Bitcoin.review Podcast | BR018 — Nostr Deep Dive Panel with NVK ft. Jack Dorsey, Fiatjaf & William Casarin
Link to the YouTube (the timestamps are based on this): https://youtu.be/GU37FW5M83Q
NVK: I have some some interesting guys here today, and we’re all working on Nostr — it’s this new thing that seems to be taking the Internet by storm. So why don’t I intro today’s panel: We have Fiatjaf [no comment]. We have JB55 [William Casarin]. JB’s working on Damus which is one of the clients that’s been well-used right now. Hello sir.
William Casarin: How’s it going guys?
NVK: Thanks for coming. And we have Jack: Jack builds the competitor to Nostr and he’s gonna join us to add some color to this.
Jack Dorsey: Awesome, can’t wait.
NVK [2:03]: Okay guys so a good way to start, because people have absolutely no idea what Nostr is — people are super-confused. I wanted to just give a very short explanation of what Nostr is — we can get into what it isn’t later — and the state of development: where we are now. JB, I’ve seen you go around now to a few shows — do you want to give an elevator pitch about what Nostr is?
William Casarin [2:30]: Sure I’m on three hours of sleep, but I’ll try: I always start off by saying that Nostr is a protocol. And the minute I say protocol I feel people’s eyes glaze over a bit because it’s like, Okay now this is some nerd talking to me. But I always try to say, Hey well we have these things called protocols, and e-mail is one example of these which we use day-to-day. I always try to say, Okay Nostr is the protocol for social media, so instead of e-mail clients we have social clients. I always like to start with that, because obviously we all know that Nostr is more than just social, but it’s a good way to ground people into something more concrete.
NVK: Yeah that’s a good beginning: we had a protocol for e-mail, we had a protocol for other things, and Nostr now is this really good protocol for many things — one of them being social media. Fiatjaf, do you want to give a quick update of the state of development for Nostr?
Fiatjaf [3:32]: Well I don’t know if there’s a “Nostr development” — there’s just a bunch of people making a bunch of clients, and Damus is the one that looks the most finished of them all. But there’s no “Nostr development.” There’s a lot of clients: people are making a lot of clients — every day I see two new ones — but everything is very not ready.
NVK: I love that. We can get into some of the NIPs and where things are and some of the very big open questions. Nostr was fairly known by a lot of people for quite some time — the project is close to two years old — but we had a huge run up in users once Jack came onto the platform and started greeting every single person that comes in. So Jack, what brought you to this platform? Why Nostr?
Jack Dorsey [4:34]: Well I tweeted out just a short post about some of the lessons learned from Twitter and work in Bitcoin, and committed to moving a bunch of — I have this foundation called #StartSmall to give grants all over the world for a few particular problems that I care about — and opened a new category which was: free software, open Internet protocol development. And as I posted that and asked the question of, Where should some of these funds go? Nostr kept coming up in the replies. I’d seen mention of it before briefly, but unfortunately back then — maybe a year ago — I didn’t look deeply enough into it. But when I kept seeing it again, I went to the GitHub, I read the original [readme.md], and it hit every single point that I had been looking at and concerned with. And then I found a bunch of the clients — I played around with it, I found Damus and played around with that even more. And the more I played with it, the more I interacted with people, the more fun it became. And that — plus the underlying fundamentals — just made me really happy! And I wanted to do whatever I could do to support [Nostr], so I had a brief conversation with Fiatjaf and William — and here we are.
NVK: Yeah there is a special feeling to this platform. I say platform, not the protocol and not the client, because that feeling — at least right now — is coming from the social media use of it. It feels like the BBS days, it feels like a bit of the IRC days, it feels like we have people who are generally interested in building something new — also similar to how Bitcoin started. It has a certain stickiness to it, and at least in my opinion I can attribute this to: you really don’t need permission to participate in Nostr, and people who build software are coming in like a torrent and building clients and relays and different uses of it. How do new protocols take off? Because everybody and their mother wants to build a new protocol, everybody can design a complete protocol and have amazing documentation — and then nobody uses it! That’s normally the fate of 99% of all protocols that have ever been created. Why is Nostr taking off, in each of your opinions?
William Casarin [7:37]: NVK, you ran the Bitcoin Hackers Mastodon, so I’m curious what feels different about this compared to Mastodon? Because I was on ActivityPub for a couple years just playing with it — I probably got banned from two or three instances just for saying Bitcoin — but this does feel quite different. And I’m trying to figure out why that is? Is it just the type of people who are joining it at this time? Is it the protocol itself? That it enables a different type of interaction? Because I don’t feel the need to censor myself or the need to worry about the instance admin banning me. It’s interesting how protocols can shape the dialogue itself, which I find fascinating.
NVK: Yeah the medium is the message — it truly is. And people shape — at least in my view — their thoughts, their expression of themselves, based on the medium that they use. You have the radio voice, you have how people write an e-mail, you have how people write in DMs — it’s all different because it’s based on that medium. And what Jack created with Twitter was very interesting because it didn’t exist before: it was the first time you had people able to send little text messages publicly — broadcast text messages, in a way — and allow this stream of conscious that’s a little bit less constrained, but also super-constraining in size so you can consume a lot more. And the reason why I’m bringing that up is because: of all the platforms out there, Twitter was the only one that you could do this — that you could broadcast something small in text. So: super-high signal, even though I’m walking my dog or whatever — the signal to people differs. And Mastodon is this massive bloat: the software is crap, the protocol is crap — I’m sorry I don’t want to be rude to the original developers, I’m sure they put a lot of effort into it, but — it’s essentially just remaking centralization to feudalism. So now, instead of having one lord — now it’s Lord Elon — you have whoever is running the Mastodon instance will often be the person who rules over a few thousand people. I bumped into that issue, for example: there were some people that I’m not interested in hosting their content — I’m not gonna block you from talking to my instance but I don’t want to host you — and people didn’t understand quite how it worked so I got in some hot water. Anyways, the point is: Mastodon just recreated the same stuff: my instance is blocked by every other instance on Mastodon Universe because we don’t do content moderation — I wanted people to be free to talk. And I’m hosting their data — I don’t want to do that. I don’t want to have people’s stuff. I don’t want to have their identity. I want people to have their own. I actually plan on closing Bitcoin Hackers — sorry bitcoinhackers.org! So that’s where I am. Jack, how do you feel about this?
Jack Dorsey [10:40]: I think you guys know, but there’s never any one single thing that forces this to be adopted or not — it’s always a multivariable problem. But if I were forced to point to one thing, it would be the simplicity: it is just such a simple spec, a simple idea, and you pair that with the humility of the approach — Fiatjaf’s original description of, This might have a chance of working, sets the tone for everything. And because of that simplicity, it’s just so easy to start coding for it and start playing with it and actually feel it. In the early days of Twitter, we had this ultra-open API, and that was both a positive and a negative: it was a positive in that anyone — 14-year-old, 13-year-old — could take the API, understand it, keep the majority of it in their head, and start programming. And then they could actually send a message via SMS that would vibrate their phone — so there was this real tangibility of, This thing is alive! And it was a negative because it was way too open — we had no idea what we were doing in terms of APIs — and people went crazy. We didn’t rate limit, we were built on Ruby on Rails — which was the original sin — and we went down all the time. So what’s different about this to me is: you’ve maintained that simplicity, you’ve maintained this desire for developers to contribute and feel something almost instantaneously — especially over this past month — and I’m sure it felt that way earlier as well. But now that you have more people, it just feels deeper and deeper and deeper, and the momentum around development will continue to grow — because the protocol spec is just so simple — and at the same time so open, so anyone can go to GitHub, see the source, see how it works, tinker a little bit, fix this, add this, and see it in front of people, and people give feedback on the protocol itself, which just again creates this incredible momentum that pushes it further and further and further. And the speed at which people are seeing change I think is somewhat addictive. And the knowledge that, if I don’t like this change I can just write my own or fork my own client and do it in the way that I wish is even more compelling.
NVK: Fiatjaf I think you were going to add a comment there?
Fiatjaf [13:25]: Yeah I was going to say about Mastodon that I think it was created as a light thing that mimicked Twitter, and then they decided after the fact to add federations.
NVK: I see. That makes sense — it never felt like it was glued to everything else. I think to Jack’s point there: protocols that are too opinionated are less likely to find other people that want to use it. Because people have their own opinions, and if it’s too opinionated it’s going to interfere with their opinions, so nobody ends up using it. That is all fine when you have, say, corporations that enter a consortium or something and then they want to have it all specced out. But when you’re building something that is truly decentralized and you don’t need permission to come in, I think the value really is on that simplicity. And the model of this is for people to have an identity based on a key pair, which to me was an absolutely beautiful design choice — very similar to Bitcoin. So you have this pseudonymous system where people can publish something — in the case of Nostr, as an event — and then everybody else in the world, in the universe that’s connected to you somehow can see that. Provided that it’s not encrypted content — but they can still see the event.
William Casarin [14:44]: To Jack’s point: this idea of simplicity — I think that’s underestimated sometimes, just in terms of how quickly developers can get on board exactly like what Jack was saying. I have studied a lot of protocols — and it’s fun to go through large protocol documents — but a lot of the time they’re actually quite hard to interact with. If you look at a lot of the standard internet protocols, they’re actually quite large and complicated — even just the WebSockets protocol. So when I first found Nostr, it was just so refreshing — and the fact that I was able to build a client in an afternoon — that was such a powerful experience for me personally. I’m like, Wow! So this idea of simple protocols is something that hasn’t even really been fully explored in terms of other types of systems — like peer-to-peer systems and things that. Maybe we should be making our protocols simpler so that they’re adopted quicker? Maybe one big reason why people aren’t using protocols as much is that they’re too complicated, and we need to be thinking about simplicity in terms of driving adoption?
NVK: Bitcoin has a lot of that, right? Bitcoin is super-complicated in the consensus level, but in the transaction level it’s surprisingly simple, especially if you’re using old pre-HD [hierarchical deterministic] wallets — so pre-BIP32 — it was super-easy to create a key pair and construct a transaction and send it out. Same for receiving. I feel like I really see how Nostr and Bitcoin overlap in that sense, and I guess to me that’s why I was drawn to it. I mean, @DocHex — my business partner — goes and builds a relay from scratch just following each NIP over the break. It is amazing to be able to do that without losing yourself in the complication! That was super-cool.
William Casarin: Yeah and building a relay is actually one of the harder things to do!
Fiatjaf [16:33]: BitTorrent is also a very simple protocol — and you can see that there is a surprisingly huge amount of BitTorrent clients, even though most people use the same three or four. There are a lot of libraries and clients for BitTorrent, and what it does is also very simple.
NVK: The torrent stuff started to work very well once they added magnets. That was the original sin, was needing a centralized point of coordination — that’s why The Pirate Bay still to this day is trying to hide.
William Casarin [17:11]: Should we have a Nostr DHT is what you’re saying?
Fiatjaf [17:16]: I don’t think the DHT works. The magnets generally have a tracker ID — a tracker — and that’s when it works well. The DHT sometimes works, but in my experience not very often.
NVK: ZeroNet was a project to do that. I don’t know if any of you guys heard about this project called Twister? It was a guy from Brazil maybe in 2014 that tried to do something similar to Nostr in concept, but it was more complicated and it disappeared — it was a Twitter clone project.
Fiatjaf [17:52]: I’ve read about it multiple times in my life but I don’t remember anything.
NVK: It was quite a while ago. Okay so let’s start picking apart and breaking this stuff down. I had this aha moment a couple weeks ago: I had my key many months ago and then I lost that key, and then I was watching Nostr’s Telegram but I wasn’t participating. And then I came back to it and I finally had my aha moment with Nostr’s topology — and I think this really confuses people: Nostr is not a server-client model in the way that most people think. It’s a many-to-many: you can have your client or your multiple clients — your multiple IDs — talking to many different relays. Some of these relays may overlap — so you may have data duplication to a certain extent, which is nice. Data retention is also undefined, and you can also have relays talking to relays — not so much now, but in the future we will definitely have that. Do you guys want to comment a little bit on the topology of Nostr? Because I think we really need to explain to people how this works.
Fiatjaf [19:11]: I don’t think we even agree on how it’s supposed to work in practice! I think the clients currently connect to a bunch of sometimes fixed set of relays, and they expect to find everything in these — I don’t think that’s how it’s going to work in the future. That’s not how it’s supposed to work, but something like that.
William Casarin [19:33]: Yeah this is something me and Fiatjaf were talking about on the GitHub repo because we were talking about the repost spec and how it’s such a mess right now. I was like, Hey something Damus does which is really dumb is it actually encodes the event into the boost so that no matter if your client is not connected to the relay where that post came from, it can at least still display it. So it opened up discussion about: Should the Damus client be opening up random connections or relays and trying to fetch stuff and trying to find stuff? It’s starting to look a lot more like a peer-to-peer system in that sense where you’re opening up connections all the time to find stuff. There’s many different ways you can do it — right now Damus connects to that fixed set of relays, and it really relies a lot on looking at people’s profiles and connecting to those relays. And just having your relays on your profile is really helpful for clients so you can see, Oh maybe I should be connecting to those relays as well. Or if they weren’t on your profile, maybe it would be hard to know which relays they’re connected to. So yeah we’re still figuring it all out because it’s confusing — Fiatjaf has one vision, Damus is doing its own thing, but maybe it’ll move slowly toward more Fiatjaf’s vision where I’m just connecting to relays and finding stuff. But yeah I think there’s many possible ways to do it.
NVK: I think it really is an amazing property of this protocol that you can make these choices on the go — we’re not stuck on an early spec of the protocol that prevents us from changing how this topology works. And at least in my view, it’s going to be different for different applications of Nostr: maybe you have an ultra-censorship resistant Pastebin application like the Nosbin thing where you don’t need a lot of communication back and forth between the readers and the author. Maybe you have many many relays versus applications where you have a lot of back-and-forth and maybe has less relays. How we choose to use the client for the purpose will, in my view at least, dictate a lot of this. We actually have a NIP coming that’s going to ruffle some feathers.
Fiatjaf: What? What NIP?
NVK [21:46]: We have a draft coming for some ideas on relay-related stuff, because that’s still quite empty. We built a relay just to test the relays, but I won’t get into it — it’s not mine. It’s gonna be great!
Fiatjaf [22:02]: I’ve been wasting a lot of time answering and commenting on NIP proposals and people asking and people proposing stuff — these things are consuming me.
William Casarin [22:14]: I used to be able to keep up with that stuff, and now I can’t even follow the repo anymore. There’s like a new one every day — it’s insane.
NVK: That’s part of the deal, right? When you’re a project maintainer, things can consume your life. At least this is not like Bitcoin, where you’re immediately dealing with money — there’s a lot less room for mistakes and allows less room for experimentation — so there’s a lot more room for people to play around and go their own way, too.
William Casarin [22:41]: I think it’ll be sad when you go to the NIPs repo and then you see a thousand NIPs, because then it looks really complicated. I do miss the days when it was 10 NIPs or so or whatever it was.
NVK [22:52]: Jack you’re funding another couple of — at least that I know — another couple of interesting attempts at creating decentralized social media: you started one at Twitter and then you have decentralized ID stuff. You have a bunch of little projects — we don’t have to get into them because I don’t want to dive into comparing them specifically, but — I want to understand, from somebody who has been playing with the Internet for as long as you have, what is the bigger goal for this stuff? What were you trying to achieve by funding these projects, and also helping fund some of the Nostr work as well?
Jack Dorsey [23:38]: Yeah it goes back to what you said earlier about the feeling of the early Internet, to me. I was in St. Louis: St. Louis had the academia, military industrial complex there with McDonnell Douglas and Washington University, and we had early access to the Internet through BBSes. And I discovered Usenet really early, and IRC, and it led me to the Linux source code when it was just starting. It was just so incredible how it worked and how it failed and how it failed over — the amount of redundancy and just excitement to build for it. And then in the 90’s, as more and more people got onto the Internet, one of the problems that a lot of companies tried to solve was the discovery problem: IRC and Usenet and SMTP e-mail and the web were all inherently fairly decentralized as long as you could host a server and find what you were looking for. Google comes along — and Facebook and Twitter — and centralizes discovery. But unfortunately the centralization of discovery also meant centralization of data ownership and monetizing that — because they’re companies — which led to the ad model and putting almost all the eggs in that one bucket, without doing a proper exploration of other models that could work. And those incentives — of course then going public and having shareholders — created a particular direction in the momentum that was very difficult to get out of. That was definitely my experience at Twitter: we built something that was entirely — and still is entirely — dependent upon brand advertising. And you’re in a class of companies where you are seen as this — by advertisers — their budget is really seeing you as a nice-to-have but not necessary, because you’re so small relative to the giants. And that puts a lot of pressure on how you develop, what systems you build, what content is on the network. And to me it really all came back to the fact that we were both the client and the protocol and the hosting of all of it, which put a single point of failure in terms of how people used it — and the single point of failure was ultimately the company. So in retrospect: when we were just an API and we were a fun little project inside of another company — those were the glory days. That’s when anything went and we didn’t have any external pressures to grow or to monetize — we were just having fun. And [the Twitter API] just reminded me of the only way this can truly be done in a way that serves all the people without all of the surface area that centralization provides 1) for advertisers telling you what they want and what they want you to do, and 2) for governments telling you what they want and what they want you to do, or what data they want for particular users that they don’t like or want more information around — Turkey was a great example the government that always wanted information about particular journalists that were critiquing them. There’s hundreds of other examples of that. But when Bitcoin came along, it looked — from the white paper — that at least some of the discovery issues that we were trying to solve for in a centralized way could be rethought, and it could be done in a way that was actually open, that was actually accessible. Obviously there’s much more accessibility we need to build on top of Bitcoin and through Bitcoin, but at least it showed a different path. And in 2009, when I was starting Square — and Bitcoin was coming out — it just really changed the way I thought about everything: it reminded me of the earlier Internet, it reminded me of why I was here and what I loved, and then it was just a matter of going against this incredible momentum that a public company incentive provides. So I think all the lessons that we learned as a company, we now have — not the technology, but — the understanding as to how to do it a little bit more correct and a little bit more right. And I see it in this [Nostr] protocol. But I also think it’s beneficial to try different things: the Web5 work, the DIDs — they’re solving identity problems. I think there’s some really good solutions that I’m excited about. I don’t expect it to be usable for everything. Bluesky is focused entirely on the social network use case, and they’re doing a different approach than what we experience here, which is: put a spec for a protocol out — it’s not necessarily as simple, but there’s a lot of thought that went into it — and building a client in a less public way, using that as a reference, and then hoping that inspires more developer activity. So I think these A/B tests for development models are generally good, and over time I don’t think they’re necessarily competitive. The reason we started Bluesky Twitter is: ultimately, we’re only as good as the content we have access to. And if we help create a protocol that people actually want to use and put stuff on — and we can get access to that content — we get better, we get stronger because of it. So I don’t see Nostr as competitive to Twitter — it would be my hope that [Twitter] becomes a client. And it would be a great thing strategically, because they have access to more content — if they continue with an ad-based model, that’s critical. But even if they explore other models such as commerce, such as subscription, it’s also meaningful that you just have a bigger network. And it’s really the presentation layer that Twitter should sit at — it should not own all three of those things: the hosting and also the protocol layer. So [Nostr] solves for all of those things in such an elegant way. And to me, my heart is more with this approach: super-dead simple, make it public right away — I love love love that Fiatjaf made it public domain. And just: what’s the velocity that a developer has from going from reading that document, being able to keep it in their head enough that they can replicate what William did, and then in an afternoon actually feel something for themselves to a point where they want to share with someone else so that they can feel the same thing — that to me is the magic path, but I think it’s important that we have different models as well.
NVK: Well different, people different applications — there’s going to be a lot of stuff out there. Fiatjaf, do you want to comment?
Fiatjaf: What do you think about Farcaster, Jack? Because people have been comparing Nostr with Farcaster a lot. Have you ever looked into that?
Jack Dorsey [31:24]: I’m not that familiar with it, actually — I’ve tried to find as much as I can. But what really struck me about what you did was: first that initial paragraph and then listing out the problems to be solved, and the comparisons you did at the end of the doc — Scuttlebutt was very interesting, but heavy in implementation. This work took all of those learnings — it seems to me — and just took the essence of it and put it out there, with the desire of: the problems that we encounter in the future will be a matter of iteration iteration iteration and new ideas. And as William said, It’s beautiful! The inception and just how it’s evolved in a short amount of time has been really spectacular to me, and to see.
NVK [32:22]: Yeah. I think this is a good point to get into incentives. In Bitcoin, the incentive is for you to participate as a good actor so that you don’t break somebody else’s bags, because if you break their bags, you break your bags — you have this mutually assured destruction, Mexican standoff-style of network. Everybody keeps each other honest and we all get to participate in that. But because it’s money, a lot of people who are interested in that topic come in — and it’s a lot easier to monetize, really. I was very surprised how reasonably easy it is to start finding income in Bitcoin as you build Bitcoin things 1) because people put more value into things that make their Bitcoin secure — this is not a nice to have, this a must-have — and they understand the value proposition very fast, and 2) because some companies are moving money around — they can take a cut. There’s many ways to monetize that — you don’t have to think about ads or anything like that in Bitcoin. And then you also have the Number Go Up, so people who have been in the space for long held a little bit so they managed to have their pile be worth a little bit more and they can self-sustain and work on Bitcoin open or closed or whatever, but people can find a way to find a living to work in Bitcoin — the incentives are much better. What was very cool to me immediately with Nostr was the fact that there was no token — we’re in the age of tokens, which is an absolutely disgusting thing — and the incentives: because there’s no centralization, there’s no toll booth, we could find a way for people to start building things. And these incentives actually go against you — trying to force too much onto the user to pay, or it seems that way because you can just move somewhere else. At least in my view, you can do, say, pay to relay in the future — there’s a myriad of ways, and you should confine monetization and find better incentives. How do you guys see the incentives around building on Nostr versus old, centralized protocols? Or versus Bitcoin, really? Go ahead, William.
William Casarin [34:50]: It would be curious to see what would happen to this protocol in a world where Bitcoin didn’t exist, because spam would be a huge issue — I’m not sure if there’s an obvious way to solve that without some type of frictionless monetary system that you can easily tap into and use for spam prevention and things that. But honestly, what’s been the most mind-blowing to me — even to something as simple as adding that Lightning tip button to the profiles and seeing the number of people who are donating, contributing to the development of Damus just for adding that button, and I didn’t even integrate that much Lightning stuff, and I’m eventually gonna have listing tips and things that, but — the amount of people who are just contributing and streaming sats into my node at home was such a profound thing, I don’t think it really clicked until there was a social network around Bitcoin. That is going to be insane — the network effects of everyone using Lightning within the system. I think Lightning and Nostr are very integrally linked and we’re just starting to see that play out at the very early innings, but I don’t even know where this is going to go.
Fiatjaf [35:54]: I don’t think Nostr is so deeply tied to Bitcoin like that — I think you could do spam prevention with other forms of whitelisting people by maybe requiring an entry pass or reCAPTCHA on your website or other forms of registration or phone numbers, etc. Also you could essentially pay via credit cards and so on. I actually don’t very much that it’s so tied to Bitcoin, because people that don’t like Bitcoin, they look at Nostr and say, Oh this is a Bitcoin thing so I’m not going to touch that. I want them to come in without thinking it’s a Bitcoin thing, and when they’re inside they start to see Bitcoin stuff in there so they convert to Bitcoin — I think that would be better.
William Casarin: We should have No-Bitcoin relays or something.
NVK [36:46]: You could say I’m a little biased with the Bitcoin thing, but I personally don’t see how Nostr wins without Bitcoin, because you end up just centralizing everything again and you end up just having KYC everything, because that’s how you find payments. It is possible that Nostr can exist without Bitcoin — no doubt — but I don’t think it would be the permissionless Nostr that I would like to see. Anybody who has touched payments as a business or whatever can understand that as soon as you start taking payments, you are now fully capturable. And that becomes a problem if you want to build a network where people can exist, interact, express themselves, sell stuff in it, do whatever they want if, say, relays are reliant on KYC’ed payments. You wouldn’t even be able to accept customers from some countries that are not allowed to do business with your country — it’s the recapturization of everything. I think it’s just so cool that the first interaction you can have with a relay could be a 402: you could have a Payment Required and it could be a couple sats. It could be sats that are, for example, subsidized by Damus as a client, that’s charging a person a fee per month to have a Damus premium or whatever. That customer per se that just came from the App Store — that knows nothing about Bitcoin, knows nothing about Lightning — can still be paying indirectly for subsidizing the relays that are taking the pounding from a Damus data stream. I find it’s just super-important to understand that we need money: without money you can’t build the coolest things — you can start the coolest things, but it’s very hard to scale the coolest things without real capital, because there’s physical costs like servers, time, devs, and everything. So I saw that very organic growth — being in Bitcoin for this long — just this spontaneous order happen in Bitcoin, on how people self-organize to monetize things and pay for development so people could have an income. Anyways, I found that pretty cool.
Jack Dorsey [39:25]: If Bitcoin existed when we started Twitter, we would not have to go down the ad model path — it’s as simple as that. So to me, what it did for the Internet — not society, but what it did for the Internet — is: it just showed another way and it showed another path that is not subsidized by an ad model. There are ad models that can work — the classified ad model at a more localized scale is brilliant, and I think Bitcoin is going to enable that even more — but the fact that this protocol and the clients that build on top of it have multiple options and multiple choices, and it creates more of a market-based winner approach instead of having to compete with the giants in a way that they are firmly entrenched in both regulation and also of course usage, is really really important. So I think that tangibility I was speaking about earlier — where in early Twitter days when we were more SMS-based you could feel a tweet in your pocket because your phone would buzz — to me that’s effectively what paying Lightning invoices on this protocol feels like right now. You put an invoice out there and suddenly you get a notification that someone paid it and it just feels amazing — it feels alive, it feels tangible — and when people can feel the thing, it just lights up the imagination. When it’s all abstraction — which, unfortunately, is to me the case with what we see with Mastodon and Matrix and a bunch of other things — the marketing to people is it’s “decentralized” and it’s not about necessarily the utility and the use case. And being able to say, You can instantly see how this global money transmission protocol really works and really feels with strangers, is pretty incredible. I think it just keeps people thinking in a different way. And it’s probably not the end-all, be-all, but the number of models and the new incentives that can be created because this exists — because Bitcoin exists and we can pair it with Nostr — is the most important thing.
NVK [41:53]: The Lightning thing never clicked for me that much — because I’m a base layer boomer — until I started doing this podcast thing with the Podcast 2.0 where I’m seeing sats being streamed to us. It is super-cool: you get that very special connection with somebody who’s a listener because they didn’t just go on your YouTube channel or whatever and put a comment — they’re streaming sats and they’re sending what they call boostagrams, so you’re getting these messages with a financial token. Tips are ridiculous when you think about it economically speaking, but they’re not empty like a Like — a Like is truly empty. And when somebody parts away with their sats, there is a special connection there. And I’m seeing everybody posting these Lightning invoices since Damus started to take off — it’s just so cool! People are paying each other, buying each other coffees and things — it really has a different feel to it than just people yapping around about how cool the protocol is.
William Casarin: Hey — a Shaka is not empty, all right?
NVK [3:07]: That’s right — it really is that feel: the Shaka is not empty. So why don’t we go around a few times here and we talk about just possible use cases? A good way I find to explore new things I don’t understand is: try to think of all the crazy business ideas I could have for it, and try to fit that new technology or new tool into a whole slew of different boxes and see if it works — that creates for me a nice mental model of all the possibilities and the aspects of the technology. We talked about how Nostr is great for social media — it is a fantastic Twitter replacement — I mean, you guys managed to even bring over the inventor of Twitter here, so that one is settled! So for example: broadcast is the one of the main aspects, properties of this protocol — it’s a broadcast system, in a way. But it’s a two-way broadcast system, so you can have marketplaces where you can post a Note that has something for sale and then people can respond to it, people can DM you, people can maybe bid for it — I’d love to see a NIP for bids. Hopefully that’s not a client thing — it’s a protocol thing. Maybe it’s stupid? You could have the Pastebin thing that we talked about with the Nosbin project where you can post on a Pastebin. And that’s fairly censorship resistant — nobody’s gonna take that bin down. What do you guys think? Do you guys want to give some awesome examples of things you can see exist in this protocol?
William Casarin [44:54]: I’ve been making a lot of posts on Nostr about potential ideas in the future — the stuff I want to build at least in Damus, or maybe even in other clients. The commerce thing is huge because the minute we have Lightning tipping where we can list the tips — the people who are buying things within the app — then you can start to open up really interesting direct-to-consumer use cases where, if you’re a clothing store, you can sell stuff in a post and you get instant feedback. If you imagine someone randomly buying something and now it shows up on the post and then they get a badge, and things like that. So there’s a lot of really cool direct-to-consumer stuff, but that’s maybe just within commerce. In terms of more at the protocol level, I still think — and no one’s actually doing this, as far as I know, but — we have this system where we can connect to multiple relays, but there’s no reason why you couldn’t connect to just one relay which is like your company relay that manages communication. And so you don’t need the relay pool in that instance, but you get to utilize all these other clients and tools to interact with your company Slack or something. You can totally imagine a Slack for your own company that’s built on Nostr and you only connect to one relay — there’s so many use cases like that that we haven’t even really looked at yet.
NVK [45:59]: William, I was thinking the other day: once there is some crazy amount of traffic coming in, clients may not be able to filter anymore the global feed that you’re connected to. So I was thinking it’s very likely you could see people running a buffer relay or something that they control — or a client that they really like controls for them — where they can do some pretty insane advanced filters and send to their client a much more contained collection of filters for them to watch.
William Casarin [46:33]: I’m very wary of filtering, because I would always want that to be a client choice and a user choice — I get very uncomfortable when relays are making those decisions for me, but that’s just me personally.
Fiatjaf [46:47]: Well I like the idea of relays filtering! The way I see it — this is just the social stuff — I see it like: you connect into a bunch of open relays full of spam to talk to random people, different people, but for your global feed or for the type of stuff like showing the replies to your posts that could be spam, you want to only accept stuff from relays that are filtered or have some way of preventing spam on your side. Other ideas: I’ve had a lot of ideas — probably most of them are trash — but I always think of having specific clients dedicated to these niches. I’ve been meaning to write a commerce client and a way to have a store and publish stuff there, but I couldn’t manage to do anything yet. But then when I saw William’s idea of sending stuff in a post, that maybe it’s a simpler way, it’s a easier way, and a way that will catch on better if you have stuff [for sale] — mix it in with the social layer, social clients.
NVK [48:14]: What if we had a kind — maybe this is just dumb because you don’t want to add this to the protocol level, but — it would be nice if the Notes themselves had some description of the intention, because I don’t want to see NVK’s Good Morning post coming into my classifieds feed, even though I may be sharing some relays that are relaying that kinds of information.
Fiatjaf [48:42]: That’s why we have the kind stuff — the [kind] numbers.
NVK: Yeah exactly.
William Casarin [48:47]: We could be much better at this. For instance: whenever you see a [Lightning] invoice get pasted into a Note — maybe you don’t actually create a kind 1 Note, but you create a kind 9735 Note or something, because right now a lot of people complain that there’s too much invoice spam in the client — if it just created a different kind, you can just filter out all the invoices from your feed automatically, which would be cool.
NVK [49:06]: Can we create a kind 402? So if you send that, that just means you have sent payment and maybe there’s a Lightning invoice in it? Maybe it’d have to be fully baked — this is not fully baked. Okay do you have some cool ideas you want to see existing on this? Even if it’s stupid now and we’re gonna find that it’s completely impossible?
Jack Dorsey [49:36]: It feels we’re gonna be in a phase for a little bit of time where we’re replicating what already exists like the Slack clients, Reddit, Twitter, obviously, and there’s more running away from a thing to this instead of running to it, for a particular new use case. And to me, the new use cases always come from mashups of different things. Just a simple example: it might be the fact that commerce and money transmission becomes so native to social, but that might be the utility that really takes us to the next level and gets it broad adoption. But I think, for a little while, you’ll see a bunch of projects that are just replicating what exists — and it’s something I want! I hate Slack — there’s a number of issues that it presents — but the use case is still important. To be able to see that on this protocol, and then to see how all these things interact with one another — that to me is where the truly unique utilities is ultimately going to come from. It’s probably not going to be necessarily designed proactively — it’ll be discovered, and something that someone just trips upon and decides to go down the rabbit hole and build. So to answer your question: I don’t know. All that’s in my mind right now are just things that we might replicate in a better way, but to me that’s not going to be the reason that people actually come to this protocol, and to the respective clients — there’s going to be some new utility, some new use case that is so undeniable that I have to run to it and I just have to be a part of it — that, in addition to the network effects and probably the utility effect. What’s interesting about this protocol is it’s not just focused on social media — it at least sets the stage for an ecosystem of different use cases working together. That’s going to be the power, is that you have one understanding of how to interact with this — it’s a very very simple one — and with that one simple understanding of the public/private key pair, you have this entirely new universe of things that suddenly work together. I just don’t think we’ve experienced that as much before. So if there were one use case, it would be this ecosystem mentality of, These things just work! I didn’t really have to do much at all: I can forget everything I know about the old password model and really get a ton of utility in my own way, my own choices, just by generating a key pair once — a whole new universe. I can have multiple identities, multiple key pairs for different things, and potentially in the future intermix them. But the power of it is: it’s a new universe — it’s not one thing.
NVK [52:52]: When you think about it, most new technology happens that way — in the beginning of Bitcoin, everybody was trying to recreate casinos and trying to make payments, because that’s what we know. And then 10 years later, people are creating ways for you to fund podcasts with Lightning and streaming money — streaming money was completely impossible before Bitcoin. So it is fascinating how we do recreate things and experiment by trying to remake what we know.
Fiatjaf [53:25]: But I think one point Jack said that’s important is the general interoperability between different use cases: you could have an e-commerce use case or Wikipedia use case, but you could somehow tie that to the social stuff, for example — the most dumb way would be to put comments on everything, but other things can happen.
Jack Dorsey: Exactly.
NVK [53:53]: For sure. That’s the thing that is so cool: with Bitcoin you have this common money, this common value, but money is pointless if you don’t have communication — two people need to communicate in order to trade. And what you invented here Fiatjaf is, in my view at least: you now have a common carrier for that coordination and communication — these things are so intertwined and so important and they’re the fabric of everything a society does. It’s like: talk to each other and figure out how to trade. Anyways, it’s just super-cool.
William Casarin [54:31]: I just want to say: I don’t know if I would actually be interested in even building Damus if it didn’t have something new to offer — that’s where the Lightning part is very important to Damus. We like to say, Oh we don’t want to push people away [by pushing Lightning] — but I just don’t think Damus would be that interesting if it was just another Twitter clone. So I just feel like integrating Lightning into a social network is the killer app I think, at least on the Damus side — otherwise it’s just a shittier version of Twitter, I guess.
Fiatjaf [55:00]: I’m not trying to say you shouldn’t have Lightning stuff on Damus — I don’t know what I was trying to say, but that was not it.
NVK [55:08]: I think your sentiment is correct, as somebody who invented the thing: you don’t want to pigeonhole the thing into Bitcoin — this is a Bitcoiner’s job to do! We’re gonna keep on pigeonholing Nostr into Bitcoin far in the future, but I totally understand that you created a thing that’s independent of Bitcoin — am I getting this wrong?
Fiatjaf [55:31]: Yeah I think it’s there in what you’re trying to say. Well I want Bitcoin to succeed much more than I want Nostr to succeed, so I would sacrifice Nostr tomorrow if you could get Bitcoin global adoption.
NVK [55:48]: I mean that’s very kind of you, but because your invention is independent of you in the way you created it — it is that special — we can say, Okay bye Fiatjaf, we’re going to continue on in this Nostr thing. It’s very similar: Satoshi got pushed out! Most people think that Satoshi is like a Jesus-type character — which I think is important on how the religion of Bitcoin is disseminated — this is a different conversation, but I think the true testament to an incredible incredible new invention is the fact that anybody who participates in his invention can be completely pushed aside and the invention continues on its own. What he did is beautiful in that way.
Fiatjaf [56:36]: Okay okay, I’ll leave.
NVK [56:43]: So guys, I’m good in a way, but do you guys want to give a few more ideas? Because one thing I’ve noticed being in technology for this long is that a lot of people — they may be great builders, they may be great business-starters, but — people oftentimes don’t have broad imagination about a new technology. I love to seed people’s imaginations, if we can, with ideas — even if they’re not baked concepts, but — things that would be really cool that you wish existed in Nostr today. We don’t have to work it through, but you guys were mentioning the commerce stuff — for example: PubNub is an incredible service. It is push notifications, essentially. In my view this could be completely replaced by Nostr — boom: you just have a whole SaaS industry that you can disrupt with having the system available to projects that don’t have a very specific thing that they need from PubNub, but they were just trying to push. You can now have 10,000 concurrent users on a chat room being pushed stuff from each other without needing the centralized entity to make it happen. Is there other very obvious things to you that come to mind?
William Casarin [58:11]: What I’ve been thinking about is: what other things could we be integrating into Nostr? For instance: does it make sense for Nostr clients to have more peer-to-peer functionality so that you can spin up a video chat? So, not necessarily building on Nostr directly, but: What are other things we can integrate to improve the experience? How can other protocols improve the Nostr experience? Another thing is: How can we apply the Nostr model to other protocols? For instance: Nostr is mainly for text, and a lot of confusion I see of people joining on the network is, How do I post an image to Nostr? How do I upload my cat photos? Or upload my 6 GB file to Nostr? I’m like, Well no — this is not what this is for! Would it make sense to look at the Nostr model and apply it to maybe a binary protocol, but using a similar topology? That’s an interesting idea that would be cool for people to explore. This idea that we need to put everything on Nostr might be misguided.
Fiatjaf [59:13]: I agree with that — we shouldn’t put everything on Nostr.
NVK: So no Base64 in line?
Fiatjaf [59:20]: Well people are doing that — I wish relays prevented people from doing that so they could give up. A guy made a file system over Nostr or something like that.
NVK [59:33]: Pre-Bitcoin, we built a dispatch software for a very very big taxi company — one of the biggest ones in the world — and being able to do that push notification and push the order into all these clients, I can see all these industries being completely replaced tomorrow. Because this protocol is so simple, so usable, it is literally a 1-week project for somebody to replace some of these industries that were built around push. I guess I’m just very stuck on that — I want to see pagers come back.
Fiatjaf [1:00:17]: The problem with push stuff is that you have to keep a connection open. I don’t know how Android —
NVK [1:00:23]: No, you can push once and essentially you can have a client do some pulling just to remain alive, and then you can have the client do that pull — it’s gonna look like a push, but it’s actually a client doing a pull. It’s just that that was not very possible before Nostr to do in any reasonable, scalable way.
Jack Dorsey [1:00:43]: The big class of use case that selfishly I’m interested in — that I think this protocol will eventually help address — is: Block bought Tidal just over a year ago, and streaming has a bunch of issues that need to be rethought. But the specific use case that I’m thinking deeply about in intersection with this protocol is the musician side — the obvious one is around payment for artists: the current system is just criminal in the way the artists get paid and the way that they enter into contracts, and all their life’s work is taken from them without any sort of financial education to put them in a better position. But the collaboration side for artists — especially new artists — is just wild, and it’s a complete gap in the world. There are very little tools — there’s so many artists that collaborate over iMessage, and a lot of them over TikTok or Instagram DM — and that problem itself has I think a lot of intersection with what this protocol enables, because you do have this fluidness between a social layer and a money transmission layer, and eventually a commerce layer. And for a new artist to really get established, they need to have some sort of confidence that they’ll be able to be paid for their work, and they can leave whatever job that they’re currently in because they can see the horizon of not only getting listeners, but actually those listeners exchanging value for their for their value. So this concept of exchange of value for musicians and how musicians work together — and you can go more broad into artists obviously, but — music is such a fundamental part of life, and it’s been so long that it’s really been critically looked at and changed and evolved. And I think a lot of it obviously has to do with the concentration of the [music] labels, but there are just no options for the musicians right now — there’s no new thinking. And if this protocol does nothing else in concert with Bitcoin, it makes people think a different way, and it just opens the door to many more possibilities — some of them will work, some of them won’t, but it at least presents a challenge to the status quo. And it’s so needed because it is so locked in at the moment.
NVK [1:03:21]: With the Podcast 2.0 thing, we finally see a way to do this: you have the streaming and then you can do the sats splits — that’s what they call it, the cool kids call it the splits — and then you can have the sats come into your pod. I’m not set up for this, but you can actually do per-episode splits with other guests that are there that are set up for it — there is no reason why music couldn’t work that way. The same way movies — and labels, you have the people who worked on a record — you could have the artists set up the splits right from there, and then people just get paid as things get consumed. But then you have the issue — and this is the issue that I’ve always had with RSS — RSS needs to be semi-centralized. There is no good way of doing truly truly distributed RSS. And Nostr comes in and it’s like, Hey! Holy crap — you could do this Podcast 2.0 thing, the Value for Value thing, but instead of using RSS you could actually do the distribution with the Nostr protocol and be unkillable. It’s completely censorship resistant if you have this in enough relays, and this could be distributed in any kinds of platforms, and you immediately have the social layer on top where people can comment and participate — I don’t know, it’s just so crazy cool.
Jack Dorsey: Exactly.
William Casarin [1:04:51]: I was just gonna say: on the label side — because I worked at a record label here in Canada called Monstercat — I honestly think the biggest issue that we’re running into is the fact that everything was so siloed. All the contracts, each label, all the splits — it makes sense a lot of the time that they’re hidden because there’s ghost producers and things that, but the minute you have a system where potentially you could open it up, it just opens up a lot of proprietary stacks. That’s particularly important when you’re trying to do attribution for stems and things like that where you want to be able to pay these people, but a lot of time you just can’t track down the splits properly — that was the biggest issue that I saw at Monstercat was: How do you have an open system? And not only that, it’s a little bit different from RSS because RSS is just an XML document and you’ve gotta sift through it and build specialized tools.
NVK: Hey at least it’s not SOAP.
William Casarin [1:05:40]: But just the ability to query stuff — I want to be able to say, Hey give me all of the splits for this artist, who to associate with a pubkey. Being able to do that with a simple query model that’s universal, and then you can attach comments to it, that itself — that queryability aspect — is super-important for these types of use cases, and it’s not something that you can simply replace by [using] an RSS feed.
NVK [1:05:59]: I mean you’re gonna be able to do that with software — that’s how you could fund software. You could do git over Nostr — you can choose if you want to push the actual commits through Nostr or not, because it’s just text. It could get pretty crazy in size, but you could at least do the meta part of git with Nostr: you could have comments, you could have payments — people using the software now could actually donate to the exact person who made the commit, in scale! This is the absolute crazy part to me, is that you can do all these cool things that are completely pie in the sky dreams that we’ve been hearing the Web3 people masturbate about for the last five years with tokens and crap — but that’s all possible now for real, in scale.
Jack Dorsey [1:06:50]: Yeah the GitHub use case I think would be incredible in itself: being able to get Bitcoin Core off GitHub — and then obviously all these projects on Nostr off GitHub — would be amazing. But when you actually get into thinking about how to fund free software and open source — again, this presents just a new idea and a new model, and there’s a lot of parallels to what William was saying in the music industry. You’re giving this new utility to creators worldwide, whether they create source code or music or stems or jingles or whatever it is — it ultimately comes back to: How do I fund my life such that I don’t have to take another job that distracts away from what I truly want to do? And what Fiatjaf said earlier about the interoperability of all of these things within the ecosystem and in the use cases — it’s a super powerful idea that I think we’ve just seen a glimmer of with the Lightning integration. The other aha moment I had was being able to move from client to client so effortlessly, and imagining being able to move from use case to use case with that same lack of effort and the lack of friction that we’ve already seen in moving from one client to another — that’s just magic to me: you don’t have that anywhere else on the Internet right now.
NVK [1:08:26]: I was going to bring that up: one of the most amazing things in Bitcoin is the fact that — through the key pair system — you’re not tied to any wallet, or the client. You just go to something else and your money just magically shows up there — it’s crazy! Try to change banks? It doesn’t work like that. So now you have that for the information, comms layer, and social is part of that. You can move communication type and layer with just that public key pair — that’s fascinating.
William Casarin [1:09:04]: The thing that underlies all these things is this idea of a simple language for computers — that’s how I view protocols. Imagine if it was a language that you could easily pick up and learn — it’s almost like algebra — there’s simple rules, you can put them together, and then once everyone knows the rules then you can build some insanely powerful things just from these simple constructions. That’s how I view it whether it’s Bitcoin or whether it’s any protocol, or even Nostr — the fact that we all have a common language now to build together, you can build independently but still interoperate because we have that common computer language.
NVK [1:09:41]: Yeah the common bus is important — you can’t build electrical things without a bus. It’s the same for money now with Bitcoin and it’s the same for comms now in Nostr. Those are the building blocks of civilization. Those are the building blocks of human interactions. You need these things.
William Casarin [1:10:02]: In some sense, centralization of these services is corrupting the language itself. If we’re thinking of protocols as a language for computers — and then you take that ability away from computers and individuals to communicate, and they control the language — it’s very much like Big Brother in that sense, but at the protocol level. It’s an interesting way to look at it from that perspective.
NVK [1:10:23]: I mean it’s a very hard a problem to make things distributed and still have a common language. Bitcoin, in my view, was the first thing that ever achieved that properly. And centralized systems, centralized protocols, centralized services are extremely efficient — that is the undeniable gain, is that efficiency. That’s why people use Visa: it is fantastic! 99% of the time it works: you go to the place, you tap your card, boom — it’s done. And until this kind of technology came about and these aha moments and these inventions came around, it was very hard to just try to attack those problems. We had the Byzantine Generals Problem resolved [with Bitcoin]. And now I think with the way Fiatjaf designed Nostr, we have decentralized, distributed comms resolved. I guess we can get into the weeds a little bit of some of the criticisms of Nostr, and some of the things that are unresolved. I think because of the nature of how this thing has been developed, unresolved things can be resolved later, but some things are — in my view, at least — super-necessary in order for people to take this very seriously with their true value. They don’t want to lose their identity, they don’t want to lose their money — that kind of problem. So for example: key rotation and delegation. I love the key pair solution, and there is a NIP for a NIP-26. Fiatjaf, do you want to just talk a little bit about the current state of key delegation, key rotation, and possibly key revocation?
Fiatjaf [1:12:05]: Jesus Christ. So, at first I didn’t give too much attention to these things — I don’t think it’s very important. I don’t know if it’s that serious as a problem. I also don’t think NIP-26 solves it in a perfect way. And I don’t know if people should be expecting NIP-26 to generate a bunch of NIP-26 keys and put one in each app and don’t care anymore if these keys are lost or not. And because it’s optional, I think clients should be okay — it should not support it and not show the linkage between two keys that are using the delegation stuff. I think for the main app that you use for your social stuff or whatever else, you should use your actual key — but I don’t know if this will work. I think that there are solutions that could be done with multisig and dedicated devices that are safer — that just perform signatures. If you’re a normal person and don’t have a lot of online identity reputation, it’s okay to lose your identity sometimes — it happens to normal people all the time: they forget the passwords, they change their phones so they don’t get to recover their account so they create a new account. But for people that have an image, a reputation on an online identity that is valuable, that’s not a good thing to lose it — so for these people, maybe more precautions should be taken.
William Casarin [1:13:52]: I personally think that delegation is probably a lot more popular in the future. It does solve a lot of annoying things such as: you don’t want to put your private key into a browser or you don’t want to put it into some random application. One really good use case is a Twitter-to-Nostr crossposter — you want that app to be able to post on your behalf. Those are really cool use cases where you can assign it a delegation and let that app create those events for you for like the next week or so. I think multisig is really interesting for corporate accounts where you want to make sure that two people sign off on a post, but it’s going to be really hard to make it work if — for each post you do — you have to do some type of multisig thing. I’m not really sure how that’s going to work unless it’s two software keys or something. But I think delegation solves a lot of problems, and I really expect more clients to adopt it.
NVK [1:14:51]: I’m a little biased on this thing — I’ve been making Bitcoin security products for so long, and I see how people lose money and how people get screwed losing their keys. I think identity and vanity — the healthy type of vanity, let’s call it that — is integral to people’s view of themselves in how they interact with the world. That’s why it’s one of the issues that I’m super keen on seeing resolved — I’m not a cryptographer, I just hope somebody smart out there comes up with a wonderful solution. We’ve seen this come and go in Bitcoin a lot: in the early days you had single key pairs — these are pre-derivable keys — and that resolved a lot of privacy issues, they resolved a lot of security issues. And then we saw between P2SH, which was the first multisig thing, and then and then SegWit. I think the evolution of cryptographic ways of doing things really help with this kind of stuff — maybe the bar seems lower now, but as this thing progresses and people have more of their life’s work and their income and everything else, it would become critical for people. But these are all resolvable problems: all you need is people to just sit there and figure it out. It’s not things that we cannot resolve — that’s the cool part.
Fiatjaf [1:16:19]: Do you think? I don’t know. I think these key stuff problems are very hard to solve. I didn’t want to trash NIP-26 — I like NIP-26 — and the example William gave is a good example of how it could be used. I just don’t like the idea of people having a hidden key — they never show, they never use — and that’s their actual key, and then they use other keys for everything. I think that sounds weird — I don’t have an actual reason.
William Casarin: The other keys are always linked to the root though, right?
NVK: I’m curious: Why would you not like people to have a hidden pairing key?
Fiatjaf: I don’t know. I don’t have a good reason. I just think it feels wrong.
NVK: That’s fair. It’s just your intuitions — it’s: I don’t like it.
Fiatjaf: I’ll come up with a reason.
NVK: There you go. Yeah I think there’s going to be many ways of skinning this cat. I was asking you the other day, Why did you choose Taproot Schnorr instead of using ECDSA?
Fiatjaf [1:17:32]: It was more because of the multisig stuff. This is basically the same as the other, but this can do a little more things with the multisig stuff, so why not?
NVK [1:17:43]: Exactly. You have a very decent primitive there with Schnorr. Once MuSig gets finalized and all that stuff, you’re gonna be able to do a lot. And you can de-risk clients, you can have co-signers — it can do all kinds of crap. Maybe people have a personal server on the Internet? Or a key service out there that co-signs things so that they’re not at risk as much? I don’t know — I really don’t know. I just think it’s an issue.
William Casarin: There’s a NIP for you of what you just described — a remote signer — someone’s working on that.
NVK [1:18:18]: There you go. I was sending the NIP-26 to Adam Gibson, who did a lot of the work on JoinMarket, and he’s an actual cryptographer. This is nerd-sniping, because this is total catnip for cryptographers, so I feel like the interest is gonna come and a lot of people who know these problems very intimately can resolve them.
William Casarin [1:18:43]: I’m sure they existed, but the fact that Bitcoin congregated all the best crypto people in one spot — I didn’t know that there were so many good people in this space — and I hope that maybe some of them come over to Nostr and help us out, because I don’t think there’s a lot of good crypto people. They’re all working on Bitcoin and things that, because it’s obviously more important for their types of skills, but we definitely need some help making better encryption specs.
NVK [1:19:12]: It is quite fascinating: there is a lot of stuff that didn’t exist before Bitcoin, even though those crypto primitives were around. For example: deterministic builds for software — this is a Bitcoin-people invention, because the threshold of security needs that happened with Bitcoin is much greater, because there’s no mulligan with Bitcoin. If you screw up, you can’t roll back. So because Nostr is attractive to so many Bitcoiners, I think you’re going to bring those people in.
Jack Dorsey [1:19:41]: I do think that the delegation point is really important, because it at least gives you confidence that you can play around with different parts of this emerging ecosystem. But there’s something refreshing in this time of not being so tied to an identity, and to me that was refreshing in first experiencing the multitude of clients and being able to generate new keys and choose whether to persist them or not. There’s definitely a gap there: there’ll be some interesting use cases that emerge from how fluid the system enables someone to be that we haven’t had access to before, because the “account creation process” was so heavy. And then obviously Bitcoin contributed a lot to this idea and this path, but to now go to other utilities and have that same sort of approach is to me super-refreshing, and I think we’ll see a lot of people playing around with that. And one of the things that makes the Global Feed so magical right now is: you can see something — it’s all serendipity — but you have no idea who this person or bot is, but that one post might be meaningful in a really tangible way. And it just really detaches the idea of this built-up identity to something a lot more atomic and a lot more free and a lot more fluid. So I just think that’s something we don’t want to lose, because it’s not present anywhere else, in my experience.
NVK [1:21:31]: Because this thing is clearly completely distributed, you could have these super-interesting relays that are serving more unorganized, more undefined content. You could have place where people are being a little bit more free, and that you’re gonna end up having your private relays where people are being super-formal — like a company relay. There’s a lot of companies out there that are going to probably not want to have their Nostr communication publicly, so they might just have private relays between whoever works at the company. But it’s immediately compatible with the broader external communication, too. It gives me hope that it’s never going to go away — you’re always going to have Nostr after dark, because that’s kind of the vibe you have now: you’re in a BBS and everybody’s sharing stuff and you have no idea who they are or where they are. I don’t know — it could still be there forever.
William Casarin [1:22:33]: I think the analogy with other aspects of what we see in the real world is in terms of: Okay let’s say there’s a public space and this might be just your park near your house, and anyone go there and scream crazy stuff into the void, and maybe I don’t have to go there. And so the analogy there is: Oh you don’t have to connect to that relay, but maybe there’s a more semi-private mall or something where you can go and have more civilized discussions without people screaming. So I’m really interested to see the breakdown between public relays, semi-private relays, private relays, and having clients be able to quickly switch between different views of that. Because some people don’t want the wild west, people screaming in your face — and being able to switch to a more private view and more civilized discussion with your friends, or with people who maybe paid some money to get in there. There’s a lot of different relay models that I think are yet to be explored.
Fiatjaf: I’m waiting for the relay that bans people that write, GM.
NVK: I don’t know who does that.
William Casarin: I’m surprised it’s not a kind yet.
Fiatjaf: They’re very awful people.
NVK [1:23:43]: So I think the same way I can see how there is a whole new business industry there for people who write AI prompts to generate pictures, you’re going to have a huge interesting both free and paid set of packaged filters that may include which relays you connect to — it could be an art.
William Casarin [1:24:10]: Is this what you’re working on? Is this the controversial thing you were talking about?
NVK [1:24:13]: Ah, no. So the way we experiment with technologies as a company is: we go and we build a thing from scratch. With Bitcoin, we built a Bitcoin explorer back in the day — it was btclook.com. And we sort of understood how Bitcoin worked by building an explorer. So now with Nostr we’re building this — because I had an idea for what I wanted to build on Nostr — I think now it’s semi-stupid, because we work through building something with Nostr. There’s a few ideas we want to play around with, but we wanted to have a better understanding of this — I don’t want to bring too many of my biases into it as we’re building it. Right now we’re doing search.
William Casarin [1:24:57]: Yeah that sounds a lot like my implementations on Damus — they’re all pretty semi-stupid.
NVK [1:25:03]: It’s easy to be stupid with new technology — the electricity comes around and people are like, Oh okay so maybe I could put a lamp in my horse buggy or whatever, instead of trying to make an electric motor.
Jack Dorsey [1:25:18]: I think Fiatjaf’s — I don’t know if this was in your Google Document after the funding or if it was in the original spec, but this idea of micro-apps and more micro-apps being built really removes the barrier for development and puts a lot more emphasis on how these things work together and truly create this ecosystem. And the cost of doing a micro-app is super super low, but the combination of hundreds of micro-apps working together has a high amount of value. And I don’t know if the community is necessarily thinking about it that way just yet, because again we’re running away from things and replicating them, but when that clicks and when we actually see people switching from use case to use case in this simple self-contained, easy approach of this micro-app universe, that’s when I think things get really really interesting, and when you see more of the “mainstream adoption”, is because that to me is the gap in the market that hasn’t been truly filled — especially when you own your data and you own and control your identity and how you participate or how much you participate. So that’s what I’m most excited about.
Fiatjaf [1:26:49]: Yeah that’s a good way to scale work, because when one guy can do a lot of stuff, but he can’t do a big, powerful, clean, pretty, performant client that does everything on its own, but one guy can do a client that does one specific thing very well that doesn’t do any of the other things— I think that’s a very good way to distribute the work.
NVK [1:27:18]: I think that was the Achilles heel of the [web] browser. The browser is in my view one of the biggest sins that ever happened in the Internet: you have this thing with millions and millions of lines of local executed code — and why? Why can’t you have this plethora of different clients that do different things completely differently based on a single unifying protocol? I think that’s why we don’t see a lot of browser competition — it’s impossible to compete with how the bloat of HTML5 happened. It’s very cool — you can do all kinds of cool stuff — but it just further centralized things due to complexity.
William Casarin [1:28:01]: Yeah just going really quick back to this idea of micro-apps: it doesn’t really have to even be a client, per se. One example of this would be just an embed in your browser. One example that I really want to build is those customer chatbots that just pop up — so it could be a small piece of JavaScript that you put on any web page and then now you have a customer chat bot you can link to your relay that you can have one-to-one conversations with on your website. So there’s micro-clients and things like that, but there could be micro-embeds we can start putting into websites. I don’t know if people are thinking about it that way as well.
NVK [1:28:38]: It’s super-simple to build: if you’re building a very restrained client to receive Notes and to send Notes, it is trivial. There’s no reason why all of this stuff can’t exist. Chatbots are awesome.
Fiatjaf: Yeah an embeddable thing like that is a client, technically, I was trying to say.
NVK [1:29:02]: I love the idea of DM bots — we built one that’s not doing all the things they need to do — it’s the operator at brb.io. Why do you need to build the UIs on the web for everything? Why do you need passwords to log in to all this stuff? So if we’re focusing on experimenting with different relay things, when you join our relay, maybe with the paid version of our relay you get a DM and the DM says, Hi! Welcome to brb.io! Please pay this Lightning invoice to continue. You don’t need any other UI — and this is fully private, too, because you have a fully-encrypted DM. The possibilities of these prompts to do very complex interactions is very interesting.
William Casarin [1:29:53]: I love integrating Nostr into my desktop. One of my hobbies is I built my own operating system desktop environment. So being able to write some code that listens to events — that people are sending me Notes — and having a little blip notify pop up. I have a little script that does that, so that’s kind of cool. And let’s say your server is generating events and so now you can use your Nostr relay as an event stream. Other examples are like: every time my Bitcoin node gets a transaction, I send an event to my private relay which I get on onto my [Nostr feed]. So just tight integrations into every computer system is something I don’t know if a lot of people are doing that, but that’s also a really fun and cool use case to automate a lot of stuff that people would normally use a service for.
NVK [1:30:40]: I think that used to be a thing when protocols were simpler and computers were simpler, but the bloat of HTML5 and everything really modern web is so big that it became impossible to build little things that just truly participated as a first-class citizen in the network. And Nostr brings that back: you can really create small things that have an outsize impact in the network.
William Casarin [1:31:06]: I don’t know if we’re going there, but I definitely agree with the fact that the web has become way too insane and bloated. There’s been talks of having a web phone and things that, but I honestly think the web is just way too convoluted to build anything except to staple on top of it. I got into programming to do game development, and I really like that model where you’re just doing raw rendering calls to your GPU. Maybe that’s the model, where we switch away from the web stack altogether and you’re just drawing stuff directly to the to GPU — that was my other client I was working on that I was trying to demo, but maybe one day I’ll get back to it.
Fiatjaf [1:31:42]: I had this amazing idea for a protocol that went nowhere before Nostr that was another way, because the nice thing about the web is that you had the URL that you connect directly to a server. And you could do chatbots on Telegram and stuff, but you’re still using a centralized provider. But if you could connect directly to a server and have an experience with that server that is a protocol that your client talks to — yeah I was trying to make something like that. It was a mix of a chatbot and other UI components that was just drastically simpler than the HTML5 stuff. And why not have a different view? I don’t know how to explain it but you get it.
William Casarin [1:32:39]: I was trying to do this with my metaverse protocol that I was working on, that instead of maybe connecting to a server to get a page, you’d connect to a server and get a virtual space — there’s so many things you could do that are cool like that. We’re really focused on text and documents and stuff in a lot of the way we interact with our apps, but there’s no reason why you couldn’t connect to a server and then have a virtual space where you’re chatting in as well.
NVK [1:32:59]: A small little business idea we had a few years ago that I wanted to deploy — but they were bigger consequences of making that happen — was: you have a lock box for a tax document, and then people pay to see what’s inside. And the issue with that is you’re gonna end up always having takedown notices and it becomes a nightmare — it doesn’t scale or whatever. But now with this you can have Notes out there that are pay to open — and they live. GDPR is completely made obsolete now. How are you gonna tell relays to not have that information exist? Maybe the ones that are near a jurisdiction. But things are gonna become very weird and interesting: it’s going to be hard to enforce copyright because the content is not necessarily deletable. I think we’re in for a paradigm change in how information ownership — and the consequence of that — happen, because of this new protocol.
Fiatjaf [1:34:03]: There’s a lot of people criticizing me on the GitHub saying that I should provide a mathematical proof that the protocol is censorship resistant. And all I have is that Note on the readme saying that there will be a Russian server somewhere!
NVK [1:34:22]: But that’s a completely fair explanation — that is it! There is no other way of doing this so simply, except for the fact that — it’s just like Bitcoin — we count on flag theory and on state actors having different interests.
Fiatjaf [1:34:41]: Yeah that’s how I understood it too, but these people wanted something academic — I don’t know how to provide that, so I can’t.
Jack Dorsey: Someone will do it for you eventually.
Fiatjaf: Yeah I hope.
NVK [1:34:57]: I think the catnip will start kicking in soon and you’re gonna start seeing a lot of very very interesting people with very specialized skills coming into the Nostr Core-ish development.
William Casarin [1:35:13]: This brings to mind something that I never considered, which is: there was an article in the Indian Express or something — some large newspaper in India — and I was trying to think about what would happen if a large percentage of India started coming into Nostr and what that experience would be like, at least in Damus? I feel like we’d almost need a relay per region, because it would be very confusing — I don’t know, I’m just trying to think of how does this scale out of your country?
NVK [1:35:40]: Oh, so part of the NIP we want to propose — Dust-identify: so essentially it’s a message for the relay and how to engage with the relay. The relay identifies itself for region and then you can find some sanity there. I mean it doesn’t have to be formal, but it would be nice if there was some way of identifying with a two or three letter code for a country, because you’re gonna have legal issues with relays — you’re gonna have all kinds of stuff. But I think the solutions will present themselves. At least identifying the language could be nice, too: once you have a lot of clients, a lot of world penetration, you want to be able to know what’s coming in. But you can also have a filter relay that just translates everything to you.
Jack Dorsey [1:36:25]: India is a very interesting use case because there’s over 300 dialects and those dialects tend to have different cultural attributes as well. So it’s not even the state understanding, it’s more of the localized dialect and cultural attributes associated with that dialect.
William Casarin: I didn’t even know that — that sounds even more complicated.
NVK [1:36:52]: And then you’re gonna have charset issues with clients: you have all these expanded character sets for obscure languages, or languages that your clients never had to deal with. And supporting them all at the same time might be a problem — those are interesting issues.
William Casarin: Your idea of translating at the relay level is going to be pretty difficult, considering all the messages are signed.
NVK [1:37:15]: But maybe the trust model is different — maybe you have some control over that relay and you’re just receiving now cleartext. Maybe your client is even lighter than a normal client: it’s just essentially doing text — it’s not doing signed event messages.
Fiatjaf [1:37:34]: That’s one of the attack vectors, I think: I’m worried about people doing clients that trust a server and you only talk to that server. There’s one client called member.cash that’s made by @bitcoincashguy and his client is very fast because it’s talking to your backend that does all the hard work on the back end, but it’s also not very censorship resistant.
William Casarin [1:38:01]: I feel like we would standardize a lot of that stuff, because I’ve been thinking about this for the stats spec: we’re probably going to need some aggregate rest endpoint — I’m not even sure it’s a good idea to do it all at the protocol level, because we’re gonna need lots of stats for different things. Maybe it makes sense to have a spec that just pulls the stuff efficiently so you can get the gains [that centralized systems have] — because I noticed his client, it actually has follower counts and accurate counts for things and it’s because he’s using a backend, but if we had that spec in somewhere, that might help clients out a lot.
NVK [1:38:21]: But if you have a no-permission network, this is bound to happen. Coinbase exists where they custody Bitcoin for a bunch of people that don’t even understand that their Bitcoin is not theirs. I think you’re gonna end up in the same place — with freedom comes responsibility. Some people are just not interested in that, and say, for example, Twitter decides to create a Nostr relay and they may not be interested in delivering actual Nostr messages to their client. now you have a marketplace of clients — which is great — but some people may choose to have a subpar, more filtered, more state actor-curated version of the global feed. I don’t like it personally, but I don’t see a problem with that.
William Casarin: Not your Keys, Not your Notes.
NVK: Exactly — Not your Keys, Not your Notes.
Fiatjaf [1:39:18]: I actually like it, but I do see a problem — so it’s the opposite: I think it would be nice to allow people to do that, but I worry a lot about it. I think centralization is a good way to scale things and do things in custom ways that is not standardized. I also like these on a gut-level, but I see it may cause some problems. But I think Nostr is fine against these kinds of attacks: even if Twitter did this, just having a little amount of people not doing it would keep the protocol open.
NVK [1:40:04]: So we’ve went all over the place with Nostr — let’s talk about how do we win? Because all technology can be cool, can be amazing for a subset of people, but winning at the global level is a whole other ball game. I guess there is one person on this call that has a little bit of experience with that: so Jack, how do we keep people interested in at least this use case of Nostr? So that we can have a driving thing so that it attracts people, capital, and other things? What would be some lessons that you’ve learned building what you’ve built, for us to create engagement, create stickiness? Because that was the problem with Mastodon: nobody stuck there. And compound this, snowball this, and then take over the world in the next month?
Jack Dorsey [1:40:57]: In the next month? Well I think the open development model — the development in public — is absolutely critical. And that the discussion is happening over the protocol itself, and the iteration, and just the guidance that is being provided in the idea is important. But it does come back to this idea of: the net new thing here — as Fiatjaf has tried to amplify — is to me the interconnection. And this concept of many micro-apps solving different use cases or different problems enables so much more than just someone who’s working entirely on social media, because there’s more of an open field feeling — it doesn’t feel like you’re constantly competing to add more bells and whistles to it, but you’re actually bringing new utility to this new fledging ecosystem. And someone might find it interesting and useful and valuable, and that encourages you to keep going with it. So there’s a sense in Twitter — when we started — that we just happened to be in the right place at the right time with a lot of bloggers, and that turned into journalists, and that kind of set the tone forevermore as you see today. With Nostr, having what feels like a majority Bitcoiner-type attitude will create an entirely new system, and I’m hoping that there’ll be some point — and you’re already starting to see a little bit of it where the conversation around the protocol takes up 1%-10% less, and more topics emerge and more use cases emerge — and that’s what makes it attractive to more people. I think some of the people trying it right now are going to see a bunch of Nostr and Bitcoin talk and think that’s what it’s for, but the brilliance of the micro-app idea is: they don’t have to see any of that — it’s just providing them real value right away, instantly. And having these little separate views on this giant canvas I think is really cool and really important to getting mass adoption, but I wouldn’t think about it as how do we get mass adoption on the social aspect — it’s more about mass adoption of the use of the micro-apps together. To me, success and winning is: what number of micro-apps is any one person utilizing at once? That’s where the stickiness feels like it’s going to come in, is: I may fire it for the social thing because I have this other thing over here, but I’m going to continue to hire it to solve these problems for me — and that maintains a relationship with the overall protocol instead of quitting the whole protocol altogether, which I think is a single point of failure in some of the other development models such as Mastodon and maybe even aspects of Bluesky, where they’re going after one particular use case, and if you’re not getting value from that use case that’s fundamentally new to you and valuable, then you’re quitting the whole protocol. Whereas at least with Nostr you might find something else that’s not social that you find value in, so you’re not quitting the protocol — you’re just quitting one little micro-app of it. And again: to me it’s fundamentally new in this time, and we’re not even seeing a glimmer of it just yet — maybe a tiny bit, but — there’s so much more to come. We’re definitely early.
NVK [1:45:01]: So what you’re saying is that you’re gonna keep on being the Nostr greeter and saying hi to every person that comes in?
Jack Dorsey [1:45:08]: 100% — that just feels good. The other thing we learned from Twitter is: if you’re not getting some tangible feedback immediately, you’re lost. So even if I had like 10 people on in the day, those are 10 people who I’ve had some interaction with where they’re going to look for the next thing: okay, so something happened — now what’s the next thing? And the next thing and the next thing. I just think it’s important that as it’s developed in public, the interactions are very public too, because people can see it and they can feel it and they understand the potential when they feel it more. It’s not abstract anymore — it’s real.
NVK [1:45:53]: I find that every successful technology has a very defined early culture — that then waters down and disappears much later on, but — in the early days there’s a culture. In Bitcoin there were free Bitcoin taps: you go there, you get yourself some free Bitcoin. Everybody gave Bitcoin to their friends and family: Try this thing out! Everybody who got that Bitcoin for free lost it, but that’s beside the point. And with Lightning, there was the Lightning torch where people were giving it to the next person, and then there was the tips. And what these things all have in common is this idea that you have a chain of humans interacting with each other in one way or another, and make it friendly and interesting right away, like you mentioned there. So yeah this is some interesting insight. William, you’re building a client that’s fairly well used and becomes this very clean experience for the system: How do you see us winning short-term? How do we keep the momentum going?
William Casarin [1:47:00]: I think it’s just showing those novel use cases. You’re not gonna win people’s attention if it’s just a crappier version of Twitter, because Twitter is pretty amazing. I love Twitter — I still use it. So we really need to be tapping into these new things that really blow people’s minds. So that’s why I love focusing on this Lightning tipping feature — which I plan on adding soon, which I got working the other day — and seeing people tip on your posts and seeing who it was. That’s a new experience that I don’t feel has happened yet, but when you imagine listing those in terms of the highest tip on the post, now it becomes this gamified aspect where people want to get to the top of the tip list — and who knows what that’s going to unlock? People are gonna love that, I feel like. And just adding more things like LNURL widgets, and just making it fun. If it’s fun then people are gonna love it — people already love it, but let’s just make it more fun. That’s the goal of Damus: let’s make it maximally fun, because we have that opportunity now where we can just do it. So that’s why I hope that we keep building these cool, fun demos because we want everyone to experience those fun things.
NVK [1:48:10]: For people who haven’t tried Nostr on Damus yet, or just the general social application of it, is absolutely, hilariously fun — people are having a blast on this thing. And people are sticking around because of that.
Fiatjaf [1:48:26]: You say that because you’re using Damus — you should be like me and use the most trash clients that people are making and try to make them better.
William Casarin: Yeah well some people are into masochism I guess.
NVK [1:48:37]: There are Android clients coming. So Fiatjaf, I’ll pull the same question to you: what keeps the momentum going? And how do we win in the short-term aspect?
Fiatjaf [1:48:48]: I have no idea. I think we need a few more Jacks to give more boosts to the thing. The actual use case I had in mind was just these people getting banned by governments and stuff, and I think it would be nice to get these banned people introduced [to Nostr]. It would be nice to get other niches besides Bitcoin — and I know no one here agrees with me with doing that, but — getting some other niches will be good. I was thinking that it would be interesting to create a separate protocol and say: Marketed only for people that hate Bitcoin. And it’s exactly like Nostr but then it grows from the other side, and then when it gets bigger, we realize that it’s perfectly compatible with Nostr.
William Casarin: Nostr Cash or something?
NVK [1:49:48]: Nostr Cash. I do think that there’s a lot of people that won’t come until you have a mute button, a block button. I think most people want censorship — they just want their censorship. And that’s fair: people may not want to see something that they don’t want to see. And I’m sure it’s coming.
William Casarin [1:50:07]: I’m surprised I haven’t really needed to implement that. People are generally pretty nice — obviously it’s not going to scale forever, but I’ve been putting off that feature because I haven’t felt the need to do it to anyone.
NVK [1:50:15]: I think it’s because we have like-minded people — it’s so early. Guys, we’re reaching about two hours here. This is an absolute amazing conversation. What other things maybe I have missed that you think we should bring up? Imagine maybe this is the only two hours that somebody’s gonna get about Nostr. And because it’s so early, I hope I can shove in as much as possible in terms of aspects of this and angles of this. Because again: protocols is a sales job, and I want really quality people to come into the space and come early.
Jack Dorsey [1:51:05]: I think in terms of the sales job and marketing: just showing more of this micro-app idea and showing more of those use cases is important, because a lot of people coming in today, the first thing they think of to build is probably going to be a client — a social client, at that — and miss this whole other universe of potential. And having more conversation around the importance of how all these things interop together and work together is a little bit missing right now, because we’re still in that run-from phase. So once those things are seen — being able to amplify them and boost them and pull more focus on it. Hopefully we’ll see more people build these simple, tiny little things and get funding for it through these respective foundations that William and Fiatjaf have have created. And then they’ll be talking about how they got that funding, which will encourage more people to do similar things — that’s one. And also Fiatjaf brought this up, but people are lazy so they go to defaults, and the defaults that are built in to the clients or the micro apps will likely create momentum that’ll be hard to move away from. So this whole earlier conversation around relays and what relays are chosen and how those are rotated or what people show up with or what they see as a Like or what they see as a boost or how they perceive what a Note is on this one view versus another I think is really going to set the tone — in the same way that the Shaka has set a more positive, welcoming tone, there’ll be other things, other defaults that just take a life of their own. So being super-mindful of what’s being built in — especially on William’s side, as you cross the chasm of the App Store — someone’s first introduction to this ecosystem is what you’re presenting them, and it’s going to stick. So anything that you feel is a significant mistake at the moment that should be reconsidered I think is important to at least think about that, because these momentums take over and then there’s all these new feature requests that have nothing to do with the original intent, and you get super reactive — I feel like there are some points of failure there that I would think a lot more about.
NVK: So, no pressure.
William Casarin [1:54:02]: I’m already kind of feeling that. It’s weird because there’s so many users and they want certain things and I’m like, Well I was just trying to build a cool Lightning tipping prototype. But now it’s like, Oh wait these people actually need a usable app that you can actually use. I do want to keep to the roots though. As a protocol maximalist, I just want to build the best possible thing that demonstrates the power of the Nostr protocol — that’s always been Damus’s goal, so hopefully I can stick to that. I want to see someone make a poker client for Nostr so we can play poker — that’s an idea I’m throwing out there at the end.
NVK [1:54:36]: You know the original Satoshi client had poker in it, right? It’s fascinating how the decentralized people cannot stay away from the poker!
Fiatjaf [1:54:47]: Is that true? I thought it was a joke.
William Casarin: Yeah [since the first release] there’s a poker client.
NVK [1:54:55]: It’s very exciting guys. Fiatjaf, I really want to hear your opinion on what do you see? If it differs from everybody else on how we win? And I just wanted to say that it’s kind of special that the biggest, most used client is made by somebody who is also a [Bitcoin] Core contributor — somebody who actually cares about the protocol. I think that’s special. It’s nice because you’re gonna have more consideration for that. But yeah Fiatjaf: How do we win? And what do you think could be done?
Fiatjaf [1:55:39]: I thought I had an answer to that, but the answer is: I don’t know. No, I answered that when I said that we need other niches, other use cases, other groups, other communities. I think Nostr can support other people — and yeah that’s it. Also on the other point: it was a miracle that William happened and liked Nostr and started building good things, so I think we wouldn’t have gotten anywhere without Damus. But yeah I don’t know how to win. I think Bitcoin is so far from winning that talking about Nostr winning is such long thing — but it could happen overnight, probably.
William Casarin [1:56:34]: We’re only two years into this thing — it’s still pretty early.
NVK [1:56:38]: Yeah but a man can dream, right? I used to call Bitcoin an experiment up until 2–3 years ago, and I don’t think it’s an experiment anymore — it’s winning in its own manner, at its own pace. But there are many wins along the way, and I think you guys had quite a few already — so much so that the interest about this stuff is amazing. So guys I guess the last thing that I wanna address is: if you are leaving somebody with something like a thought, an ask, or a piece of information or whatever, a lot of people know something that you want, that you need, that the protocol needs — go with that, and then maybe just tell people where they can find you? Why don’t we start with William here?
William Casarin: Sure yeah. So I’ve been working on this iOS client Damus. And it’s open source: if you go to damus.io/code it’ll redirect you to the GitHub — help out. And then we have this Nostr fund: I’ve been writing down all the names of everyone who’s doing the most work and helping me — and I’m doing that with other clients as well — so if you just start contributing lots of code to all these clients, we’re watching and we’re putting you on the Santa list for the Nostr fund. So help out and you’ll get paid. So thanks Jack for providing that, and yeah just help out and join the fun.
Fiatjaf: Same thing, and also please try to not bloat the protocol, because one client can make a difference and create bad things for the other clients. One client can implement a thing that puts a burden on everybody else, and this kind of stuff should be carefully handled — that’s my request for people that are coming and want all the features in the world. Just please: take care.
William Casarin: Don’t back up your cat photos on the relays, please.
NVK: I guess it’s early — there are not a lot of specs so it’s like, Be gentle: don’t abuse the protocol, because otherwise things start to get closed down.
Jack Dorsey: William said it earlier, which is: keep it fun, keep it weird. As soon as we lose that, I think we lose some of the potential. And I have no doubt that this will stay fun and stay weird, and that will be the driving factor. For me, my experience with Bitcoin, is it just took me back to the old Internet where it was fun, it was weird, it worked sometimes, it didn’t work other times — Bitcoin obviously worked from day one, but the weirdness of it and just like, I’m getting into this because it’s fun, it’s refreshing, and it feels new, and I can actually contribute to it in some meaningful way. And it doesn’t have to be code. I think the thing that will help the client development the most, and also this idea of the micro-apps, is getting more of the designers engaged around key management, around settings, around relays, around interactions. When they get in and they can pair in a frictionless way with an engineer — and more and more designers are hybrids of course, but — when you get that element in, there’s just a new level of thinking. The community thus far has set the tone for weirdness and this joy that we experience when using it, and I just hope that we can persist that as long as possible, as more and more people come in, and as more and more utilities — that are serious — tend to be developed upon it.
NVK: Yeah thank you so much guys, this was an absolute blast. I loved the conversation — I think we covered a lot of ground. Who knows? Maybe we do another one of these in a few months from now and touch on where it’s gone and where it’s at. So I encourage anyone listening to this that haven’t tried Nostr — there’s a bunch of resources: we’re gonna put a lot of the links on the show notes. Download Damus or some other client and do come say hi — it is very satisfying. Do try to change clients — it’s also very satisfying. This protocol might have legs, and I think we’ve got something special here. So with that, I’m gonna sign off and I’m gonna put also information on how to reach out to each person here in the show notes. Thank you so much guys.
