having signs of life is always a good thing. I was mentioning yesterday, we were just talking about this on a Twitch live stream, my first time ever appearing on a Twitch live stream. Maybe not the last. Yeah, exactly. I think Brian Armstrong has like really good perspective on this kind of stuff where it's kind of the highs are never as high as you think they are and lows is never as low. [01:09] And I think if you're going to make it in crypto, you really have to try to keep his even keel. And so I really try not to over rotate one way or another. But. [01:18] I am a little happier if the Dow chart is going up. I definitely think that that is an impact on my mood every day. I think I can... [01:27] at least every time I see you fairly, which isn't that often, but periodic enough to confirm the

[01:32] stability of mindset. Okay, so I want to start today not explicitly on Farcaster, but on this notion. So Peter Thiel has this famous line, we were promised flying cars and all we got is 140 characters. And for a long time, the notion in Silicon Valley was sort of [01:48] confirmed especially like 2019 era it was like let's work on something serious let's work on hard technology let's work on things atoms whatever it might be we're now sitting here in november 2024 and it sort of seems like 140 characters is pretty important [02:03] Well, there's a deep irony in that the guy who's building the equivalent of a flying car, you know, Starship is probably as close to flying cars as it gets, happens to also own the 140 character network. Yes. And one might argue is having a harder time making the 140 character network. [02:19] work that these amazing feats of, of kind of engineering and physics with, with, [02:24] SpaceX. Yeah. So I do think Thiel is always really good about saying something that kind of is provocative. Get 50 percent of people to kind of say he's wrong, which I think usually is a sign that you're you're more contrarian and or it's a good statement. But. [02:38] Look, I think that there's an argument to be said that – [02:43] Elon's [02:44] version of Twitter and the support for Trump and that kind of [02:48] ability to influence the algorithm relative to the status quo that existed a few years ago is now going to unlock the [02:56] a whole bunch of regulatory change in the United States, which will make the world of atoms actually move a lot faster. You could you could say that, you know, having a Department of Transportation

[03:07] that is extremely pro self-driving, which very well could be the case over the next four years, gets you to the flying cars. And so you need it. You need 140 characters before you can actually get to, you know, the flying cars. So I think obviously Teal was early on that bet. Right. 2016, he he was making that better and said he kind of had lost it. [03:37] as an angel investment. He's eight years in, and if J.D. Vance happens to run in 2028 and ends up winning, that's a 12-year to IPO from someone who is directly working for him to having the most important job in the world be someone who's a direct accolade of you. And so this goes back to this just like kind of long-term planning. And I think both Peter in the stuff that he does, [04:07] When you actually have a goal over a long period of time and you're willing to kind of go and work on it, you can achieve it. [04:13] Elon and Peter, in this case, to me at least, obviously Peter was very early on the Trump stuff specifically. Interestingly, though, Peter, at least historically, and I think you could probably say he was [04:23] has generally not been super pro public messaging and, [04:28] in some sense is like maybe underrated direct attention and direct distribution maybe that's implied in the quote maybe it isn't elon on the other hand quite the opposite somewhere along the line 2019 whatever figured out with tesla owning the distribution matters a lot there's one view

[04:42] And I might be generalizing this too wide, but there's one view that says, yeah, attention matters and cultural reception of things, distribution, all these things matter in the short to medium run. But in the longer run, 10 years plus, technology is just sort of this inevitable way that's going to happen. My read of the situation, especially given Elon's calculus and what he's spent the last three or four years focused on, is that. [05:06] We're actually seeing in real time a case against the sort of technological determinism of flying cars, self-driving cars, whatever it might be. Rockets. These things are just give the science and technology enough time and they will come around. You don't need to worry about. [05:19] what people think, you don't need to worry necessarily about whatever you wanna call it, distribution, propaganda, attention, et cetera. Am I drawing a line between Peter and Elon in this way? In a way that feels, you can make a case that actually, if it weren't for Elon, [05:32] Effectively aiming to control the airwaves and use, to some degree, his attention to swing this election, Peter would have been wrong, despite being... [05:41] so ahead of the curve ideologically otherwise. [05:44] I'm making a few jumps. Yeah, no, I think that's the Foundation series. Familiar. Okay, so Harry Seldon has this concept of, you know, it's like psychohistory. It's like macro... [05:54] Big, big predictions and any one individual doesn't matter. The great man of history doesn't matter. It's like these kind of like big swings. So you could say, though, if Peter's betting on that, having Elon show up as the path dependency to actually have this whole vision manifest, you know, Trump turning his head during the assassination attempt. You saw those memes this week of like, you know, World War Three versus Occupy Mars. And you're talking about inches. Yeah, I think that's a fair point.

[06:19] kind of, [06:20] critique of it and saying that if you don't have Elon, maybe none of this happens in [06:24] He ends up being wrong. But I do think I'm a big believer in the great man of history. [06:28] in in the sense that i do think you have these exceptional individuals who actually push civilization forward [06:35] For better, for worse in certain circumstances. And I think if you just read the Elon Musk biographies and just realize like how close Tesla was to failure and SpaceX on multiple occasions and to think that, you know, where we are now, right? [06:53] Tesla's kind of leading this whole electric car revolution. [06:56] And then SpaceX is catching rockets with chopsticks and Starlink is, you know, ubiquitous internet anywhere. Like, yeah. [07:04] I think anyone who's actually been in the driver's seat of trying to build a company, [07:08] regardless of what you think of Elon's antics, politics, personality, has nothing but deep respect for the level of commitment and accomplishment that he has kind of made with those companies. And yes, there are all the smart people who work there and it's a collective effort, but you can't just assemble that group of people and assume, either the case of Tesla or SpaceX, and assume that out of that room will come these companies. It's very, I think, founder and leader led there. And [07:37] And so I do think you need these individuals. And to a certain degree, I do think Trump, [07:43] You know, he's got his own set of quirks and flaws, but... [07:47] relatively easy to lose the election in 2020, January 6th, all that kind of

[07:53] stuff that fell out of that and for him to to [07:56] Then in 2022, to have the election or the midterm election go really poorly, people were saying that night, it's like, this is the end of Trumpism. And he got back in the driver's seat and campaign and won the, [08:08] nomination for his party and [08:10] And so I think like, [08:14] To go back to the – it was Teal right in his bet. I think part of it is you are betting on individuals in the same way that with a startup, you are betting on the founders, right? Like early-stage investing – [08:23] I think the media likes to make of it as like this brilliant idea. And you and I both know, but most of it, it's not a brilliant idea. And it's kind of like there's a certain level of, you know, naivete of like just the founder being a smart, energetic person. Just ready to wander for a while. Yeah. So I think in that sense, I think Teal very much understands that. Yeah. Like it's like a bet on the person, bet on the founder. I mean, literally the name of the fund is Founders Fund. Yeah. [08:53] working on something squarely in that. One of the... [08:56] One of the ideas that you've shared with me that I kind of find most interesting that ties into maybe a little bit of the motivation around Farcaster is this notion of if attention does matter, at least on the short to medium term, maybe on a 20 year time horizon, 30 year time horizon, technology is going to come anyway. But if attention really does matter, who controls the. [09:15] information flow matters a lot too and whether or not somebody can control information flow. And people may be familiar with this idea of bring your own algorithm. Jack Dorsey has talked about this. Elon Musk has talked about this. I would argue they put more lip service to it than sort of substantively working on it. Although Elon, you could make a case, cares a lot about controlling the algorithm. He just decides that he's going to control. Yeah, he's bringing his own algorithm. Exactly. And obviously part of this too is a question of state and revealed preferences,

[09:45] Do people actually care about controlling their algorithm or is that just what they say they want? It turns out people want engagement. And frankly, the answer might be that 90 to 99 percent of people don't care. That said, you shared with me a metaphor that I found really interesting as maybe a lead leading point into Farcaster, which is this notion of how we get information in different contexts and specifically how the improvements may be on how we've been able to get information on health and nutrition. As an example, the last 30, 50 years, whereas in one world, [10:13] you kind of trusted the food pyramid and like whatever the government told you about health and nutrition and that was kind of all you had. Now you have a world where you actually, thanks to the Internet, can go – [10:22] choose your own adventure in a sense. And yet with information flow in a place like Twitter or most of our news feeds, it's like you're walking into the health food store and they're blasting cocaine and sugar in your face. Can you talk a little bit about kind of your initial frame for this? I'm going to ask specifically about RSS Plus in the beginnings of Farcaster, but why you think this is an important problem? [10:42] Yeah, I... [10:43] So to the degree that you brought up the food pyramid, obviously what you put in your body has a huge impact on your overall well-being from a physiological standpoint. But I think like what you put in your brain also has a pretty big impact on kind of like your mood day to day, which I actually think that then itself has an impact on a lot of other things in your life health wise as well. But also just where in a world that's so driven by knowledge work. [11:10] what you put into your brain over a long period of time, [11:13] is going to have a pretty material impact on your ability to succeed in a very kind of knowledge-rich work environment. Maybe even who you are. Absolutely. I mean, you know, political beliefs and all that stuff. So I think the –

[11:28] I've been addicted to Twitter since 2007. And I think part of it is, you know, I've kind of talked about this type of personality that like just kind of craves that stimulation of like new new information. Like, what can I learn a little bit about the world? That's actually one of the reasons I don't think Twitter has been as successful as like a mass market consumer thing. [11:47] I think a lot of people... [11:49] Don't find the like terminally online poster in group joke that is like highly memetic relative to literally whatever's happening. [11:57] on the internet that day, right? Like the AILA, [12:00] chart right that everyone knows and the people on this podcast that can visualize that charts like congrats like you're terminally online like there's a lot of context that you need to have to be able to eugene way has this old line where it's like there's no lead up to the joke anymore it's just punchlines yeah so so the the thing is and there's actually a really good book on the kind of like history of [12:21] of how written communication and increase in literacy has changed Western civilization, the weirdest people in the world. Yeah, yeah. Joseph Heinrich. Yeah, so I highly recommend reading that book. [12:31] But but you can kind of extrapolate like Twitter is the extreme of that. And and the weirdest people are like five percent weird people, Western educated. Yeah. It's like five percent, I think, of people or so. Well, obviously, there are, you know, whatever percentage of people who live in the West. And then there's like, you know, the people that qualify on all those things. Right. And I think the. [12:51] The way to... [12:52] Think about like something like Twitter is like you've just now injected a collective consciousness of the world into your brain. There is really no editorial component other than who you follow and what the algorithm is willing to show you. And a lot of you on posts these days. Right. And so we can kind of get into that. But.

[13:08] Just contrast that to [13:09] If you're the same type of person, but you were 30 years earlier, like where are you getting your information? You're reading a daily newspaper that in some cases people were getting, you know, they had an evening news. You could get two newspapers a day. There's the television. You have the radio. [13:26] Maybe as a New Yorker. Yeah, yeah. It's just like the quantity of information that was available to you is just not – [13:33] there and so [13:35] I think... [13:36] In a world where you now have total abundance of information, too much information, the control of what's being shown to you, because time is... [13:44] zero sum, that's really impactful. And it's kind of crazy to think that [13:50] Up until recently, because I actually think Elon showing up at Twitter has fragmented the ecosystem in a net positive way. Fragmented. Tribally or politically? Oh, I just think in terms of they're at scale or growing decentralized social networks. Yes. That allow for client choice, which we can kind of talk about in a second. But ultimately, that's algorithmic choice. And so then I think if I was to fast forward. [14:13] 10 years from now, we're going to go from a world where – [14:15] kind of it's Twitter, Zalgo, [14:19] Like, yeah, you could use the following feed, but they seem to always snap you back to the for you feed and the fight videos and Elon posts or whatever. And critically, you have no control over the algorithm in the for you page. Yeah, absolutely. And my perspective at this point is actually the average even a power user, like they might like the idea of knobs and they can kind of control it, but they don't actually know what they're doing. Right. Whereas I think you're far better to actually have.

[14:42] It's almost the best way I think about it is like a newspaper. Like a newspaper, the Wall Street Journal and the New York Times are effectively covering – [14:49] Some... [14:50] Shared reality. [14:51] For the most part, right? It's like if there's some something happening in the Middle East, both are going to run a new story. They're going to cover it differently. They're going to have different sources, different facts that they prioritize and, you know, different political agendas, even if they're both claiming to be impartial. [15:07] But the point is, is that there is top down editorial judgment that happens at The Wall Street Journal versus The New York Times that results in different ways. [15:15] Just like vibes. Right. And over time, I think that they've actually gotten closer. Right. Right. And this is something Mark Adresen always talks about is like 1950, you could actually find. [15:24] Different university like, you know, the history department at Yale and Harvard actually probably had different. [15:29] Whereas the 2020 ecosystem. Yeah. So, so, so you, you've had a homogenization of a lot of these kinds of institutions. Yeah. [15:37] And I think that that's why naturally the – [15:40] kind of terminally online info for type is on Twitter, A, for the real-time feedback, but the ability to actually kind of follow [15:48] you know, the weird, a non-internet or a Twitter account that you can find it and like go deep on a topic on any topic, right? Whether that's, you know, [15:56] genetics, which can get [15:58] pretty spicy if you if you kind of like dig into some of the details and that's like a whole subculture on twitter but you can also be in like the nba right like or you know political forecasting twitter like where [16:09] You know, 30 years ago, at best, you're going to get like one story a day in The New York Times talking about polling. And instead, now we have like Polly Market mixed with Nate Silver fighting Keith Rappoy over like what the election result in Florida. All that's available to you. And so I think...

[16:27] a terminally online person is willing to kind of like work through some of the clunkiness of the algorithm. Well, and this is very Twitter like 2014. The only way to use it was you had to wade through the mud of all of this. You had to really put in the work. Right, and I think even like – [16:42] serious Twitter people today still do that and then they get they kind of like get through the algo they've kind of turned the knobs and tried to figure it out and it's like okay you turn off the retweets for that really noisy retweeter because that ruins your feed how many people are actually doing that but my point is is that most people don't do that yeah and so this is actually even most power users sophisticated people yes and having now built one of these apps myself like I very much realize that [17:07] Any feature you're going to do, max, you're going to get 5% to 10% of people to use it. And that's, like, a good feature. Most is going to be very low. And so – [17:16] The more sophisticated your product gets, the less likely actually any of these features are going to get used by your consumer, even if they're sophisticated. And so what really matters is like what's the default algo? And so I'm of the mind that I could give you all the knobs you want for your algo. [17:34] And you might take the time to set it up once, but then you actually probably won't want to have to set it up again. And so then you're stuck with that algal. Yeah. [17:40] Whereas I think a way better version of this would be if there are 15 different apps and there's the this this is the app that looks literally at the same graph of data and it makes me smarter. [17:52] where it's like it uses AI to remove. Yeah, I have one chance for an express preference at the very beginning. Yeah, yeah. Or even better, there's no preference that you get to pick. And it's just like, this is what we do. It's like, our whole goal is just like, this just makes you more intelligent. Like the economist. The intent is before I, in choosing the app I choose, I'm actually. Yes, the brand is the outgo. And that's the point of the economist, right? It's like once a week, you don't need to get caught up in the day-to-day news. We just tell you what was important. It's almost like I'm choosing paleo,

[18:22] And to the degree that you want to flip over and get the random fight videos that you would see, you choose the more degenerate. It's literally like opening up Comedy Central. Like television channels are another good concept for this, right? Totally. You know, an average Zoomer would be like, what's television channel? But like. [18:38] Growing up, you could watch MTV if you wanted to see music related stuff and flip over to ESPN. Not to mention tribal TV channel. I'm curious to get to that a little bit later in terms of the different Twitter variations and the ways they've become more tribal to double click a little bit since you're already starting to talk about it. [18:55] You wrote an awesome blog post called RSS Plus Pre-Farcaster. Do you want to talk just a little bit about one of the things you highlight in the RSS context? One quote maybe I'll read because I think it expresses both the ideological element of RSS and also the fact that it's a withered technology to steal from Gunpei Yokoi of Nintendo. And it's from this awesome, like, rise and demise of RSS posts. [19:25] programmers with tech blogs and occasional journalists. Though, of course, some people really do still rely on RSS readers stubbornly, [19:32] adding an RSS feed to your blog in 2018 is political statement. That little tangerine bubble has become a wistful symbol of defiance against a centralized web, increasingly controlled by a handful of corporations, a web that hardly resembles the syndicated web of Warbac's imagining. I'd love to, one, if there's anything reflecting on the RSS Plus piece, but two, as we actually get into what Farcaster is and kind of the founding ideas of Farcaster, how did you get from this RSS Plus idea into closer to where you are today?

[19:59] Yeah, so – [20:00] My background is blogs were the first thing I kind of fell in love with about the Internet. And I think I still read blogs today, and I read those blogs in an RSS reader. And it's pretty amazing that this – so there's a file you have. Files is a foreign concept at this point, but it's called an OPML file. And I literally have had this same OPML file adding new blogs and removing blogs for – [20:23] As long as I've been on the internet, that's kind of crazy. It's like a single file that's actually kind of, you know, it's like having the same like wallet for 20 plus years. But what's amazing is I can take that file and I can put it into any RSS reader. And there's this like whole cottage industry of like old school nerds who still care about RSS readers. What do you use? Use Feely? I use a service called Feedbin, which is hosted. They actually have a really good like default experience. And then there's like Unread and I'm forgetting the other name, but it's like an iOS. [20:53] to email is you can actually host [20:55] your RSS service in the cloud, [20:57] And then you can actually like choose the best client if you want like a good one on Windows. I don't use Windows, but like. [21:02] a Mac client versus an iOS client. But yeah, it's also like I read Tyler Cohen's marginal revolution. I've been reading it for, for 15 plus years, daring fireball, like old school blogs where people are still actually doing that. And every time I, by the way, yeah. Yeah. So for the newer people, this is what sub stack is except, [21:22] Substack's interesting is one is it pushes stuff to email, which to me is, [21:26] I get the appeal of an email newsletter. Part of this is because RSS was a little too hard to use. And that's what I talk about in that kind of post. And,

[21:34] I think a second component is like, okay... [21:37] The goal for the writer is to just get in front of you some way and, you know, [21:42] If they can't get in front of you on social media because the algorithm is not favorable to links or, you know, a long form post on why. [21:51] you know, mid-range shooting in the NBA is under, like if that's just not going to play well in the Twitter algorithm, [21:56] then naturally what the writer wants to do is figure out whatever platform that can kind of give them that edge. [22:02] newsletters make a lot of sense, right? And it kind of like strips away. I don't have to run a WordPress blog and then hook up a different service. It's just like I open up a thing, [22:10] And then I get it to my reader's inbox, which is kind of an interesting thing if you remember the website Medium. [22:15] Medium, [22:17] So history is Ev Williams actually sold Blogger to Google first. [22:21] Made money there, then started Odeo, which was a podcasting thing. [22:25] He was pulling a thread. And while they were at the kind of [22:29] podcasting thing, which just happened to be too early. I mean, it was basically, you can think of it as like all the podcast stores that exist today would have been Odeo, right? But it just wasn't the [22:39] Jack Dorsey was an employee, and that's where Twitter came out of. So this guy has been around kind of like every major information dissemination platform. [22:47] technology that we've had. And so then after [22:50] Twitter, he started Medium, which – [22:53] At the time... [22:55] was kind of like this interesting thing because blogs kind of got killed by, and that's why I think RSS didn't do as well, is like Twitter just ate it because it was way easier to just type a tweet, 140 characters. You immediately pushed it out to everyone who else was addicted to using this app. It worked really well on mobile. You could write it on mobile.

[23:14] And so it naturally just did the same way that Instagram kind of killed Flickr and like the whole. But in many ways, early Twitter really was a lot like RSS in some of the core ways, which is that unlike Facebook or other social media platforms, it was wasn't bi-directional. It was one directional. So you were following a person. You were basically saying, I want to opt into this person's content. [23:32] And the difference from RSS, of course, is that you got it all in one feed. [23:36] that was a little more real time than, [23:38] Sort of like looking through an email inbox, I guess. Well, I think the big thing is notifications. So which, by the way, we're not. So originally every Twitter account had an RSS. You could like dot RSS and consume it. I used to get. I know they had the phone number thing or you could at least text. Yeah, you text. But the whole point is, is it was like way more configurable. Like what what, you know, Farcaster and Blue Sky and these open kind of decentralized protocols. [24:01] that we're building today are Twitter was that at the beginning. Yep. And then Twitter has increasingly gotten more kind of centralized and closed. The point I think with RSS though, is like, [24:13] Basically, everything on the web at that point had an RSS feed. You could get RSS feed from the New York Times because none of these companies were thinking about social media as a major distribution. And there weren't algorithms, right? Yeah, Facebook basically was the first one to really push an algorithm. But Twitter, for the longest time, was literally just a following feed. It's like who you followed in reverse quantological order. [24:31] which naturally... [24:32] that's how your RSS reader works. Like there's no algorithm for RSS readers. And so like, [24:36] News junkies and people who just like want to sit online all day, like would just love that because it's just like you get it's like tapping into the stream of consciousness of of, you know, collective Internet. Yeah. And so.

[24:48] Basically, Twitter wins because it's kind of a vertically integrated experience. The bidirectional feedback that if I reply to you, [24:55] you get a notification now. [24:57] there are a whole bunch of like weird quirky like ping backs and other things in blogs but it just like didn't work yeah and so [25:04] Naturally with mobile, you were able to kind of like build this dopamine flywheel loop of like, oh, I post something. [25:11] And I think it really clicked when Ashton Kutcher got, you know, raised CNN to a million followers. But the point is that a celebrity could like kind of put a thought out. [25:19] And this is like before Instagram ever became the kind of like a behemoth like that they are today. Basically, it was just like this amazing thing that if you are a famous person, you could put something out and immediately get all this feedback, which it took Elon a while to kind of figure that out. But then I think once he did, he was like, oh, my God, this is amazing. Right. And so. [25:36] I think that is the thing that RSS could just never compete with. Is it naturally appealed to like more thoughtful, introverted people? [25:44] less dopamine-seeking [25:46] behavior. And so obviously you're going to get outcompeted by the thing that is just like a dopamine casino, which is Twitter, right? When you put a banger tweet out, like, [25:55] Feels good. Like you get this like, hi, you open up the app, you get an every time it's like 20 plus notifications. Like you click it like that. [26:02] The whole thing. Even if they're kind of bot accounts, you don't even really care. It just feels like you're winning. Yeah, the talent show. Right. So that doesn't happen on blogs. If anything actually... [26:10] Blogs had comments, and most blogs had to turn off comments because they were just littered the stands. That's false? Yeah. Okay, so that's RSS. That's why RSS loses. What is RSS Plus kind of – what was the genesis of that idea and how did it sort of materialize past –

[26:27] You even in that blog post, you talked about the notion of trying to go kind of like iMessage style and say, can we actually like bolt this on top of RSS? [26:35] in the same way that iMessage gracefully kind of turns into the green text if it isn't working. That's not the route you guys eventually went. So I think it – [26:44] My... [26:46] idea was like, okay, you already have all this distribution. And by the way, all podcasts, this podcast, if you're listening to it, will be distributed via RSS. So RSS is a lot. With the one exception now, I guess, Spotify could like kind of maybe kill RSS. Is non-Spotify podcasts kind of like the last remaining thread? Well, the question is, is if your podcast is on Spotify, only hosted on Spotify, can you get it in another podcast app? No. But even Spotify is like creator platform. They'll syndicate it on RSS. Yeah, yeah. Okay. So that's kind of how the [27:15] My thinking was, okay, you could use this huge installed base that exists of, you know, old school people. All these apps would be able to make it work out of the gate. Like podcast apps. Yeah, podcast apps, it would kind of work, but you'd have all these RSS readers. I just mean to say podcast apps are the one experience maybe people today could actually relate to what RSS was like, which is you choose your client. Yes, exactly. And so the idea was just like, could you start with this protocol? [27:41] And just start adding tooling on top of it to like kind of make it more interactive. I think we explored that direction and then realized that it didn't. [27:50] Basically, you were going to be building your own protocol anyways. And so there's a certain aspect of like... [27:55] You could try to go convince a bunch of old school people who have not migrated to Twitter effectively or they're they're simultaneously using Twitter. Do you even want them? Yeah. And I think it was like, OK, let's just try to start from something at the beginning and actually just be much easier.

[28:10] If you have a completely new stack, [28:12] Because the thing that was exciting to me about starting to work on Farcaster, there's two reasons. One, I think it's really important to have the kind of like public square of the Internet not be controlled by any one individual company or country. Right. I think Elon on on the margin. [28:29] is better than the previous regime because you're at least dealing with one person and he's pretty vocal about his decisions. Yeah. But it's still – [28:37] Just as bad, in my view, is that it's whatever his whim is, is what is happening. Sometimes it's a whim. Sometimes it's really thoughtful. Sometimes it's a whim. Yes. So this is what is a monarchy, right? I think that the public square should be something that is a little bit more democratic. And if anything, is like a credibly neutral protocol, which I think we have some great examples of this in cryptocurrency, right? Like Bitcoin and Ethereum, I think, are at a level of decentralization, credible neutrality. [29:04] The Ethereum one and some of these newer chains, maybe a little bit more centralized. But the point is that I think that's how the public square of the Internet should be. [29:15] That was 2020. I think Elon actually showing up buying Twitter. [29:20] changed my perspective of how that would shake out. Because I thought basically the way this would happen is there would be kind of like a clear protocol that would get built while Twitter was kind of in this stagnation phase. Also censorship concerns. Yeah, it's just like – We're in a very different world now. It was kind of this thing over to the side that probably wasn't going to have mainstream appeal.

[29:42] And basically one protocol would build, would start to actually get some momentum, and then you'd actually have a legitimate competitor to Twitter. Right. I think Elon – [29:52] buying Twitter, [29:53] basically gave a huge tailwind to the entire space which [29:57] gave like meta for the first time in their history saying okay maybe we should go build a short text [30:03] platform with threads, which say what you want about it, but it's got 275 million people using it every month and they're adding a million people a day. [30:10] I think Blue Sky has gotten a lot of growth as a result of just like people who don't. [30:14] don't like Elon. And so now... [30:17] I think you've seen a bit of a political fragmentation of if you think of like the big networks are. [30:22] Twitter, Twitter. [30:23] And threads are like the two biggest networks. I think threads is Instagram, Normie, left-leaning. And then Blue Sky is more left of threads. Yep. [30:33] And then X is now right. This guy is sort of like, we're really mad that Elon runs Twitter and we hate Trump. And so we are leaving Twitter. Yeah. And I think that they have some international growth. Like, I mean, they're doing a really good job. Like they're growing. Also notably founded by Jack Dorsey while he was at Twitter. Yeah. It's like funded by. Yeah. So this is kind of mess. And he's now not involved. Yeah. And then I think you have X now is right of center. [30:56] whether it's the user base or just the algorithm is like very favorable to Elon, very favorable to Trump. [31:02] which was not the case, you know, four years ago. And then I think... [31:07] Where you have some other subscale, like Farcaster is one of those. It's like, Farcaster, we're not explicitly political.

[31:13] Like we have plenty of left-leaning crypto enthusiasts and right-leaning. I think the average person would say if you're crypto, you're right-leaning. But, you know, I interact in my network every day. We have plenty of different political opinions there. And then I think Noster is another example. That's like more right-wing because it's Bitcoin. And then you have like Parler and Gab and Truth Social, which I think are like far right. So I do think there's actually quite a – [31:37] political fragmentation. And so I'm actually... [31:40] My updated opinion is I just don't think there's going to be one protocol, and it's going to be this kind of archipelago of different – [31:49] Pick your flavor. Some will be decentralized, some will not be. [31:54] I forgot Mastodon in there. Mastodon kind of gets lumped in with both Blue Sky and the political side, but also it's technically part of the Fediverse, which is where... [32:01] where threads is. So my view is we're just going to move to an era of like you, you pick your newspaper base, which I don't actually think is the worst thing, right? Because, because here's the thing in a world where some of those networks are, [32:14] default open from an API standpoint, right? So Fediverse, Blue Sky is actually built on a protocol called AT Proto, Farcaster. You could go build a client that sucks in information from all of them. And so I actually do think we're getting closer to this world of like RSS+, where because the APIs are now going to be open on a lot of these networks, [32:32] you're at least going to be able to, as a developer, build a client that could actually have a specific algorithm for at least some percentage of the content. I think increasingly Twitter...

[32:44] And the thing about Wims is he could change his opinion on this again, but like – [32:47] Twitter's API is as locked down as it's ever been in the history of the company. [32:52] I have a lower belief that Twitter is going to change, but Substack is another good example. Substack has its own like weird, [32:58] Twitter version for its writers that actually has a lot of users, like from what I've heard. And if you go on there, it is like robust discussion. [33:05] They're built on email, which is a pretty open protocol, and you can get an RSS feed, so [33:10] Does subsack at some point federate into one of these protocols? Yes, great. More competition. Yet another – [33:17] part of the internet that is now kind of in this like open version that is remixable and open to different developers, which the moment that that is the case, [33:26] And it's a little different now also since I started in 2020 because of AI. Because I actually increasingly just don't believe, I don't think people are going to be using pane of glass scroll of feed. [33:36] Um, [33:37] in a world where I think AIs can get really smart about what you want. How they feed you information. And they're going to ask you kind of like, what are you feeling? It's like, you know, and then basically they're going to be able to provide it on the fly. Sort of like the Google Reddit search. Like everyone sort of realized you should append Reddit to a Google search and now perplexity just does it for you. So I think... [33:56] I'm optimistic that we're basically going to be moving to an era simultaneously of like the internet getting locked down because people, if they have valuable data, they don't want AIs to just like train on it. And so, you know, Reddit has a famous deal now with. [34:09] I forget which, I think it's Google and OpenAI. [34:12] pay them to get access to the data. Yeah. 60 million for the first one. Right. So, so like, you know, and if you're anthropic now, like Reddit's off limits to you, which is not how the open web worked before. But my sense is that these open, uh,

[34:26] social networks are going to actually have to confront this. You know, I think there's actually been some, [34:31] I saw a story where Blue Sky users... [34:34] We're kind of demanding to understand how the blue sky data. Remember, blue sky is like an actual centralized thing built on top of the protocol. And as they said, they're not going to train anything from blue sky data. But then people were like, wait a second, if everything is available over the API, like people can just take it, just take it. And so. [34:52] I think we're going to kind of like people are going to have to kind of figure all this stuff out. Yeah. But I am optimistic in aggregate, and I'm still very bullish on what we're doing with Farcaster. But I think the where we're going to be over the next five to 10 years is like a lot of the Internet is going to be kind of lit back up in terms of your ability to, as a developer, build your own algorithm for information sources. And I actually think you'll probably see some equivalent, which I've been wanting this for a very long time. [35:16] of kind of like a, [35:18] Just like organic food. And so this idea of it's like slow food movement, like get back to the basics. And so you could imagine RSS. [35:25] these kind of like open social networks that you can actually pull this feed in. And then you have an algorithm that's like, don't show me anything that's going to be like rage inducing. Only show me interesting things that are academic or interesting. And so I think that that will exist and I think it'll probably be AI driven. We're making progress. Yeah. Okay. So maybe feel free to correct me if I'm wrong. As I understand Farcaster, there are kind of like three core ideas. One of them, as we just talked about, information should flow in protocols. That's starting to happen both due to what you guys are doing and more broadly.

[35:55] you've talked a ton about, but I think is really critical and resonant. It's just this notion that like distribution is almost always underrated when it context of building a new thing. And one of the ways I've thought about this is people sort of assume that composability will drive network effects. And in reality, it's kind of the other way around network effects are what make composability useful. You've said this is sort of like the field of dreams idea or the Q DAO versus the tech stack. Do you want to talk in people aren't familiar field of dreams, the baseball movie, do you want to talk specifically about like, [36:25] Was this something you kind of always knew going in? Clearly it started to work with Farcaster. How do you think about this is also the sort of product over protocol idea? Like what was inside of this and when did it become obvious that this is what you needed to focus on rather than just building the most robust open tool? Yeah. So the idea is taken from Teal's zero to one, you know, feel the dreams. I actually think he cites in either the class notes from like masters or in the book. But the idea is like if you build it, they will come. That's a fallacy. Like not going to happen. [36:54] And all companies, a lot of people in crypto are a lot of developers. Generally, they think it's the brilliant idea. And the idea itself is the thing. And crypto is even worse because you have these extremely smart technical infrastructure people. [37:06] that see the success of, of a blockchain like Bitcoin or Ethereum. And then you go, I'm smarter than that group. And if I build the right one, if I build the best design, it will win, which we, we know it doesn't actually work. And, [37:17] And so EGRSS, arguably. Right, right. Yeah. The best tech does not necessarily win. If you have any, you know, any student of history, you know,

[37:26] We'll see that JavaScript is the most popular programming language in the world and was like a sleep deprived programmer in 1994 who put it together in like some crazy short period of time. And so the network effects of these things last a long time. And I think the. [37:42] the kind of like way to think about, [37:45] the field of dreams thing is, [37:47] It's it's sales is what actually like you. The tech part is necessary, but not sufficient. And the sufficient part is actually how do you convince people? How do they become aware of what you have, regardless of what you're trying to sell, whether it's enterprise software or consumer app? [38:01] and then get it in front of them get them to use it and then keep them around right and [38:06] So that to me is sales is something that you can do individually, right? Like I can show, come up and try to sell you a car. I could try to sell you enterprise software. [38:15] Harder to sell you a social network? I think you did personally. Yeah, I mean, in lieu of having a really, you know, like lightning in a bottle idea, basically you have to do hand-to-hand sales. [38:24] But generally, the way to do mass sales is marketing, right? People think marketing is like advertising. Yeah. But I think a lot of that is just like, [38:33] Clear value prop, like how do you get the word out, whether it's word of mouth or a referral program or whatever. [38:39] You know, PayPal had this famous you sign up, I think it was five dollars, but it's like they gave you five dollars and the person who signed up got five dollars. [38:47] Generally, it's a pretty good way to get people to sign up if you give them free money. And so the growth actually goes fast. Uber famously just like crazy discounted rides and all that kind of stuff. And so I think that is the...

[38:59] the go-to-market and kind of like sales component of any great company, [39:04] actually typically is just like not [39:06] well covered or well understood because it's much more sexy from a media standpoint to talk about the brilliant it's like this edison complex or the nikola tesla it's like brilliant engineer in a room comes up with the idea and then it just sells itself which [39:22] Go to market is a lot more messy, right? And there are there probably some shenanigans that happen one way or another for a lot of these companies. And so I think like Google is a good example of where it is this company that they just like built this better search engine and then it just went viral and or Facebook, another one where, oh, just is this idea that took over the world where the reality with Facebook is. [39:42] Yes, there was some good, like, you know, kind of went viral at Harvard and they were smart to not have it expand too quickly. But very quickly at Facebook, they developed. [39:51] on a very aggressive growth team because engineers don't like being called marketers. So they can tell the term growth. But, but like, how do you, how do you kind of like get around, you know, a school that, that might not have penetration, like get the friends at other schools. And there's a very famous story at, [40:06] at Facebook, you know, it's apocryphal, but basically they slurped down the entire AOL instant messenger graph after a partnership and they shut the partnership about four days later. Oh, my God. And AOL didn't even realize at the time, didn't realize it. There's this kind of insinuation that that Chamath then, like, realized that he was AOL and then went. But but or, you know, he but the point is, is that. [40:27] Basically, people just think of like, oh, social network, it like works and then it just grows viral.

[40:31] I don't actually think that that's the case, and especially – [40:35] If you're building a developer platform, which is a second order. So in order to get people to actually have interest in your social networking protocol, you need users because the developer doesn't want to come and build on top of your tech stack. Get the users, which is the hard part. Right. This is the field of dreams. You can't you can't expect people to do that. [40:55] I think Blue Sky is actually a very good example of this. Like the people in the Macedon activity pub, [41:01] They're a competitor, right? Like they're both these like federated networks and they've all written critiques. It's like blue sky is actually not decentralized. Like blue sky is literally referring to the app, not the AT proto. And, [41:12] I have every confidence that that team will actually execute on getting to a good level of decentralization. I think that they're actually pretty principled people over there. But – [41:20] The whole thing is like people are using that app. And because there's heat there, there are users, people are building on top of it, right? And so I – [41:29] The... [41:30] Most important thing for a social or consumer developer platform is like, do you have people using it? Windows is useful because there is a Windows desktop, you know, in the 90s on every worker's platform. [41:43] Place, you know, desk. And if to the degree that they had to do like Microsoft Office is the like primary reason you buy it and that's vertically integrated. Great. That's what they did. Yeah. And then then you get Adobe. Then you do get the gaming and all these other things, web browsers, because they've actually done the hard work of getting the installed base. The other analogy I always use is like the iPhone.

[42:04] The iPhone didn't have third-party apps for the first year. [42:06] And so Apple did the hard work of marketing marketing. [42:09] and selling 5 to 10 million iPhones with all first-party apps, including Google Maps and YouTube. So they partnered for the content, but they were the ones that built the apps. [42:19] Yeah, it would be pretty easy if you knew less to go back and credit the early app store as driving iPhone off. No, the whole thing got started because they were able to, you know. And the... [42:30] path dependency of the iPhone as a result of [42:34] the company being in a way healthier place because of the iPod and like building the supply chains and understanding the, the Apple stores, like, [42:42] You can't just like magically go from like brilliant idea of the iPhone. [42:47] And actually, if you go look at the history, they thought they were going to do an iPad. So they were like, how can we take an iPod and mix it with a Mac and... [42:56] And they were going to do a bigger one and it wasn't. And then they were able to like actually make it small. [43:00] And and what's crazy is how fast they were able to achieve it. But so all that said is, I think is very, very strong point of view now. And this applies to crypto because of all these very technical people who think that they're like their technical brilliance is the reason people are going to use their stuff. [43:16] is, no, it's like [43:17] Tech is necessary, but not sufficient. The sufficient part is actually how good are you at sales? And I actually just don't believe that most of the time you can outsource that. [43:26] It's like founder-led. Like that's what you need to do. Could you argue that Uniswap... [43:32] Is like an example, maybe like a counterexample? Yeah, but you could argue Uniswap.org. It's also an extreme example of utility. Right, but does Uniswap work if Uniswap.org doesn't actually have good design and like works nice? Yeah, it's a good question. Which I think is like it's hard to find that counterfactual. It's a good question. Right, if it was just a bunch of API docs and like a white paper of like, look how brilliant this like automated market maker is.

[43:56] I actually think you get your lunch eaten by someone else who builds the thing that... [44:00] You know, it just allows people to swap whatever token of the way, right? [44:05] Hey, Jackson chiming in here. Just wanted to let you know this is the beginning of the section where we start to go much more deep on crypto, status within it, and Farcaster specifically. [44:16] Again, I think it will be enjoyable for anyone who's curious, but if you're not as interested, you can skip ahead in the show notes about 45 minutes. [44:25] Well, this gets into my read of if the first was information flow and protocols, the second is this network effect idea. Maybe the third kind of core idea inside of Farcaster is the notion that ultimately we're kind of playing status games, and that's what people seek. This gets maybe into... [44:42] Even if Uniswap could have worked on the utility basis, something else is happening on social platforms. I know you, I think probably both of us are quite influenced by Eugene's... [44:51] status as a service. And interestingly, in rereading it, I was reminded like how heavily he compares social platforms and cryptocurrencies, and that they basically both have this sort of emergent form of capital, a proof of work around it, and thus increasing scarcity. On some level, like you kind of have to like believe in this new thing, whether it be a [45:09] money or a place to spend time, and they both critically reward outsiders. This is something that gets very much into what you were saying around marketing and growth and so on. I'm curious for you, [45:22] 1. [45:23] You, I think, have done a good job of rewarding outsiders in the sense that you brought in new, you talk about this all the time, you brought in net new people to Farcaster. But two, very specifically, what does it look like in your mind to build the premier arena for status players?

[45:37] for crypto people. [45:38] Because that seems to be maybe maybe I have that wrong, but that seems to be where you've landed today. [45:43] Yeah, so we very much have tacked [45:45] to being a [45:47] Social network focused on crypto. I don't believe in a post-Elon approach. [45:52] kind of black swan event for short text-based social networks, that there was going to be one. And I think our ability to win on a political vector is... And that's changed since you started, Park. Yeah, yeah, yeah. So in 2020, it was compete against Twitter. Twitter is stagnant. And 2024, it's... [46:07] we are not going to win on a political vector. And so in that world, I think... [46:14] And that's what's driving growth outside of Twitter. It's because of politics. And so if you can't benefit from that tailwind, [46:22] You have to find a different wind. And I think our wind, which... [46:27] seems to be picking up now, going back to the beginning of the conversation, is if we're tied to crypto and we live and die by how crypto sentiment is, and it's an area I've been in for 10 years, so I'm pretty comfortable in weathering the kind of ups and downs there, that is, I think, the bet to make for Farcaster. [46:43] And so it actually informs a lot of our product decisions is to make sure [46:48] We are leaning into what makes us unique and differentiated relative to all these other networks. And it's the fact that every user has a crypto wallet under the hood, even if that's kind of abstracted away from them. [46:57] And the programmable aspects of the network work and pair really nicely with crypto. And so that is like the updated... [47:05] you know, strategy for Farcaster. And on some meta point, you have at least started to be a place where crypto native status can be earned.

[47:16] Yes, although I would say that it's still an extremely hard network effect to fight. And to the point of the essay – [47:25] We still have to compete against crypto Twitter. It's like CT. Like that is the status arena for crypto, which wasn't always the case. So I started at Quendez in 2014. Crypto social media was Reddit. [47:38] Our Bitcoin, our Ethereum, like Vitalik was posting, you know, but but at some point kind of around 2017 and certainly into the next kind of bull market. [47:49] things shifted over to Twitter, which if I was to kind of [47:53] My best bet is increasingly people are on smartphones, more and more time on smartphones. Reddit is just not a smartphone. It like barely works on a phone. Whereas like if I think of the amount of time I spend on Reddit, [48:07] Well, the power user use case, especially since they killed Apollo, I think the Reddit home, the home view, ironically, feels a little bit more like TikTok and Twitter. But for the way people like you and I use Reddit, I would I would I would totally agree. Right. Yeah. [48:20] So I think that, [48:23] That was the opportunity for Twitter. It wasn't like I don't and maybe maybe this has been reported, but my sense is Twitter didn't explicitly target these people. It just was like more and more people very much in actually like post Trump. Right. He had been using Twitter in the 2016 election. So like kind of the world started to get like more on Twitter. Is it possible that crypto also, at least in 2017, started to enter a phase that was more about evangelism than it was about kind of the insular community?

[48:53] And then you started to have crypto native funds and stuff like this is the money started to get a lot bigger. So naturally, more people working in it professionally, at least. So that is where the status hierarchy in crypto is. It's very much on on Twitter. [49:08] I mean, look, Elon in the last bull run is talking about Doge and Ethereum and Bitcoin. Yeah. So – [49:14] That is who we're competing against. I do think it is a competitor that we can chip away at, though, because – [49:22] One, it's small enough within Twitter's overall kind of... [49:26] portfolio of users, right? That they're not like, [49:29] specifically focused on crypto people. And if anything, they actually don't like it. It's like that's where all the scams come from. A lot of the spam have crypto. And so. And they've just picked more of a side in general. Twitter, unlike YouTube or even Reddit to some degree, Twitter is way more opinionated on a whole bunch of things today. This goes back to why all these competitors are good as two. Well, it's funny. The Stagnant, the only area that I was impressed with, there were two in the previous regime. One, they were able to clone Clubhouse spaces. [49:59] from a like pure, just like, okay, you could get it with a much bigger graph and distribution. And then the second one was they actually had a crypto team. And so you could add a crypto NFT profile, like a little hexagon. [50:12] And we had added that at Farcaster, like, is one of our early features. And we thought it was going to be, like, kind of this differentiated thing. And then they go to add this. And I just remember being like, oh, man, what are we going to add that, like, could actually be differentiated? And then Elon shows up and just, like, totally veers them away from crypto. It's just not a focus. That whole team got let go or disappeared. And so –

[50:32] I think we're still very much fighting... [50:34] crypto Twitter for the status game. And where we've been unable to move is anyone who has a big audience, which is naturally makes sense. It's like you, you've won a crypto status game on crypto Twitter. You've got 200, 300,000. Musical thing. Yeah. Yeah. Why am I going to move over? Yeah. [50:49] And I actually think that one thing, one riff on Eugene's essay is, [50:53] that I like to think about from kind of like a history perspective is just think about the people who moved to, [50:58] The colonial United States. There was no one in the elite. Yes, we might have towns and states and cities named after the elite. You had to be kind of desperate. Yeah, you basically it's like Puritans were this weird, you know, marginalized religious group. You had a bunch of basically... [51:16] thrill seekers, adventurers, who are the people moving to California in 1849? The frontier is always going to be populated by people who don't have power and status. [51:26] in the existing world. [51:27] structure right and and so [51:31] What I think you've [51:32] get is like, to the Eugene point, like you do start to mint your own aristocracy, your own status hierarchy. And so I think where we are finally starting to click is, [51:44] Because we've been around for four years and we have users who have been around for four years and have accumulated larger followings and have have stuck around in this new thing because they have an incentive to kind of make it work. And part of it is we also seeded the network like a lot of our early users stuck around were developers. Right. [52:01] So now we're getting kind of this like you build it. The developers won't come. But if you go actively sell the developers, maybe they will come, which I think. Right. And or they become your users and then they're hanging around and they're basically made friends. So this is just like a big group chat for them at this point.

[52:14] But getting, you know, four years later, now we're kind of entering a bull market, expansionary market for crypto. [52:21] People are starting to play around with things. [52:23] because everything is programmable. And now naturally that's pulling people back into Forecaster, because I think that the quickest way to develop status in crypto Twitter is, [52:35] is did you deploy something on chain that makes money? And so the money game is actually relative to most social networks. Which is pretty different than others. Which, again, goes back to the idea that our network is – [52:47] differentiated in that. It's a very fine line though. [52:50] because there's a network, BitClout, that had a kind of brief moment in the sun, and I think it's very... [52:57] tough to do explicitly financialized status games. Because if you do, basically anyone who [53:03] wants to actually be really high status, money is an indirect component of status. And what you want to be – Can't make it explicit. Yeah. It's just like even a successful rich business person wants to be known for their business accomplishments, not for the money. Right. Right. [53:19] and be known for just money. We don't talk about the money. Yeah, it's very Puritan. But the point is, is that this is like a very Silicon Valley thing. It's like you have all these are really rich people, but they're wearing sweatshirts or whatever. And the whole point is they want to talk about ideas and the things that they're working on. [53:32] And yes, they happen to be rich. Whereas like you think of like the classic guy in like, I don't know, Vegas or Miami is like – [53:38] bling bling, like I've got a lot of money. Yeah. [53:41] But everyone who's high status thinks that's like money is a second, secondary thing. And that's not a judgment, by the way. I think it's just, you have to have that honest assessment of like, when you're building a social network, you have to deeply understand like how, how do the status hierarchies here work? And,

[53:58] If you tack too far in one direction. You're going to exclude a whole bunch of people who, yes. Exactly. And look, right now on Farcaster, there's been a tack. [54:06] to the kind of like what people are doing on chain which is trading meme coins and there's like this whole kind of like [54:11] meta that's happening that are like Farcaster native meme coins, which are arguably in a lot of cases, there's like some more sophistication. Like there's a class of meme coins on Farcaster that like power applications that you can use. Whereas like, I think like a pure meme coin tends to be pretty nihilistic. It's just like, will it go up or down? That's it. Pushing this [54:30] Base level activity, slightly higher status. But then there's a whole bunch of high status people even within crypto who go, I'll forecast there's too many meme coins. And so – [54:39] I think... [54:40] You can get mad at one side or the other, or if you're in the business of like trying to build the social network, you just have to. How can I please both? And like, how can I make sure that the people who aren't interested in that when they open up the app, they're seeing different types of content algorithmically than the people over here? Because the easiest way for people to stop using their app is they open up the app. Nothing they see interests them. They scroll a little bit and they do that three or four times. They're just not going to come back. Yeah. We talked a lot about, obviously, the underlying infrastructure and design. We talked about the tribal component of it. [55:09] Do you want to talk a little bit about, at least today, what Farcaster actually is? You've described it in a handful of ways, at least recently, that I find interesting. One is just like broadly it's an economy. I think you said it couldn't be the single best way to interact with anything in crypto. It's social plus crypto wallets plus programmability. Obviously, a lot of these things are inside of this. Are there any of those? If you're meeting somebody who had some familiarity with crypto, is it just it's a better version of crypto Twitter? What is it?

[55:36] Yeah, I think that the... [55:38] The layman's [55:39] pitch, which many people would just say, hey, I'm not interested, but here it goes. It's just it's a social network for people interested in crypto. And if I give you that pitch and you don't think you're interested in crypto, then you probably. So what if I what if I'm interested in crypto and I use Twitter? [55:53] Right. So then the question is, I already use Twitter. I follow some people from a crypto standpoint. [55:58] Yeah. Why would I care? [56:00] And I think that for those people, it's this is where you're going to find the alpha. And now whether that's the case, that's that's more work to be done. [56:08] But I think more recently, even just like in the last couple weeks, there's been more alpha on on Farcaster than ever before. And naturally, user numbers go up. Right. And the fundamentally that the. [56:21] Behind every good consumer product, there's a seven deadly sin. That's like a famous Sequoia framework. [56:25] uh, [56:26] Farcaster is greed and not everyone on Farcaster is greed. And I don't think that that's like the ethos of the network. But if you were to say like, why are, what is driving growth? [56:35] It is people... [56:36] people think they can make money. In a little, in a sense, it's almost the combination of the two things Eugene talks about as status. It's the combination of the crypto status, like the actual money alpha and the cultural status combined, like layered into one thing. That's pretty interesting. Well, it's funny you say that because I did a bunch of user interviews about six weeks ago, and I was just really trying to figure out, this is before any of this kind of current meta, like what's the direction to kind of like really bet the company on? Because we had had a lot of growth at the beginning of the year and that had really leveled off.

[57:06] Thank you. [57:07] And in talking to people who came in in early 2024, a lot of them had come in when we had launched Frames and then there was a new asset on the – What's DJ? People not familiar, Frames are like mini-apps on Bargaster? Just think of it as like if you're older, like Facebook had the app platform, Zenga, Heyday, Farmville, that kind of stuff. And if you're a little younger, Telegram and Discord now have these things called mini-apps. So it's – Any tweet, any cast can be basically a little application. Yeah, launch it and be able to be interactive. [57:34] And so at the same time as frames, there was an asset that popped up on the platform. You can call it a meme coin, but it's somewhat more sophisticated called Degen, which, [57:43] And what was interesting about it is anyone on Farcaster was eligible to claim [57:47] this asset that it's always funny talking to people who might not like think about crypto, but it's like, [57:54] Just like the government can print money, like cryptocurrencies can kind of just create more supply. I asked somebody to ask me, how many coins are there going to be? Yeah. And so basically the idea was Degen had a set budget and they said anyone in Farcaster can go claim their Degen. Based on Farcaster use and engagement. Yes. And then you could tip it to other people. So there was actually, you know, natural utility from day one rather than just like purely a coin to speculate on. Right. [58:16] And so if you talk a bunch of people, they came into VARcaster as a result of that because DGEN, [58:21] was a potential thing to make money. They heard that you can make money through frames. [58:24] And then a lot of people left after that kind of big growth spurt. But then those people who stayed, [58:31] They're like, oh, I made friends. Or a classic story in crypto. Right. And so – [58:35] So one frame that I think – you can't use that word. I also try to use the word paradigm, but it's not – I don't have it. So one way to think about it is –

[58:47] the kind of inclination to find alpha in crypto. [58:51] right, is very high. Like that's why people sign up for Telegram or Discord or pick your flavor of whatever social media. Hunting. [58:59] Goldrush, let's go. That is what this... [59:02] category, [59:04] From a just like pure audience standpoint. Yes, there's a bunch of people building. They're trying to make money too, by the way. Is speculation just the oxygen in the air if you're going to build a crypto product? Can you possibly build a consumer crypto product without speculation being? Well, I spent four years trying to do that as much as I could. And I would say is one – [59:21] The incorrect bet there was if you were going to go do that starting in 2020, you'd be able to do that. [59:25] You needed to build a left of center social network knowing Elon would come in. [59:30] because then you would have benefited from all the refugees of that. A lot of people have asked, why haven't you gone more anti-Elon in response to this? I just, I'm not going to be that kind of founder, right? I have immense respect for Elon. Like, again, antics to the side, because I think I can separate art from artists. Yes. My wife worked at SpaceX for a bit, so I got to see a little bit of the inside of, like, how Starlink, you know, rolled out. And now, having been a founder for four years, like, [59:55] the difficulty of doing one company, like having that even just succeed to have that work [1:00:01] In the physical world, both Tesla and SpaceX, like basically. You actually mean literally challenging. I just meant to the sense of counterpositioning. My point is, is that I thrive. And I think you know me well enough at this point is like I have to be authentic in my positioning. And there's no version of the world where I can counterposition against Elon.

[1:00:21] That is like. [1:00:23] He's a... [1:00:24] He's bad. I think he's, [1:00:27] complex and do I endorse every opinion like no of course not but my point is is that [1:00:32] I think Elon has improved Twitter on the margin. I think there have been some decisions I wouldn't have made. Where has he improved it most? Censorship? [1:00:41] I think that, [1:00:42] People don't appreciate how... [1:00:44] like suppressed and censored Twitter was before. Got it. And, [1:00:49] What I think people, you know, and I'm sure I'll get feedback for this section of the podcast from people on Farcast or whatever, but they'll basically say they'll be like, oh, he just let Nazis on the platform. [1:00:59] And it's just like, no, like that is... [1:01:03] such a small segment of whatever group of people that are on Twitter. And I think what is just more is the entire Overton window of Silicon Valley is [1:01:13] has shifted as a result of Elon coming into Twitter, basically allowing people to kind of like speak. So now that is not so like group mob left mentality. You I think maybe you wouldn't say this. I would at least my anecdotal experience would be that it started to swing in a pretty good way. And now at this point, it's basically like a right wing. I think the algorithm is awful. It's literally every time I open it up, it's a tweet from Elon. I'm like, OK, like he also was like, I'm going to swing the election. [1:01:43] publisher prior to like the modern era to do right like but my point is that [1:01:49] It got Silicon Valley out of this crazy...

[1:01:52] like woke lockdown. I mean, Brian Armstrong with his blog posts, and I think there's like a kind of overall trend this way. Yeah. But I mean, I think it probably had an election impact more, more than hard to measure exactly. And I do think I also think firing 70, 80% of the workforce, like that was a like kind of a signed, it allows the other CEOs to, [1:02:14] Silicon Valley to actually, you know, we're talking about founder mode and all this. Like the reality is that Elon's actions there allowed a lot more leeway for people to express the full range of opinions. And and so I think that that is a net good. [1:02:30] in our society. [1:02:32] And I think if anything, it's also good that there are alternatives. Like if you don't want to be on Twitter, there are at scale alternatives. Maybe to your credit too, like if I were building a network, I probably wouldn't want my core user identity to be like, we hate Elon. Yeah, yeah. That's not a very strong. I think founding something on an anti-sense. [1:02:50] mentality is a losing strategy. It might be, it might be a good way to organize people. I was listening to Lulu Meservi talk about this recently, and she, she kind of made this point, but it's, it's, it's exactly that, like it actually being the founding story, it might be a good way to like find some people to go start chatting with. Yes. But if that's the core thing. Yeah. And I think those are the types of people who are going to complain about everything. They're going to be, and so my, I have like much more, and this is probably where I'm more aligned spiritually with

[1:03:20] like EAC is kind of like this cringy meme, like effective accelerationism. But like my whole point is like, [1:03:27] I want to see civilizational progress, right? Like I want to see us build 100 nuclear reactors. Counter positioning against Elon with a Twitter clone would be probably unlikely to. Right, because that group of people is happy. The guy who is catching rockets with chopsticks and electric cars. High overlap with crypto. Right. And so I think it's just not going to be authentic, right? And so there's no version of like a strong counter position. I can counter position of saying like programmable, like Twitter APIs, crap. And like it's totally locked to town. [1:03:55] like the two couple examples we've seen the last two weeks, it feels obviously frames is interesting. You've launched frames V2, which we can get into, but particularly the last couple of weeks with some of these AI meme coin things, we've seen some like pretty novel stuff, even if you don't like the speculation part of it. Yeah. And, and I, I view it as like a, it's, it's kind of in a line of progression towards like, who knows where this is going, but yeah, [1:04:15] The point is that with Farcaster, you can spin up from the command line, so something that's very easy for a developer, an account, and you can just start... [1:04:24] posting to the network and and like the apis are completely open there's no rate limits like it's crazy like twitter charges like to [1:04:31] have like a barely functional app. [1:04:33] on Twitter now, and you couldn't have other users use it just to like run a bot. It's like $500 a month. Like an analytics tool. [1:04:40] And they limit the amount of calls. It's crazy to think how close the API for Twitter is now. Their premium plan is – Elon's also very afraid of – he's obsessed with the human symbol thing, which maybe is part of this. Yeah, and he also doesn't want to leak the data to open AI because they had scraped everything. And so it's $42,000 a month for even anything that's come close to Farcasters. If you spin up a hub on Farcaster, you can get that for free. Wow.

[1:05:05] So, [1:05:06] I think that that is a counter positioning and that's authentic. And so I do that all day long. It's like Farcaster is programmable social. And so what we've actually seen is like there's been this AI – [1:05:15] Agent. [1:05:16] On Twitter, Truth Terminal, which was funnily enough, $50,000 worth of Bitcoin sent by market tracing to some random anon. But that is actually started to play out on Farcaster because people saw that idea. They're like, cool, I can do this on Farcaster. And it's actually easier to kind of manage. And then one of the AIs on Farcaster started deploying coins on Farcaster. [1:05:38] an Ethereum L2 base. And then the two AIs talk to each other. Someone like basically kind of nudges them. And then the two bots started going back and forth. And so the one that can create the coins created one for the AI. That is a little bit more just kind of like an art project philosophy oriented. And, um, [1:05:56] So I just saw this today. That one's called Ethernet. Ethernet now owns 1%... [1:06:03] of a meme coin that now has like a $50 million market cap or $40 million market cap. This is higher? Yeah. [1:06:08] It's related to that. But the point is that like [1:06:12] That AI... [1:06:14] basically has now funded itself for like a, like as a, as like a seed round just by creating like meme coins. So it's like, [1:06:21] if you look at that, you kind of go, okay, that's not sustainable or that's not scalable. But if you squint and kind of look back, it's like, where could you see this going? It's, [1:06:28] It's like you have these like proto... [1:06:30] AI agents that are actually interacting on an economic basis. And this is where actually I think people who totally don't buy into the crypto stuff like this is a truly novel Internet money thing. Right. There's no way. Yeah. But like so for example Stripe this week or two weeks ago they released something called like their agent kit. And he's like you can tokenize credit cards and you can use it for shopping and perplexity release this. Okay.

[1:06:53] It's far more likely to impact you day to day right now as you can go to perplexity and you could say, like, get me airline tickets or buy this like shirt. [1:07:00] Spend my money for me. Like, go do this and whatever. [1:07:02] But like, [1:07:03] you can't actually give like an AI agent, like a bank account. Like the way the world works today is you have to KYC. It has to be like this, this AI agent, [1:07:15] if that private key is online on a server that the agent has access to make a call on truly like literally like has a bank account and could control and like could. And so if you kind of extrapolate AI progression from there, then you're, you're going to have like, [1:07:29] weird stuff start to happen and and to you know chris dixon's like what you know what people are tinkering with on the weekends like this is what everyone's going to be doing in 10 years like [1:07:38] I kind of look at this and go, like, this is a total playground for mixing, like, what is the leading edge of... [1:07:45] A much better playground than something like Twitter because it is native to code. It's natively programmable. Yes, and every user has an Ethereum address. And so it's like the whole network is designed like you can't actually have. [1:07:56] an AI agent, [1:07:58] on Farcaster without an Ethereum address. Like, it's just like the way the architecture works. So even if you didn't want the payment, someone can actually send money to your AI. So it's like you got to put programming into like... So I think you're just going to get weird emergent things to come out of it. And if I can predict... [1:08:12] what to go build that would be this, [1:08:15] thing that actually was important, I would go do that. But part of the excitement of working on something like this is if you give people the actual Legos,

[1:08:23] that they can assemble, they're going to surprise you. I mean, that's the actual field of dreams bit. [1:08:29] Yes, and we're starting to get there. We're like, you know, this whole recent meta and like inbound interest as a result of the stuff that people on Farcaster are doing on chain. [1:08:38] is a result of these AI agent [1:08:40] You know, hackers who... It's a positive flywheel. They didn't raise venture money for this. They literally, like, kind of hacked it together in a night or a weekend. They're here because they found Farcast to be a place, as developers or interesting technical people, that was interesting to spend time. [1:08:52] They started experimenting with things. That emergently produced an alpha that people elsewhere cared about, and now you have a flywheel. This is the first time, I mean, other than maybe the frames-degen combination, like, that's the first time we've actually had this, rather than us, like, grinding and trying to get to this. [1:09:07] And AnonCast is another one where someone took one of these coins and they basically said, in order to use this app. I have AnonCasted for those wondering. Okay, there you go. But this is pretty cool. It's like they basically didn't even create their own coin. They took a coin that existed. [1:09:20] And they said, in order to use this app, you need... [1:09:22] 10,000 of this coin. And then they use ZK proofs, zero knowledge, which kind of geekabrain math that allows you to like prove that it's you have the balance without revealing who you are. And like smart math people would tell you that it works. It's basically like Yik Yak, but you have to own the coin to be able to post it. Yeah, basically. And so they've already like iterated like multiple times on it. They have a Twitter account that actually is bigger than their Farcaster account. So. [1:09:45] you can kind of view it as the Twitter account grows. [1:09:49] To stifle for the alpha. Well, it's even more is you can actually now, even if you don't want to use Farcaster, you might go buy this coin.

[1:09:57] Just to use the distribution on Twitter. Because it allows you to post to either Farcaster or Twitter. Yeah, and so now, well, you actually still have to post to Farcaster, but then you get promoted to Twitter. But the point is, so Vitalik. [1:10:08] participated vitalik bought this coin people freaked out i think it was the first quote unquote meme coin vitalik's ever purchased obviously i think you and i would say this is something different than just a meme coin well my point is i could care less about the meme coin as much as i i think what's interesting is he's using this app and so now at the very least he could just to just to make it super explicit for people but the sort of implicit premise here is that any anon cast could be vitalik that is the most interesting thing and so now imagine if you add so they talk about the anonymity set like who who who are the holders enough to to have the [1:10:38] mouths to be able to close. And so [1:10:41] What happens if you got like 10 more really interesting crypto high status, like, you know, people have got several hundred thousand followers on crypto Twitter. Right. And you know that they're in the set. [1:10:51] Or even made a new account. New small groups. And so now what you're starting to do is get to a potential where you can, let's say you created a new version of this and you only allowed 20 people who have at least, [1:11:03] you know, 500,000 followers on crypto Twitter, arguably the most influential people on crypto Twitter. And they all have the coin. And you know that the math works, that no one can actually get doxed by even a central server. [1:11:14] for posting to an account, everyone would follow that account because now it's you're getting a new type of speech. The water cooler. Yeah, it's actually the Federalist Papers, which is a very, very eyebrow, like, you know, like important to getting the Constitution through. And it's James Madison, John Jay and Alexander Hamilton under one student in Publius.

[1:11:31] And everyone in New York is reading this to be like, okay, why should we pass the Constitution? Slight tangent, but I'm quite interested. Pseudonymity is something I've been interested in on the Internet for a long time. It's one of the kind of the first areas of crypto I got excited about. [1:11:43] You guys explored a little bit with NFTs early in the Farcaster era. You even had, I think maybe in the RSS Plus thing, Why Stop at Blue Checks, many different badges for verification. You haven't really pursued that, but does something like a non-cast or even some of these other versions, does it start to open up a world where maybe I could form credible reputation and identity on Farcaster that isn't I'm Jackson Doll? Yeah, so that's starting to happen emergently. So there's a bunch of different tools in the ecosystem that basically give you a score. [1:12:13] based on your score thing. Yeah. And there's one called open rank and, [1:12:17] Um, [1:12:18] I think that this will all become emergent. And so we actually allowed you to link a Twitter account. And so that data is now on the profile. You're like on the protocol. So it's actually kind of an interesting thing. It's one of the few APIs because all the data is permissionless and you can kind of hit it and not. [1:12:33] to have to let anyone know you're doing it, but you could actually go do like an API for an airdrop, for example, that has nothing to do with Farcaster, that just checks every single Ethereum address. Does this person have a Farcaster? What's their Farcaster score through a different API? Do they have a Twitter account? Like what information about that? And so assuming you trust the data that we're putting out there, which I think most people would. But this is the benefit of an open protocol. Yeah, people just can remix and do it and you don't even have to let us know. So I think all that is happening.

[1:13:02] Going back to this non-cast thing, [1:13:04] I think... [1:13:05] If you just kind of play it out, you could look at the meme coin stuff and you can go – [1:13:11] Okay, this is stupid. This is nihilistic. This is just gambling. [1:13:15] But I actually think when you pair it with a set of developer Legos like Farcaster has, you start to get weird emergent new things. It's like part art, part business, part like. When there are stakes, the thing that people miss with explicitly the money piece of it, obviously it's tradable, whatever. But especially in a world of AI and abundant content and AI slop and all these things, in some sense, the monetary aspect of crypto is like a little bit of a counterweight, a form of physics that can layer on stakes. [1:13:45] If anyone anywhere could post to a non-cast without any stakes, it would be much less interesting. [1:13:50] Exactly. And every improvement and interesting experiment in crypto, [1:13:55] is immediately available. [1:13:57] you know, at least within Ethereum today, is immediately available. [1:14:00] to like plug into Farcaster. So it's, it's, it's kind of like composability and reinforcing mechanism. [1:14:06] that [1:14:06] another decentralized social network that's not [1:14:09] native, it's just like, it's too hard to couple. Whereas every forecaster user, by definition, like the whole thing works because there's an Ethereum wall under the hood. Totally. So by definition, every forecaster user is like natively interact with the native stuff. This is, I think this is what people dreamed of in the early days of the internet is just that all this stuff would work together. I have a couple more forecaster questions and I do want to get through them. So we have a few other things, but just quickly, um,

[1:14:34] You obviously just launched Framesv2. We haven't really talked about channels at all, but this is an element of the sort of subreddit idea being applied to Farcaster. I'm curious what... [1:14:43] you're either most excited about what you think is working best. Obviously there's this programmability stuff. Is there anything on the horizon? Is it V2, frames V2, [1:14:52] I think our bet is – so channels, I think have been a good – [1:14:57] Social exploration, I think it has not [1:14:59] been a growth driver for us just because I just don't think we've hit a community or topic that people think we have the best like content on the internet for especially outside of crypto yeah for certainly not outside of crypto it's been good for increasing the breadth of conversations on farcaster which I think helps with retention of like oh I can hang out with friends here on a variety of topics but in terms of like I want to join farcaster because it's the best f1 racing channel [1:15:22] that that that isn't happening and so i think where we've [1:15:26] Arguably a little confusing. They've changed a bit. Maybe people don't totally know when you post into our channel. I think you can always make the UX a little better, but what I would say is things that work tend to work even if the UX is bad. And if they're not, making the UX better is an easy trap. [1:15:42] And so I think where [1:15:43] The thing is working on Farcaster is like, [1:15:45] people are doing stuff on chain and we've taken the brain damage of like [1:15:50] building [1:15:51] on Ethereum for the last four years and like moving slower, having more kind of like problems to solve for relative to a different architecture. And now it's fine. And I think that's paying off. And so it's like, let's lean into that. Like, and, and, you know, we have a bunch of other tools that we can do to kind of like fix, you know, to the degree that someone's like too much trading, too many, too many mean coins. Okay. Let them dial their algorithm, whether that's like an

[1:16:21] you know, more intellectual or whatever the topic that they're actually interested in. Yeah. And so that's how I kind of think about it. So frames frames. [1:16:28] V2 is far more interactivity in the kind of app. So you can think of like an app within an app with a big focus on allowing you to do an on-chain action transaction transaction. [1:16:41] As easy as possible. Play a game, anything. Yeah, yeah. Vote, like do whatever. You can think of this as like a verb with everything that you're going to do. And just like that should be able to be accomplished on a blockchain. [1:16:53] instantly, cheaply, invisibly. [1:16:56] And I think we're actually finally at a place where the infrastructure exists for us to be able to go do that. When we started in 2020, it didn't. This is also starting to paint the picture. One of the first when we were talking about what is Farcaster, one frame of it is it's the single best way to interact with anything in crypto. This is obviously you could imagine a world where Farcaster starts to look more like a wallet or more like one of one of these frames for like the portal to crypto. Yeah. In a way that is pretty. [1:17:26] will look more like browsers because like it's one thing to have a balance and number go up. Yeah. Feeling very much like a bank account. [1:17:36] I think it's another thing to feeling like, oh, this is my launchpad. Front door. Yeah. Or, you know, whatever. And so I think the browser – [1:17:44] I think every app in crypto will look more browser-like. And whether you started as a bank account wallet and move towards a browser or your social network.

[1:17:52] And it feels more like a browser. And I think if you just think of a feed, it's like a browser. Like it's a different type of browser. [1:17:59] Yeah, that's where it's all going. Value capture, I figured I should ask about it. Obviously, being upstream of intent in crypto is pretty interesting. It probably means you're upstream of transaction. Is that something you're... [1:18:10] Thinking about more these days, I mean, I'm sure you get asked about it. Yeah. One component of that is when is the Farcaster token going to come out? Maybe you don't want to talk about that at all, but I'm curious. You are a business. Yeah. Or at least Merkle is. You're building this hopefully useful thing for everyone. But I think this is a question that comes up for people. Yeah. So I think the way to think about it is it's a couple different layers. The most obvious place to monetize is attention, which is like, to your point, you can show an ad. [1:18:36] which is pretty crude. But like, ultimately, that is a like, if your eye looks at it, and I can convince advertisers that, [1:18:43] You have looked at it a certain level that they're going to pay me. Intent is actually a lot more interesting because there's a lot higher willingness to pay. Especially in crypto. In crypto, right. So if like you see something like a trade or an NFT or whatever, and you can buy it on one of five marketplaces because it's just on a blockchain, they may compete. [1:19:02] to like be the default provider. Yes. Which – [1:19:05] So you could go build it yourself or you can just get the other people to pay you and then [1:19:09] they make some profit and you actually capture a lot of the value. Right. Most famous example here is Apple pays Google. And this has actually been ruled by the DOJ as illegal. But I think it's a little different scale when you're Apple. But the point is, is that. [1:19:21] It's known as TAC, traffic acquisition costs. And Google would pay Apple somewhere on the order of $20 billion a year just to be the default. So you can change it just to be the default. Defaults are really important. Clearly, it's worth it. Right. But ultimately, being in Safari and then having that, you know. And so –

[1:19:37] And iPhone users are valuable, right? They tend to be wealthier. So I think that equivalent in crypto, there are a lot of different ways to do that without even having to insert yourself into the transaction. I think there's another layer down where, sure, if there are common actions, you could also, and this is like the classic way the wallets monetize. [1:19:54] because they already have you in app. And when you go to click swap, they take a fee. The difference with Farcaster is I'm not only making the intent, the action, I'm finding out about what I want to do. Right. And so what I would say is that discovery and intent is usually above your credit card. [1:20:10] And so that's one way to think about it. Like it's worked the way it's worked in crypto because crypto is this kind of I have a bank account app and then I interact with the page and then I have to like explicitly link it every time I want to do something. That's changing. [1:20:24] And I think that's going to change in a world where I think that you're going to kind of like be browsing and you're like, oh, this is kind of cool. Interact. And then it actually doesn't even matter if you kick out to another wallet. Like the wallet is not in the place to decide how to route that transaction. Right. And so. [1:20:40] You know, really basic example. You – [1:20:43] want to buy, you want to swap ETH for USDC. [1:20:46] You can go do that on Uniswap.org. [1:20:49] Uniswap.org gets the fee, pushes it to MetaMask. MetaMask doesn't get anything. You can do that [1:20:55] from a command line directly with Uniswap, [1:20:58] There's no fee. Like you get the cheaper fee. You're paying it directly to the, you know, the liquidity pool or whatever. [1:21:05] If you do that same swap in MetaMask,

[1:21:08] They get to take. So it's all about where you start. Right. And so I think this is the thing that people maybe haven't fully internalized around why Farcaster could be quite. If it works, if it's scale. So so well, but I would view that as like a Farcaster benefit to any app. Right. So it's like any amount of attention in the Farcaster ecosystem benefits any of the clients that exist. Totally. I think I think at the protocol level. [1:21:30] Um, [1:21:32] the natural place to monetize is in terms of like, [1:21:35] the storage fees. So the idea of like, there's a finite amount of space on a decentralized social network. You delete cast still every... After like a thousand casts or something like that. But just super basic. It works for Ethereum and Bitcoin. There's not an infinite amount of block space. Yeah. So people pay gas. Now it can be cheap, [1:21:54] And you can do it once a year in the case of social media, like because that's a better pattern. Yep. [1:21:59] But if you actually want something to be decentralized, [1:22:01] And not have exorbitant costs for the size of how much you need from a storage standpoint to run a full node. You actually have to limit the amount of space. [1:22:12] And so the design of our network allows you to just charge rent per year to users. Now, we're, I think, keeping it lower for now, but I think over time, the market will dictate, like, what is the – [1:22:22] the value of a Farcaster account. And if there is alpha to be found or the reach of the social network is great, there will probably be some market clearing. I don't think it'll be exorbitant. I think it'll be $5 a year or $10 a year. ENS does this today. And this, by the way, can happen at multiple levels of the stack, which is to say there might be clients that are potentially better at getting alpha. It might be the network level piece of it. There's different. Yeah. Yeah. And then I think the other thing is channels is like a very obvious one where,

[1:22:49] namespaces are valuable. [1:22:50] I think for us, we've opted for the username space, because I think it's important in the Ethereum ecosystem as people can use [1:22:57] an ENS name, which they're already paying for. So that's actually Blue Sky has the same thing. It's like you can get a free name from Blue Sky or you can use your own domain. Like Farcaster, you can get a free name from us, or you can use the paid for ENS. - Right. - So I think user namespace, [1:23:11] Not necessary to monetize. I actually like it. It's like free or sovereign. Good options. But for channels, that's a Farcaster-centric feature. [1:23:19] Um, [1:23:20] monetize that namespace. And people right now are paying $25 a year for that. So 13,000 channels, like you could imagine. I mean, Reddit has several million subreddits. [1:23:29] Very few of them are actually still active. But the point is someone could be paying... [1:23:33] you know, because they could be economic hubs. There's a bunch of. Sure. So the point is, is that I think that there will be natural places at the protocol to to drive revenue for the protocol, which I think is important. And then I think. [1:23:44] Token, people always ask like when token. And I think my answer has always been, and it's consistent is, [1:23:50] What would it do? [1:23:51] Yeah, another question is when should a protocol ever launch a token? Well, that's my push has always been – [1:23:56] You need to actually have something valuable. [1:23:59] like before you have a token right and i don't think we've actually gotten to a place where farcaster [1:24:04] is kind of like a Lindy level value. Like, it's like this thing is going to exist. We know what the... [1:24:10] you know, we haven't even charged the storage fees yet. So it's like, why would you add a random asset that basically people have a kind of just assumption that the number is going to go up when you haven't kind of actually figured out? And you run the risk of damaging the network. Yeah, yeah, yeah. And so I think like with crypto people, like in the past, they just launched token, you know, they probably get liquid somehow through that. And then it's like, okay, other people are falling in the bag. And I just don't think that that's our...

[1:24:38] Yeah. [1:24:39] like mentality. It's like build something actually valuable. [1:24:42] And then maybe you could have a kind of discussion of like, what is the best way to decentralize ownership of the network? [1:24:47] but that's like, [1:24:48] It's like you wouldn't IPO a company. [1:24:52] if you haven't even like found product market fit, like that's basically adding a token. So it's like, [1:24:57] Talk to me if we have product market fit, like then I might have a better opinion on like tokens. [1:25:01] On that last note, a few people, when I mentioned I was interviewing you, asked about sufficient decentralization. That's been kind of your guys' North Star, at least on the decentralization front for a while. There's been, I guess, questions to what degree that's still a priority. I heard somewhere that maybe still a third of the team is working on decentralization and protocol. What is the sort of short story? We don't need to spend a ton of time on it, but where are you today on that? Yeah, it's funny. In retrospect, I wonder if we should have ever wrote the blog post because I think it's very misunderstood. The term is very catchy. [1:25:31] I mean, we made a play off of the fact that sufficiently decentralized was a term like I think is like Hester Pierce at the SEC used to refer to Ethereum. And so our our point was. [1:25:41] Sufficient decentralization of social networks means you should... [1:25:45] As a mentality, don't start with build the most decentralized thing. It's actually start from a centralized architecture and then work your way to the point of sufficient. Who's going to hold you to that? [1:25:54] It's squishy. And I think like, [1:25:57] If the market actually cared, no one would be using Blue Sky. [1:26:00] And the reality is, [1:26:02] I think a lot of people aren't going to like that answer. No, no, but yeah, let me, I think what I would say is like,

[1:26:10] One, the reveal preference of developers is if they believe that the – [1:26:15] The the aspiration of the team, the team has trust and credibility. [1:26:20] then you can actually go very long without actually having hit a level of sufficient decentralization or de facto or practical, whatever, whatever way you want to call it. Right. As long as they, they kind of trust that the team is actually going to get there. And so from our standpoint, [1:26:33] I think we've delivered on all of the things that we've promised and anything that we have delayed. If we have, we've been pretty upfront and shared the reason for it. [1:26:42] But we last fall, we. [1:26:44] And a lot of people were pushing us to do this, and we did do it. And anyone can sign up for the network directly with the smart contract. [1:26:51] Like you don't have to ask. It used to be permission. You have to ask me for an invite. Big deal. [1:26:55] Basically, no one does any sense. So it's like we did a bunch of work to like make a couple of like people happy. I do think it adds to the credibility network. I think it's like hard to remove some of the growth that we had. [1:27:05] This year, as a result of people kind of believing that it is credibly neutral, at least in that regard with signups and the underlying data and APIs hubs that that actually works. And so it's not like you're hitting our database to get stuff. [1:27:17] That said, like, [1:27:19] Okay, so our messaging system is not decentralized. [1:27:23] Okay. Like, does that like you could define, oh, well, I think a public broadcast social network like Twitter also needs to have a messaging system. Yeah. And so. [1:27:35] So that's an ongoing evolution. And it's like, I think, just an honest assessment. And I would say, yes, I do agree with you that that actually should be there. But that wasn't part of the original scope, right? That was added later as a retentive feature. Channels. Most of the stuff is available via API, but it's not in a smart contract.

[1:27:51] is that sufficiently decentralized? Like for the average developer, they're fine as long as we kind of guarantee the availability of those APIs and then have a commitment to getting it into kind of like more robust decentralized place at some point in time. But every time I'm having engineers work on decentralization, [1:28:09] Those engineers aren't working on... [1:28:11] the next frames feature, the thing that could potentially grow the network 10 or 100x. And so I think we just balance it. And every time I ask developers, the ones who are actually most committed to the ecosystem, [1:28:22] Like full-time jobs, like working on something Farcaster related. [1:28:25] Would you rather have 10 to 100x growth or me work on this feature over here? [1:28:31] And then have it later. They always say, make the growth happen sooner. Yeah. I think part of that is also they trust us. Yeah. And it's part of why having this type of conversation every so often is – [1:28:41] Helpful. Yeah, totally. One last kind of fun question. Somebody asked, Jamie Hoffman asked, why hasn't Dan created a token on Clanker, but did create NFTs on Zora? [1:28:53] Um... [1:28:54] I think creating a token has significant... [1:28:59] gray area [1:29:00] legal ramifications. Is that why NFTs are dead? Well, there's no secondary market for NFTs for the most part. Some can, but there's no expectation... [1:29:11] that you're going to really trade it. Like the Zora NFTs are pretty basic collectibles, right? Like you're paying like three bucks. You could mint as many as you want. Like, so I think it...

[1:29:22] Facts and circumstances matter. And I don't think anyone minting an NFT... [1:29:27] And by the way, they're not directly for me. The money that I've generated for NFTs, which is like some crazy amount of money, [1:29:33] for the protocol NFTs are all going to the protocol treasury. Or like selling, I don't know, yeah, selling some kind of company merchandise or something. Well, I mean, if I'm creating a clanker and it's going to the company or going to the protocol, it's effectively a protocol token. That wasn't very thoughtful, right? So that would be my answer. Okay. One of the things you've written about in the past and talked about a bit, I think originally goes back to Bology, is this idea of the idea maze. [1:30:03] very much in public with live public feedback. And then on top of this, there's this kind of classic Mark Andreessen post you've referenced around that, like the only thing that actually matters, team, product, market. In the end, it's actually all market. [1:30:16] kind of pulling us all down to this notion that product market fit is when the market is pulling product out of you faster than you can produce product. [1:30:25] You... [1:30:26] Obviously have had some progress. You've been navigating this idea of Mays publicly for a while, but you also seem to very much embody the like Andy Grove kind of like utter paranoia. Kobe jobs not finished vibe. I guess my question would be one. Where have you felt the pull the most on the product market fit standpoint? And two, do you have any kind of broad reflections on the idea Mays four years in? Yeah. So. [1:30:50] Let me talk about like, [1:30:51] the overall idea may is. And then maybe more recently, what I think the closest thing to product market fit for us, I think,

[1:30:58] You will not... [1:31:00] be able to [1:31:01] evaluate everything on like [1:31:03] Petzl and paper up front like you can't just like ideate your way through the maze and [1:31:07] As much as you can read history, know all the different startups, decisions they made, read all the blog posts and stuff like that, I think very much Brian Armstrong has a line, and I'm sure it's either his or someone else said this, but it's like action produces information. And I think the sooner you have contact with reality, the better your decision making is going to be. [1:31:29] And I think there's always a balance because you don't want to ship, like, a true, true MVP that, like, barely works, especially in consumer, which I think there's a certain level of, like, design and taste that you need. But at the same time, like, if my forecaster experience has indicated anything, like, you just need to get out there and, like, figure out, like, is there anything here worth even, like, making good design for? E.g., Elon Musk might buy Twitter two years in or a year into the thing, and a lot can change, and you need to get feedback as soon as possible. Yes, and so – [1:31:57] And that's actually another important thing is don't overhire. We've raised a lot of money relative to the progress in the scheme of things, and we've been really frugal and just determined to not – because we're not money constrained, so we could go hire a big team. I could go tell you I have 50 people or 100 people, and we're trying to figure it out. But I think that the scarce resource for us is just founder time and attention. [1:32:21] And we're at a scale with 14 people right now that basically Vern and I can do [1:32:26] be in the weeds on pretty much everything. And so you're getting like one to two founders, like input, like trying to like hone the best decision you can possibly make alongside an engineer or two who are very much empowered to like own a pretty beefy feature because it's not like some like crazy team with some, you know, long quarter long deadline. Right. So that speed is really, really important for kind of navigating the maze. And you should not be thinking in

[1:32:56] like month long. Like it should be, you should have like a rough six week plan. [1:33:02] And then you should be evaluating things like weekly and then potentially adjusting edits daily in terms of scope. Like, because every day actually matters. And you're saying that even today with Farcaster. Yeah, and it's been a grind in that sense. But I give my co-founder Varun a lot of credit. Like, I'm probably a little bit more... [1:33:19] I get distracted with a shiny object or can go off. And I think it's been something that I've adopted as a result of, you know, working closely with Varun is like, how can we cut a day off this? Like, if we want to ship it on a Friday, we ship it Thursday. Like, is there is there some way that we can just what can we say no to? What can we simplify? Yeah. And it's is every decision you're going to do as a result of that. Good. No. But like it actually just starts to create a culture of like. [1:33:44] Okay, what is the thing that we need to actually go ship? And especially once you have something, most features like, [1:33:51] 80% of it is actually what you need to get out. And then you can observe and it's like, does this work or not? And if it does, okay, let's double down and improve it. I think we've started to improve. Like we have a designer now on the team. So like the average quality of output that we're shipping from a feature standpoint. And to be clear, the engineers on the team are all really great. So like what they're shipping is quality. It's a little bit more of like product thinking requires a lot of time. And so a lot of the product thinking we're doing is like the 80-20 version. [1:34:21] week planning, like we'd think through all the edge cases and you'd actually be able to like [1:34:25] present the engineer with like a much more polished thing to go ship. Whereas Angar bias is like, okay, this is simple enough. Let's get it out there and see what people will complain about. It also happens where we have like a social network where people give real time feedback. So there's just less of an obvious correct decision at almost any time in a social context. Yeah. And the other thing is you can just copy other features. Like it's like the most copied consumer industry, right? Like everyone has a stories feature. We actually don't have the stories features, but like Snapchat invents it, Instagram copies it. And then everybody else

[1:34:55] I mean, LinkedIn has stories now, right? And so it's like, I think, [1:34:58] Feed is like a thing that everyone just copied, right? And so I think with social, like it's very much about the velocity of output because you're looking kind of for things that work. We didn't think frames would drive as much interest as it did. [1:35:12] right? [1:35:12] And so that was the closest thing we've had to catching lightning in a bottle. [1:35:16] And I think where we... [1:35:18] There's a version where we could have just kept going with frames this year because it was the closest thing we've had to like kind of product market fit. [1:35:25] But I think the challenging part is one, [1:35:27] It very much rode a kind of like asset wave with DGEN. And so hard to know how much this was frames versus DGEN at that point. Right. That was continuing the growth. Right. I think that there was an initial like frames clearly was the only thing that was doing this. But. [1:35:43] And there was a macro like Bitcoin and it had hit an all time high in March or whatever. And then [1:35:48] the macro changed. And then so I think frames weren't getting as much use. So even if we went and built frames v2 then, [1:35:57] Hmm. [1:35:58] we could have had all this additional work and then kind of nothing happening. Whereas [1:36:02] The market has picked back up. There's more stuff happening on chain. Now is a very good time to launch frames because you going back to the market recent framework that applies also to features, right? It's like launching a feature, like sequencing your features and like making sure the feature is actually hitting the market at the time that you need a little different in enterprise because it's like requirements tend to have a different crypto. Yeah. Crypto has a more. Yes. Yes. But. Right.

[1:36:28] But the point is, is like, [1:36:29] Any company, though, like knowing what feature to work on next is, [1:36:34] is like actually much harder than it sounds, right? It's not just like you just ask your customers and then seven of the 10 say they want this as the top feature and that's the obvious thing to do. You probably have like three features that three of them want and then one person wants something random and it's like, okay, well, which one's the most important? And- [1:36:52] At any point, you know, one or two features could actually be the thing where getting those right and just like really driving those from a sales standpoint is the actual growth engine for what you're doing. That makes sense. [1:37:03] You... [1:37:04] I don't know if you expected this. Do you think 18-year-old you would be... [1:37:09] surprised or would it make quite a lot of sense the broad arc of the career you've had? [1:37:14] I think it roughly line up with like I wanted to be in business. Okay. I wanted to be in technology. [1:37:22] My original plan was to major in computer science, but [1:37:24] My parents convinced me that I should, for a period, that I should go to law school. So I ended up being an English major, which was, in retrospect, a bad decision. I don't know. I might have paid off. I'm not sure exactly how. I ended up working in management consulting. I met my wife there. So that was a pretty big life decision. But... [1:37:39] I think if I was to kind of like go... [1:37:42] all the way back to 18 and say, like, would I be working in Silicon Valley or having worked in Silicon Valley and then be working on a startup of my own? Yeah, I would think so. I wouldn't. [1:37:54] I wouldn't have predicted... [1:37:56] Up until, frankly, 2021.

[1:37:59] like late 2020, 2021, when Coinbase like was preparing to go public, that Coinbase itself would be the, the like, [1:38:07] $100 billion IPO rather than a $10 billion IPO. I actually don't think anyone would have really. Yeah. It's a nice surprise. Yeah. 10X is a very nice surprise, but I think, um, [1:38:18] To be fair, though, if you kind of go back, it's like of all the options I could have worked on in tech in 2013. [1:38:24] by far the most contrarian was crypto. Yep. Like it was like Bitcoin, but like, and so you get paid in Silicon Valley by being contrarian. Yeah. Which AI... [1:38:34] We're, you know, the people at OpenAI who are benefiting from this now. [1:38:38] they were joining when it was kind of like they were lost in the wilderness. People forget, like, OpenAI was, like, making, like, StarCraft bots and, like, you know. Yeah, Dota. There was no, like, clear linear progression on the LLM. [1:38:51] thing, right? [1:38:52] A couple of them went to a talk at Google. They were like, let's go try to do this. They did it. GPT-2 was like a curiosity thing. [1:38:59] And then like three was like, oh, wow, this is, you know. Overnight success. Yeah, exactly. So I think, but to that point, and the idea may be, [1:39:08] If you're thinking of your career over a decade-long horizon or 15-year horizon, you're [1:39:13] The thing you're working in today is, [1:39:15] If you want a lot of upside... [1:39:18] Should feel a little weird. [1:39:19] And if you're in something that's pretty conventional or like smart people or MBAs are joining, like you might actually want to consider it's like you might end up having a fine career from a stability standpoint. Yeah.

[1:39:30] But if you want outside is upside. It should be actually... [1:39:34] contrarian or unpopular. Yeah. Yeah. Is there a reason you think distributed systems, let's say, to maybe draw the most zoomed out take on what it is? [1:39:44] It matters so much to you. Maybe it's just pure coincidence, Coinbase and then Farcaster, and obviously there's compounding there. But is it sort of very, very broadly freedom of speech and freedom of encryption of value? Is it something else? Is it a libertarian-esque view? Is it – [1:40:01] I think that the thing that I like about the Internet is it is kind of decentralizing from a power standpoint. [1:40:07] Right. Like in the pre-internet era, you have, you know, X number of newspapers, X number of television stations, the government. And like that, that's kind of like a very. [1:40:15] centralized power structure, I guess, business. Whereas with the internet, you've kind of unleashed the creativity of like, [1:40:24] anyone with a smartphone. And I think that that's a pretty... [1:40:29] powerful check on... [1:40:31] kind of like these big centralized institutions. I tend to find big centralized institutions to be ineffective, but, [1:40:38] at best and, and like, you know, massively incompetent in a lot of ways. Is there any irony to the fact that a whole bunch of people in tech now, uh, [1:40:46] think they should be running big institutions? [1:40:50] Well, I think a lot of the smart people are just trying to build new ones, right? And I think that's part of the ethos of startups is you don't go work for the big company and change it. You just compete with this one, right? And I think what the most romantic of a lot of these now are like these hardware companies, right? Like an Anderil, SpaceX, Tesla, Boom. Like if you just think of like these like iconic like post-war American companies like Boeing, it's like total mess today.

[1:41:20] or boom in any of those. And you look at it and you feel like this, this is the future of America. And none of those people went and worked at the big company and tried to like make it great. They just, just the cycle of how things have to be. Yeah. So creative destruction is, is I think like a, [1:41:36] I think there's a lot of appeal to that. I think it's much more challenging on the [1:41:40] government and kind of anything that the government heavily regulates, like healthcare, education. And so... [1:41:50] I'm not like one of these crazy radical people who's like, let's tear down all these institutions tomorrow. [1:41:56] Um, but like, I'm very excited to see like what Doge, for example, will, will do like, I actually think it's like a huge, a huge challenge because I think that. [1:42:08] And maybe they're not underestimating themselves, but like... [1:42:11] I just think Trump's going to be a bureaucracy. Trump's going to be like, yeah, that will be a little bit of an X factor. But my point is the bureaucracy is like it's going to want to survive. New level. Yeah, it's like, yeah, you're fighting a new boss. But my point is, is that. [1:42:25] I don't know, in my lifetime, we've never lived through like an opportunity to like massively shape how the government like the government has only just grown. That is like the constant. And so like a refounding of some of these like kind of like core federal institutions. [1:42:39] Yeah. Agencies, regulators, regulators. [1:42:41] that I think could be like a big unlock, right? Like I see it from my own perspective of like how detrimental the, [1:42:48] lack of regulation and policy or regulation by enforcement, the SEC has kind of done in the crypto industry and like all the debanking. And, you know, it's funny. It's like Marc Andreessen's on Rogan.

[1:42:59] This week, and he's talking about, like, people getting debanked. And everyone's on Twitter being like, [1:43:04] Wait, what? And like Brogan's like, what? People are going to get banks. And I'm like, yeah, like I was on the forefront of that in tech. [1:43:11] Because I was in charge of the factory. Coinbase. It's like, yes, I'm well aware of this. [1:43:17] And I don't think any of those people are. I mean, there are actually a few, but I think generally most of those people who are in those regulatory agencies for the financial system are not malicious. [1:43:26] They're just risk-averse. And like, there's no incentive. They don't get a star. [1:43:30] or promotion for encouraging innovation. They get... [1:43:35] you know, kind of rewarded in the system for just keeping things stable. And, you know, [1:43:40] I think stability is good under certain axes. And then others, like if you just tried to keep the economy stable all the time, you'd be Europe. Like you don't have... [1:43:50] Like, civilizational progress. Creative destruction feels like the right word for it. Yeah. And look, you can't have everything have that, right? Like, people need some level of stability. But the beauty of capitalism is it creates reasonable stability if you pair it with, like, good government while at the same time moving things forward quite a bit. [1:44:09] Can you tell me a little bit about what it was like? You joined Coinbase in 2014, you left in 2019. Obviously things started to go pretty well in 2017. [1:44:18] Maybe for the first time. I guess when you came in, it was decent. No, 2017 was the first good year. What were you feeling and what was it like in 2015 and 2016? And what did you learn? It was pretty brutal. I think it's like 50 people at the company and like 25 of them left on their own volition. It was a lot of turnover. I got close to leaving. Yeah.

[1:44:41] It just... [1:44:42] There was nothing happening. And it took us actually adding Ethereum in 2016. Fred actually was the one that kind of really pushed for that because the identity of the company was a Bitcoin company. Right. Like I think it's easy to say Coinbase today and think crypto. It's like Coinbase was a Bitcoin company. [1:44:59] Coinbase, the term is from like the Bitcoin, you know, technical, you know, literature. And so – [1:45:06] It was, I think, good for me in the sense that I had a job that was like, [1:45:11] higher stakes, even though the company wasn't necessarily growing as I was managing these bank relationships. So I was learning a lot. And, um, [1:45:18] It gave me good visibility within the company in the sense that I was working on an existential problem. [1:45:24] But, um... [1:45:26] It was like crazy the amount of... [1:45:28] learning I had [1:45:29] call it 2014 to [1:45:31] end of 2016 and then just like what in 2017 through call it mid 2018 it felt like i got another [1:45:38] two or three years and, and, um, you know, the short period of time, uh, [1:45:43] But yeah, like... [1:45:45] The biggest takeaway for Coinbase, though, is I just saw what hypergrowth looks like. Yeah. I can remember being the startup where we could all fit two lunch tables and to then see it be a thousand-person company where you just don't actually know anyone anymore and you understand – [1:46:01] Um, [1:46:02] The kind of crazy hyper growth challenge of like at any given point, 30 percent of the company is like less than three months older. Like and so your culture changes like you bring in these execs. And so I think that was that's really informative. And what's funny is like that's a skill set that I have, but I haven't actually had to apply to Farcaster.

[1:46:21] in the pure companies because the company is small. But there's actually a lot of lessons and parallels to like growing a social network. It's like the culture on the social network changes. Like how do you manage that? And like, you know, we did a lot of like employee like comms and thinking about that. And now Farcaster, I've just done this every day for four years. But like I think I have a decent sense for like managing messaging to a large group of people and doing it pretty quickly and directly. But like so you do kind of realize that some of this stuff actually translates. [1:46:51] Hmm. [1:46:52] Are there any anti-lessons from Coinbase, meaning stuff that you might be tempted, or others might be tempted to generalize, looking at its success that you think were... [1:47:01] Very specific to that situation. [1:47:03] Yeah, I would say two – and you try to criticize Coinbase, and the reality is it's an extremely successful company. So take everything with a grain of salt. But I do think that the company – [1:47:13] And part of it is just like you couldn't attract the senior people, but – [1:47:18] We did not hire enough senior engineers early on, and that became a very hard problem. Like, no one wants to come into a company with, like, a bunch of junior engineers if they're senior. And so that, like, that really hurt us, I think, during the hypergrowth phase because – and – [1:47:33] To be fair, a lot of the people who are now senior engineers on my team, I kind of went through Coinbase and learned on the job there. But I think it's very challenging scaling a company if no one has actually been there before. And so that was something that Vernon and I were pretty determined to be like, we're going to really focus on hiring a senior. I also think that the other thing with junior employees generally is there's a lot less emotional stability.

[1:48:03] and stuff. That is just like, as a result of like, [1:48:06] people basically treating it as an extension of college. And so, [1:48:10] I think we haven't had any of that. [1:48:12] with a small team and a lot of senior people, which has been nice. And I think that the second, um, [1:48:17] is I think from a strategy standpoint, [1:48:20] We... [1:48:21] We tried a lot of different things between 2014 and 2017. And in retrospect, we just focused on the core business, like, which is easy to say, but like, really just like, [1:48:32] Just focus on the on-ramp, like expand, like don't, don't like do all these other things. [1:48:38] And, and, and, [1:48:39] I think the company is actually fine now because their competitors like have basically been vaporized from regulatory standpoint because they were doing illegal stuff like FTX. But that might not actually be like the learning that might have been maybe coincidence is not the right word. But well, I think it's just like a good lesson in like focus it and and. [1:48:57] It's easy to want to like, OK, we've made it here. Like, let's go work on a portfolio of things. And it's like, no, just just be good at like one thing and just keep driving that. [1:49:07] Until it like really hits scale. And so I think that that's a lesson for me that I will, you know, continue to be relevant. Yeah. It's just like, [1:49:17] fewer priorities and just try to do them really well. [1:49:20] Brian seems to be someone who's both hired a number of very opinionated, otherwise strong-willed people who, as far as I understand on things like USDC and maybe even BASE, has not necessarily agreed. [1:49:33] and yet has been able to disagree and commit and also like oftentimes either empower or even maintain people. You think about somebody who's still there like Jesse.

[1:49:43] You've talked a lot about learning from Brian in the sort of volatility sense. Do you have any analysis of what enables him to do that so well? Because that seems pretty rare. At least we have a lot of CEO prototypes of the sort of great man, hero, top-down, Zuck, Elon thing. Brian seems to... [1:50:01] be pretty good at finding these people who probably wouldn't otherwise work for someone [1:50:06] and empowering them. [1:50:08] Yes, I would say so. He really started this like kind of obsession or maybe it preexisted. [1:50:15] Coinbase, but while I was there, and this is kind of in post-2017 growth, he was very focused on repeatable innovation. He didn't want Coinbase to be a company that only could do one thing. Mm-hmm. [1:50:26] And, and... [1:50:28] So he kind of like went and talked to a bunch. He's very good at soliciting input from other wise people in Silicon Valley. Like basically he's a good student like that. And so we'll get the meetings or dinners or whatever and go and talk to 10 people and then come back. Patrick Carlson is also pretty well known for doing this. But Brian came away with this kind of Google framework that he likes where it's kind of – [1:50:49] 70-20-10. So it's like 70% on the core business. [1:50:52] And this took a little while to develop. And I think... [1:50:55] Um... [1:50:57] Is the right thing for now, their scale? I was going to say, maybe it wasn't right for 2017, but it's paying off now. Yeah, so I think... [1:51:04] What I would say is there wasn't the same level of rigor in the framework in the 2014. It was a little bit throwing more spaghetti at the wall to hope, hope, hope stuff would work. Yeah. But 70% of the core business, 20% on kind of like, I'm like emerging, like businesses that, that are kind of like working now. And then 10% on like completely out there. And the goal of the CEO is capital allocator to like, kind of make sure that the projects have the right funding. Yeah. And then on the 10%,

[1:51:32] the thing that the CEO founder especially can do is protect them from [1:51:37] the inevitable like execs who like have going to kill them yeah for whatever political reason uh one way or another and you need one or two of those you need base or usdc to work and yeah and and so each one is always going to have like a weird set of circumstances like i'm sure there are war stories as to like what actually happened at coinbase for base to exist yeah like i don't think there was an obvious like win i don't think anyone thought actually it was going to be as successful as it was maybe other than jesse and and i think similar to the usdc like [1:52:04] So not my story to tell, but that that was like a very interesting kind of like how it came to be. And, you know, biology was very involved and kind of like making the case for it. [1:52:13] Which is kind of funny to think like biology is the reason USDC is like really successful. Like, you know, given its US dollar or whatever. But my point is like, [1:52:20] I think... [1:52:21] You do need a little bit of luck in some of that stuff. Speaking of Balaji, if you could direct his... [1:52:27] gaze, let's say his attention to one to think about work on or even just talk about one thing for a year. What would you choose? [1:52:36] Well, I don't think he'd want the job, but he would be my pick for – [1:52:40] person to run Doge. I think, you know, a lot of people, that would be a nightmare. But I think having seen him work at Coinbase, [1:52:50] he would literally be the perfect person to do that. Like, cause he would drive the team to do the right, [1:52:57] level of like, yeah. Yeah. So I don't know if you could work alongside like Elon and Vivek in terms of just let him in there for a month. Yeah. And frankly, it's like, you know, they're big, all big personalities. But but yeah, I think that would be.

[1:53:16] When biology gets obsessed with something, it's like, just make sure it's not, you're not in the path. Right, right. That was, that was implied in my question. You, you know. [1:53:24] Seem to be pretty plugged into the sort of like Silicon Valley group chat. [1:53:28] as an idea. Um, I, I've always said like along with Twitter, the best social product ever is the group chat. Um, I have two questions on this front. One, [1:53:36] To what degree are you thinking about maybe capturing the group chat opportunity that Twitter didn't, which is so many WhatsApp signal, even text group chats, originally known Twitter, weren't captured by that product. To what extent are you thinking about it? [1:53:48] Twitter. And two, [1:53:50] What makes a great group chat slash what have you learned from the best group chats? [1:53:54] Yeah, so – [1:53:55] I would love Farcaster to be able to have that. I actually think... [1:53:59] if you're not a dedicated messaging client to hit the level of like just polish you need for people to use an app like think of it just like how buttery smooth telegram is or whatsapp yeah let alone signal which has you know the best in class encryption and you know the kind of brand that people feel good about like saying something they don't want yeah you know also very easy to move to another app in today's world yeah so so but i think with that i think that [1:54:27] people underrate how difficult it is to build that level of UX. And those are not good businesses. WhatsApp is a good business because it's part of meta and it's got crazy scale. But I think... [1:54:39] I mean, like they're good businesses like relative to other, but like Instagram is an amazing business. YouTube is an amazing business. Right. And so like Telegram, I think is something on the order of like 900 million monthly active users, a billion monthly active users. Yeah.

[1:54:50] And I think I saw a stat. It's like, [1:54:53] Call it a billion dollars in revenue a year, but it's got a lot of costs. And so it's going to create tremendous network lock in at least if you can get them like for a fire caster, let's say, or webcast like. [1:55:02] Right. But then there's the issue is you need to get to scale to even have that work. And so there's like this, like there are a lot of messaging apps on the way. So there's a little bit of like survivorship bias. But [1:55:12] The point being that [1:55:13] like Instagram is, I don't know, generate 35 billion or whatever crazy or YouTube, like the amount of value you can generate with a feed based product as it relates to ads is significantly more. I think so. So. [1:55:25] I think that's really challenges like one just to hit the UX bar. [1:55:28] requires a lot of time and energy and goes back to this idea of focus. Is this actually helping growth forecast or what, where it's uniquely differentiated? [1:55:36] Probably not. It's a retentive feature for Forecaster, right? Like chats keep people around. That's where I'm sure for Twitter, it's like if they look at like people who use DMs on Twitter, like are more likely to come back every day, even if their group chat isn't there. [1:55:48] Um, [1:55:49] with [1:55:50] What makes a good group chat? I think... [1:55:53] So my understanding... [1:55:55] is I'm sure there were some group chats. [1:55:58] kind of like pre COVID, but I think group chats really took off. Hmm. [1:56:03] Right. During COVID. Like I was in a few before and you kind of just, it's like getting invited to a party and it's like, you have one person and they add you to the chat and, [1:56:12] you know, kind of there's a level of like, OK, what's in the chat stays in the chat. And like if you're some type of person who's talking about, [1:56:20] you know, what they heard in a group chat and people hear that, then you don't get invited to the next group chat. So people, people like Silicon Valley, people are very good at like optionality maximizing. So you like, don't want to like lose your ability to get invited to group chats. That's why you don't see like leaked signal messages. Like journalists, right? You get a lot of like

[1:56:38] third-hand stuff, but people are very sophisticated and not... It's game theory optimal, actually. Yeah, and you don't want to mess around with people who can hire really good security people, or they have the N number of group chats to know that you were the one that... Yeah, yeah, yeah. [1:56:51] So, yeah. [1:56:52] I think that that level of trust in an environment where there was like [1:56:56] kind of like... [1:56:57] the like woke movement [1:57:00] Very much like, [1:57:01] Trump 2016 through [1:57:03] You know, I would say peak is summer of 2020. It used to probably happen on Twitter. Yes, increasingly cancel culture, all this kind of stuff. Like if you just like look like Naval used to tweet like, you know, all the time. And then it was like gradually reduced down. [1:57:18] That's an example of like someone within Silicon Valley who – [1:57:22] you know, just moved his [1:57:24] His... [1:57:24] those tweets moved to group chats, right? So think of like the kind of people who have the most followers in Silicon Valley. And then I think what changed during [1:57:33] like peak during COVID, but then it really kind of accelerated. And then I think now it's broken out in the open is once Elon bought Twitter. This is going back to what I said before. It really changed it. Like they kind of killed group chat culture. Do you think people are back to posting publicly? Oh yeah. Like way more than they were. Oh, interesting. Interesting. Right. I mean, just look at Mark Andreessen's like, [1:57:53] podcasts that he did on Rogan. Like, if you just listen to that, [1:57:56] The amount of things that would just literally have never been uttered by anyone in Silicon Valley for fear of being canceled. [1:58:03] It's uninhibited now. Yeah, and I don't think anything was like crazy unhinged. It's just like these topics, like you just could not talk about them publicly. I think that the group chat culture like kind of was like a Trump era, first era, where it was like peak wokeness. Right. COVID very much like drove the...

[1:58:20] Just a pure response to the public forum, the town square no longer being a quote unquote safe place. Yeah. And then I like I think it's a fair criticism to say it's like obviously it's safe. It's just like was not a like the Overton window for what you could talk about. Right. Somewhat limited. Right. And so it just naturally you got all these people creating signal groups and WhatsApp groups. Right. I think the other big thing, though, is like signal took some time to grow. And then what's happened? [1:58:45] went and it encrypted and like people trusted that. Like, so it's like people were like, okay, this stuff is like, you know, disappearing messages is important. Like people just want to make sure that this stuff is not hanging around. Yeah. Who do you think is more powerful? Elon, Zuck or Donald Trump? [1:59:00] I think the... Or Peter Thiel. We'll throw him in, given our conversation earlier. Yeah, I think... [1:59:08] Trump, I mean, [1:59:10] you know, call it Biden to Trump, like that, that role is still pretty powerful. Like you can do a lot of damage. If Zuck could switch today, he would take presidency. [1:59:18] Zuck? Yeah. [1:59:20] I don't know him well enough. I think like, [1:59:22] If his goal was power and influence. Sure. I think any of those guys would, if he could just magically be president for four years. If you're running like, I mean, Zuck runs a two billion person network state, right? So he'd be like, yeah, let me try my hand at doing this. I mean, Elon basically is doing that quasi with Doge. I don't think Elon actually wants to be president, though. [1:59:42] Nor, though, I think you just saw it. Yeah, I think at a certain level of your like that stature, like that, that's like I could do that job type type mentality. Again, I don't know these people directly. But are you speaking personally?

[1:59:53] No, I think people are interested in being like, what would I do if I was in the shoes? I think the... [2:00:00] The point, though, is that the job is a lot more ceremonial. [2:00:04] other than a couple of specific instances which can be very impactful given that you can control the nuclear weapons but like [2:00:10] You know, you're beholden to Congress on most things, and there's a lot of horse trading that goes into that role relative to a company like Zuck or Elon can just make – [2:00:20] you know, snap decision. And that's, that's the way it works. Right. Do you, [2:00:25] Clearly, you're interested in politics on some level. Do you have a view on kind of where things are going from a either the fate of the Republicans or the Democrats or just broadly? We have former years of Trump, and then it feels like we're in totally new territory. Is this one you think about at all? [2:00:39] Yeah, I... [2:00:40] My sense is populism is very much on the rise. I think that is fueled by the Internet. [2:00:45] I also just think the post-war coalitions. [2:00:49] That, you know, it's like union people are voting Republican and like, you know, educated whites are increasingly Democrats like that. That was not the case. Right. 50, 60 years ago. [2:01:03] you know, minorities increasing their vote for Republicans. So I think that the country is going to tack – [2:01:10] more populist. I would expect another version, probably younger version of Bernie Sanders to run on the Democratic side, maybe AOC, which is a slightly different [2:01:21] you know, flavor, but like, I think the...

[2:01:23] people who are good at and going all the way back to this discussion we were talking about before of just like going direct. [2:01:28] having distribution. Yeah. I think, um, yeah, the podcast president, I think like if you're, and you see this with AOC, like, [2:01:38] whether or not she's she's good on policy or whatever she has an instagram audience she's she's [2:01:44] learning that now she gets feedback, she goes on live streams. Yeah. Yeah. So like that's, that skill set is, is far more valuable than a rally speech. Yeah. Like, which is crazy to me that we're even spending time like, [2:01:57] showing a little clip, you know, it's because there are still boomers out there who are like watching TV. But I know who trust the preciousness and scarcity of that small format versus like 16 hours of Trump on a podcast. Who cares? Like, right. But that's why Trump had a huge swing. 18 to 29 year olds is like they don't watch. [2:02:16] the rallies. Like, they don't watch MSNBC or CNN. And so I think, like, [2:02:21] That will be a big shift in politics. I actually said this during the Clubhouse Mania in 2020. [2:02:28] It's just so clear to me that that format, [2:02:31] of audio direct, like long form. So you feel like you can get not edited. Yeah. And you just you understand that person a lot more like this. Like, I think someone's going to have a far better understanding of like who I am as a person versus like if I go give some canned five minute YouTube video. Yeah. [2:02:48] So I'm increasingly convinced like that populist direct –

[2:02:54] to consumer. [2:02:56] uh, [2:02:57] candidate will outperform. [2:02:59] And I think that both parties will kind of blend. So I think populism... [2:03:05] inherently is [2:03:07] not focused on the issues that elites care about. Right. So the Democratic Party, I think, is very captured by like a very small group of elites, like whatever set of interests. [2:03:19] And if you look at anything from the shift from 2020 to 2024, [2:03:24] is that, [2:03:25] you know, kind of like [2:03:26] lower socioeconomic and minorities. [2:03:31] shifted hard right on a relative basis. And so the populist policies are [2:03:37] whether you're Republican or Democrat, need to actually resonate with people who like they don't want Paul Krugman explaining to them that inflation is transitory and actually that the Biden economy is better than. [2:03:48] the Trump economy. Like that's like an expert opinion that doesn't matter. Cause like, he doesn't have to think about that where it's like, you go and the price of milk and eggs. And like, yeah, I went the other day, I got a coffee in Venice and it was like $7 for a black car. Like it's just, it isn't a fancy place. It's just, they can charge that. And, and so I think the [2:04:11] And that's not even like average normal person, but like the fact that just like the economy is, [2:04:16] And within a very short period of time, everything feels like it's doubled in price. Meanwhile, also, the other side of the Democrats is like, hey, you're a part of this, like, identity politics group that you should feel exactly how all the other Mexican Americans feel. Yeah, I don't think that that's popular either. But but I do think that the.

[2:04:34] populist policies are going to be focused less on whatever culture war issue that people are terminally aligned. Want to fight about. Yeah. Right. Which I think like, [2:04:42] If you look at how the... [2:04:45] Kamala's campaign, like if you were only viewing it through even Twitter for a while, it felt like she was winning. [2:04:52] Yeah. Right. [2:04:53] Like it was like brat and like think about just like how good she was online. [2:04:59] on TikTok, on like all these places relative to. Even the debate, post debate, I think. Right, and the debate is even a great example. No one actually watches the debate. Right. But the online people watch the debate. Right, right. And they think that. [2:05:10] There's some like framework to say, okay, she's, she won the debate. This special moment, this constrained amount of time. And people are making their decision on the debate. That's so antiquated. It's like, no, like the fact that. [2:05:21] Trump did Rogan and Kamala didn't want to. [2:05:25] And then did this contrived, like, call me daddy or, like, whatever. I don't even watch that. I think that's the wrong example. She just didn't do enough of it, really. But that's my point. It's like, okay, we're in 2024. It's like you should be on every single one of these podcasts. And you need to... [2:05:39] And here's the thing. Podcasters are not journalists most of the time. And so they're going to be, if you go on Rogan, he is going to be actually, he is not going to be hostile. No. Right. We're going to need more of that, by the way, if this continues, as I think you're actually going to need to have people, need to have people who can push, but. [2:05:56] Right. But the other answer is like, do consumers actually want that? Like the reality is like, do I want to go listen to some? I think we have to aspire to more. At least you have to aspire to have somebody who can really push Trump as an example.

[2:06:10] So I understand that sentiment. But the question I'd have is what incentive does he have or ignore Trump? Because it's like if I say Trump, people just get so mad. Sure. J.D. Vance. Any candidate. Any candidate. Why would you go on hostile media if you didn't think you could go on there and perform well? Because the whole goal as a politician is to get people to think you are doing well. [2:06:40] person at some point, if they're just going on to get layups over and over again, over and over again, like people will see through that. Yeah. But, but, [2:06:49] Part of it is you can't ask the guy or, you know, the candidate gotchas every five seconds. Of course. Of course. Right. Which is, I think, one of the things about the debate that is like, OK, you're going to fact check one side versus not. And like. [2:07:00] Whether that's true or not, who cares? The point is that that's the perception. It's just not. It's not a great point. [2:07:06] And so I think that the, [2:07:09] regardless of if everyone thinks we're going to have this like civic society with like tough, tough journalists or whatever, if media doesn't work like that, there is no – [2:07:20] There is no legal requirement that you have to deal with people who you don't like. It's a bit – maybe making a stretch, but it's a bit like the Field of Dreams thing. It's like you have to actually have people show up and have people hear it. And if people aren't getting – if the candidate isn't getting in front of people in some kind of meaningful way that they actually care about seeing, then it's all moot. [2:07:38] And I think that people...

[2:07:41] Journalism, as in the kind of like post-war era, is a function of... [2:07:46] a limited number of television channels, radio stations, and newspapers because distribution costs money. Yep. It's like Ben Thompson. And so as soon as you move to this, like, you can pick – [2:07:56] Well, there are people with massive audiences, right? Like, what's stopping Mr. Beast from having one of the candidates come on there and then asking five? They're going to fight each other, yeah. But that's the... [2:08:05] And so what do I think will happen is [2:08:08] there will be someone who realizes that there's an actual market for [2:08:13] a nonpartisan, [2:08:15] Tough but fair. [2:08:17] person who will ask candidates, let them speak, and then let the voter decide, not with the gotcha that you asked, but with the bumbling answer that the [2:08:26] Canada, right? Mm-hmm. [2:08:28] And I think you will get the candidate to show up if there's audience and you will get an audience if you can. Like, so first you have to build the distribution and then you're able to actually ask those. You have to trust the viewer and the listener to. [2:08:40] You have to give them some credit. I think Rogan, to his credit, historically, has done a fairly decent job. Right. And that's why I think that it was a huge mistake that she didn't write. Right. One of the things I really admire about you is you you're like a great version of the like strong opinions loosely held broadly. Is there anything you've really or materially changed your mind on in the last, let's say, year or? [2:09:00] I think I had an opinion that there would be a single winning decentralized protocol, like a single kind of town square, because that's how Twitter worked. And I'm increasingly of the opinion that,

[2:09:13] it will be pretty fragmented. [2:09:15] And so I think that that's had a pretty big impact [2:09:17] in terms of our strategy as like, it's like, I don't think we can actually win [2:09:21] at scale. [2:09:23] As as this kind of like Twitter replacement, nor do I think others, I think just be fractured. And so I think we we've actually tacked into the saying our thing, at least in the near term, is going to be smaller. [2:09:34] than maybe what we were aspirationally hoping for. But yeah, [2:09:37] by being more focused and by being more crypto native. [2:09:40] It can be valuable and grow from there. And so I think that that's been a tough one because I think going in, it was like, oh, building the thing to replace Twitter is... [2:09:51] on the long term. And I actually just don't believe there will be one thing. It'll be, it'll be many. And so what I'm hoping to build is, [2:09:59] is something that is. [2:10:00] valuable and specific enough in crypto. Do you remember the either period or the moment where that [2:10:09] switch started to happen, we were kind of forced to finally really see the reality on that? I think it's the last three or four months. [2:10:15] I think I've kind of had an inclination before, but I think it's pretty crystallized this fall. Yeah. So that's been an up and down for me. Anything... [2:10:23] My penultimate question, anything recently that really moved you or that you found beautiful? [2:10:29] I get that daily with my two kids. That's a roller coaster of emotions as well, right? So I don't do much. I work and I spend time with my family, but I have a three and a half year old and a soon to be two year old and we have another one arriving fairly shortly. Wow. Three boys and...

[2:10:47] Yeah, I think... [2:10:49] the [2:10:49] The pure joy that they can create is a feeling you you lose as an adult for the most part, especially if you're spending a lot of time working. You don't like have joy happen. I don't get joy sitting in front of my computer most of the time, if any ever. And I think with kids, it's like you can go from a crying, annoying bratty kid to a moment of joy within a five minute span. That's that's. [2:11:13] Especially if you step back, it's... Yeah, yeah, yeah. That's cool. Yeah, and then I recently... I mean, this is just... [2:11:20] kind of silly but i recently went on vacation with my family um we were in mexico we went to the beach and like [2:11:27] there's just like one night there's a beautiful sunset with the whole family. And that was, that was a very nice, just, [2:11:34] calm, but kind of like... Stop arguing with strangers on the internet for a little while. It's like a reminder of what it was about. All right. My last question is, what is it going to take to get you to change the Warpcast logo? Yeah, you've been on this for a while. Mike Rainbow is also a big believer. For those who aren't aware, the Warpcast logo, which is the main client of Farcaster, how you use the product, is a purple square with a white W on it. Yeah, [2:12:04] Yeah. [2:12:04] Yeah, I mean, at this point, it's pretty iconic. You know if you hit that, you're going to need some dopamine. I refuse to admit it's Lindy, but... But yeah, I think we will... [2:12:13] How about this? Product market fit, we'll have a shining. Wow. You'll know product market fits happen when the new logo drops. Yeah, when the market is pulling it out of you. Dan, thank you very much. Anything you want to tell people before you go? Obviously, you guys are announcing stuff left and right. I think you just announced starter packs, channels V2, or any broader PSA. No. If you want to follow me, you can follow me on Farcaster, which is DWR. And then I'm also DWR on Twitter.

[2:12:43] We'll get some of the crypto newbies to pay a little more attention. Cool. Thanks, man.