David King | What’s new with Stacker News and Nostr? A conversation with Keyan Kousha and Max Webster
Link to the YouTube video (the timestamps are based on this): https://youtu.be/wLF28WcCkb0
Intro: Welcome to the Curious DK Podcast, where we talk about Nostr, Bitcoin, AI, and other technology topics I’m following. I’m an angel investor and entrepreneur with over two decades of experience in Silicon Valley building relationships, technologies, and products to improve people’s lives. In this episode we catch up with Keyan, who is the founder of Stacker News, a subreddit-like platform that incorporates Bitcoin as a signal of information quality in a media platform. This platform has obvious relevance to the emerging technologies being built around Nostr, so I was excited to catch up with Keyan and hear his thinking and learning about the latest with Stacker News and Nostr. We’re also joined by Max Webster, who is the founder of Hivemind Ventures, a venture capital firm that invested in Stacker News, and is betting big on funding companies within the emerging Lightning Network and Nostr ecosystems. I had a great time discussing and learning in real-time about the possibilities for these new technologies. I hope you enjoy listening along and learning with us. This is an ad-free podcast. If you find it valuable you can send me a tip at my Lightning address: dk@stacker.news.
DK: Keyan, I met you through Twitter — because I saw Stacker News and the first DM I sent you was, Thanks for building this, I’m really excited about this — and then I was fortunate over time as you were putting together some fundraising, to get more formally involved. I’m a big fan of Stacker News, but I think I’ve become an even bigger fan of you as a thinker who’s built Stacker News and could do all kinds of interesting things. And then Max I think I met also about a year ago or so, probably. That was around when we were getting together on the Discord a lot and just wanted to meet more people looking at how to build things to make Bitcoin better, and the ecosystem around things that could be built now that we have this Internet currency. And yeah so we got connected and have had a lot of fun chats, and it feels like we have stuff to talk about every day now. So thanks — but anyway, that’s in my words. Why don’t you give a little bit of your own intro? Keyan, do you want to go first?
Keyan Kousha [2:20]: I’m the founder of Stacker News. I usually describe Stacker News as like Reddit, but the karma is Bitcoin — and that’s just the highest-level thing. There are other differences, too: you have to spend a little money to post and comment — which Bitcoiners call Sybil resistance — but I think the main point is you pay to participate on Stacker News but you also get paid, and that we think is kind of new. There are other little things we do: we use Bitcoin to determine the ranking of items and have some ideas around how that might increase the signal of posts. Also any revenue we earn, we give back to the best users on a daily basis using a trust algorithm called called web of trust for that — that might dovetail into some of the Nostr talk we’ll have later. That’s the broad overview: it’s a Reddit-like platform that tries to integrate Bitcoin as deeply as possible, and tries to do as much as it can with Bitcoin. One of my big gripes is Bitcoin gets treated as a static asset, and I just think it can be used for so much more. That’s what I’m really passionate about in Bitcoin — I mean I love it as a store of value too, but it can allow us to do things we weren’t able to do without it.
DK: That’s great! Thanks Keyan. And Max, do you want to give a little bit about yourself?
Max Webster [4:08]: I am the founder of Hivemind Ventures, a pre-seed and seed fund focused on Bitcoin, the Lightning Network, and increasingly I’m excited about Nostr. Even before I met you, Keyan, I was a user on Stacker News — I was lurking back in Fall 2021. I thought the site was a beautiful site — the best UX I had seen for a Bitcoin Lightning “app” to date — and I saw some early community development there, a lot of really exciting conversation. I had the great opportunity to then invest, leading both Keyan’s pre-seed round and then seed round last year, and I am a daily active addicted user to Stacker News. But like you, DK, even though the site is incredible and I love all the different ways it’s developed, I’m more of a fan of Keyan and just the way he thinks about things. And yeah I’m excited to see where he’s going to take it.
DK [5:00]: I feel like some of the stuff that Keyan always reminds me on accident is: I don’t even realize these things that he just has so in his core that really shape the way I see the site and see his thinking. For example: I invested and I didn’t realize the whole thing was open source! I love that it is — I think that’s the right call and I love that Keyan is principled and understands why that’s important — but I was sort of like delighted after the fact to find that out.
Max Webster [5:32]: And not to spend too much time bragging on Keyan here because I know that’s probably a little awkward for him, but I will say: it’s also just phenomenal how fast you shipped here. When you had this idea around web of trust and then immediately you shipped the algorithm, and then we talked about this value rank using satoshis to help influence signal — I put out this piece called How to Disrupt Google that you gave me some great feedback for that we’ve been discussing — and then within a couple of weeks it was just, Oh by the way I’ve already shipped it. And then — I’m sure we’ll talk about Nostr later on — Oh, what if we could do this Nostr identity verification? And within a day I was like, Oh it’s out. I’ll stop now.
Keyan Kousha [6:11]: To be fair, it took me a little bit longer on the Nostr stuff because I was working on something else for a while, and it took prodding from all of you — the users included — to get me to implement it. So I want to at least admit that.
DK [6:28]: One thing that would be a fun starting point is just to understand what have you learned so far from operating Stacker News? Maybe there’s some surprising challenges along the way that you didn’t expect? It could be surprises to the upside, surprises to the downside. What have you learned in the process of building it?
Keyan Kousha [6:51]: It’s difficult getting people onboarded to Lightning. When I first built Stacker News, users would get two free posts and five free comments. And the idea was that most people want to get started really easily and maybe don’t want to load up a wallet — or maybe they don’t even have Bitcoin on Lightning. Before I built Stacker News, I hadn’t done much interaction with Lightning in general. I was excited about it, but part of building Stacker News was interacting with Lightning and figuring out what it’s capable of. That’s how it started out — at some point people started abusing that a little bit, and so that was kind of surprising. We decided to introduce no free posts — you have to start paying. And that ended up slowing down growth: we noticed new users were signing up and the friction was just too high to load a wallet — the ramp up for a new user onto a Lightning product needs to be a very shallow incline. That was something we learned, and it’s probably true of any new technology: it just needs to be dead simple to get people started. One of the things we learned is you can’t just force people into a new paradigm — you have to mix it with the existing paradigm. That was another decision why Stacker News is centralized and started out centralized — we’ll be talking about more decentralized social media, but — the UX is familiar, it’s easy. Maybe Stacker News — the way I envisioned it with all of this incentive structure based on Bitcoin — wouldn’t land with people? Maybe it wouldn’t make sense? And if it doesn’t land, then the decentralized version certainly isn’t going to land because the UX is going to be so much worse. So that’s another example of having a familiar experience to introduce people to new experiences in, and I think that’s maybe one of the bigger learnings. But the whole thing I try to think about a lot with Stacker News is just incentives: How do you structure the game theory in a community such that people behave in more the community’s interest than they might otherwise? I think that’s really interesting, and that’s what fascinates me about Bitcoin is that it just runs and it encourages people to live their lives around it with its incentives. And so for Bitcoin builders, that’s really the goal post, is: How do you build a product that’s as good as Bitcoin in terms of its incentives? That’s what I think about a lot. But I’ve learned a lot: UX matters in a consumer app — I mean that’s obvious stuff, but it’s really that that new user path needs to be really dead simple.
DK [10:36]: It sounds like you used to have a few free posts and then you shut that off because it had some abuse — is that an area that you’re constantly iterating, exploring, and testing? Or do you feel like you have that roughly how you want right now?
Keyan Kousha [10:51]: Right now it’s fairly good. What we do right now is we partition users based on whether they paid for a post or not, and then we have a decent portion of our active user base who participates in onboarding them with Bitcoin. Their posts are hidden when they’re free, and then like 10% of active users will go through and give them Bitcoin to get started if they’re posting in earnest, and then they’re able to participate like anyone else on the site. So we built this ramp for them, and that feels pretty good. Eventually at scale on those free posts they probably need Web 2.0 defenses — captchas, spam filtering, all that kind of stuff — but for right now this is a good enough MVP to protect the main site stuff.
DK [11:52]: What’s top of mind for you? Do you want to grow the site or improve other features on the core site or look at some of these more avant-garde topics that we’re a little obsessed with? Or where’s your head right now?
Keyan Kousha [12:08]: This is probably the biggest struggle for me as a founder, is: I have so many directions getting pulled on at a time. Within the last 24 hours I probably had 10 feature requests. Where do I go? It’s really hard to build confidence around a direction when someone’s like, Look at this — this is really important to me. I try to make myself as vulnerable to empathetic feelings as possible, because I feel like that’s what I need to do. It’s a struggle, but I constantly have to re-center myself around the idea of: I’m building this for me — as the generalizable version of me, not the esoteric version of me — and what that person wants, and then also fit that with what excites me and gets me to move fast. Right now occasionally it’s like biding time and filling in some of the gaps in the product where people have been complaining for a long time or whatever, while we figure out bigger strategy things and where to do experiments and stuff like that. So just honing in on more the core user experience working right now: onboarding for a new user, how to make the conversions a little better when they come, how to make it so they understand the site faster and can get started — those kinds of things are important because you’re constantly fighting churn in communities, and if you’re negative growth the thing tends to die so you really need to introduce new users fast. So we think about those kinds of things just as like core product stuff, but there’s so many ideas for things that I want to do. Strategically, oftentimes it makes sense to just refine, and that’s what we’ve been attempting to do for the last few months or so is just refine certain features and make them work better. We talked about the ranking algorithm and improving that, always revisiting the trust algorithm — that will be a never-ending pursuit probably, but — making those things better is what I normally spend my time on. But I think a lot about: What do other communities on Stacker News look like? What other post types look like? How do we decentralize Stacker News? How do I avoid some of the pitfalls of centralization if decentralizing it isn’t an option? These kinds of things.
DK: Do you have a process to work through those questions? Do you have a sounding board or send out to a couple friends on a Signal? How do you think through those?
Keyan Kousha [15:29]: I think I complain enough to you guys in the investor updates! But the way we do this is I lay out big ideas in a Notion dock and then it’s like: number of users impacted, what it unlocks, how long it takes to work on. That’s the template we use to lay out an idea — that’s roughly the process right now. It’s sad or maybe not super mature, but whatever I feel very motivated to work on — because I know I’ll just get it done — is important too, and so that plays a huge role in nearly everything and in what I end up deciding to do. Because if I’m not motivated to do something, it’s a slog and I’ll burn out or whatever.
DK: Do you feel your motivation is usually coming from more like a technology idea that you hear about? Maybe there’s a new feature in Lightning Development Kit? Or is it more like a user’s complaining about something and you want to address that? How do you handle the trade-offs there?
Keyan Kousha [16:48]: Honestly it’s mostly risk that attracts me, which is kind of weird. But I think that’s what draws me to being a founder: I really loved working on web of trust, I really loved messing with the ranking algorithm and sats — I like doing things that are new. Trying to copy something that already exists elsewhere is just so boring. They have 20 engineers working on an editor at XYZ company that makes their editor great, and that’s what I can hope to achieve as a single engineer. That’s great — I could pat myself on the back for that. But did I move anything forward for anyone, doing that? No! It bores me. It’s the things that someone hasn’t done yet or very well yet — that’s what I think excites me the most.
Max Webster [17:48]: That’s something actually I think that’s great about you Keyan is just: you follow your intuition, follow your gut, you’re experimenting with new things. As I think about all of this stuff as an early stage venture investor, incremental is never going to build the next great big company or the next great big project — it’s going to be something brand new that’s never been done before. I think you’re absolutely right to trust your instincts. I don’t want to get too far ahead of ourselves, but as you think about some of these different big moves you can make next, it seems to me three things really jump out of the page, and I’m curious how you’re thinking about all three of these: 1) getting people more sats seems like a huge problem — have you considered just integrating with MoonPay or something like this? We’re starting to see other services are doing that now. Is getting people sats still just the biggest baseline challenge?
Keyan Kousha: Oh like to get them started? No it doesn’t seem so much that — once we recognize them as a human user or whatever, the users are very generous on the site and will get them started with sats.
Max Webster [18:53]: I guess my bigger question is: because you see we have tens of thousands of people — whatever, 50k, 60k, 70,000 people viewing the site every month — but I always think to myself: How easy is it for them just to realize, Hey I can get tipping immediately. I can go from zero Bitcoin to attempt my first post.
Keyan Kousha [19:15]: Yeah we don’t have maybe a ramp for that, because right now it is: you sign up, you post a bio, maybe, you get some sats if people like your bio, and then you can start tipping. And so maybe that loop is kind of slow, but it’s also difficult where you don’t want to give people free sats when they sign up without verifying they’re a human. So maybe having something easier that’s fun to get them to prove that they’re a human to start — which will be harder and harder as time goes on, given all the AI stuff that’s happening, but — that’s something we’ve thought about. We haven’t thought about it more recently. Right now it’s just converting the users who are lurkers: I’m trying to convince them a little more to participate and nudge them to participate is what we’re trying to do.
Max Webster [20:09]: And just to be clear there: although I love the idea of helping users earn those sats via capture or whatever — that’s a great idea, I meant just a very basic [new user flow]. Like Alby integrated with MoonPay, so now I can just put in my credit card or a bank account — a user flow I’m very used to: I just buy, whatever, twenty dollars worth of sats.
Keyan Kousha [20:31]: Yeah that’d be interesting if we can do that easily. It’s just high friction I guess, but I suspect for a certain type of user it’s the friction they’re familiar with and wouldn’t mind going through. I think that maybe makes more sense in a world where Stacker News isn’t geared so specifically to Bitcoin content. Right now it’s maybe not the right move, but it’s something to continue thinking about. I know a lot of products in the space do that, that are trying to fit in a wider market.
Max Webster [21:10]: For what it’s worth: Fountain recently integrated it and allowed you to buy sats with your bank account — in the last release a week ago. And I will say: Look, it’s just already — for a lot of people I’m talking to — way easier to get them over that hump. Maybe they’re Bitcoin-curious but still not there, so something potentially worth chatting with Oscar [Merry] about.
Keyan Kousha [20:31]: Yeah that’d be interesting if we can do that easily. It’s just high friction I guess, but I suspect for a certain type of user it’s the friction they’re familiar with and wouldn’t mind going through. I think that maybe makes more sense in a world where Stacker News isn’t geared so specifically to Bitcoin content. Right now it’s maybe not the right move, but it’s something to continue thinking about. I know a lot of products in the space do that, that are trying to fit in a wider market.
Max Webster [21:10]: For what it’s worth: Fountain recently integrated it and allowed you to buy sats with your bank account — in the last release a week ago. And I will say: Look, it’s just already — for a lot of people I’m talking to — way easier to get them over that hump. Maybe they’re Bitcoin-curious but still not there, so something potentially worth chatting with Oscar [Merry] about.
DK: I wonder if I could mention the n-word?
Max Webster: I was holding off — I was waiting for you, DK!
DK [21:37]: Okay, because I know we only have a half-hour planned and Nostr’s been on our collective minds a bit, and so I’ll just offer at least a few thoughts to get us going, but really this is meant to be more of a brainstorm among us to think about what [Nostr] — as a potential new primitive for the Internet — might mean for Stacker News. We all were talking about this briefly about a year ago on that Discord we were apart of. It’s a compelling and simple idea, but then until you actually have a client that’s pretty good, that’s usable, it’s hard to really know what it would actually look like in practice. Now we have that client: Damus is a pretty good example of that. I think my excitement around Stacker News is that it helps us tie information and publishing and communication to value in a way that doesn’t exist before — that it gives us a view into what that future looks like. And then when I look at Nostr and the Damus implementation of Nostr, they’re also exploring a similar idea but with a different architecture underlying it. You can do Lightning tips, which are forms of stacking — I actually take my Lightning tips and push them into my Stacker News wallet. I have it nicely integrated because of the NIP-05, which we can talk about. But I feel like some of the core beliefs or hypotheses around what Stacker News is really good at, I think we have a new architecture for accomplishing different but related things with Nostr, and the way it’s demonstrated with Damus. Given that as a little bit of preamble, I wonder: should Stacker News take this as a moment to consider more integration points than NIP-05? What have we learned from the NIP-05 integration? What other things might be next on the roadmap there? Or is it a sideshow, distraction, we don’t want to do that right now and it’s more just keep the eye on the current path and keep pushing on that? I think that’s why it’s very relevant, because vision-wise it’s the same, but architecture-wise it’s new — that’s what gets my Spidey-Sense tingling, and that’s why I wanted to get us together to put our heads together and see what comes of it.
Keyan Kousha [24:29]: I think it’s really exciting! I know DK, you and I had a chat before, and we’ve been discussing Nostr and Stacker News since the beginning of Nostr, probably. And Fiatjaf has been in my replies — Wen Nostr? Wen Nostr? — for like a year. Yeah I’ve always been very excited about Nostr. I’m a fan of reading Internet and technology history, and it fits well with my understanding of grassroots technology development, and that’s what I really like about it. I don’t think it’s a sideshow or a distraction — I think there is something really important going on. The biggest problem with a lot of these protocols that preceded Nostr was the lack of an identity that was decentralized, and they’ve solved that with a really easy thing that a lot of us are familiar with now, given that we have Bitcoin — they use a similar concept for identity. They’re in that thread I shared with you, DK. There are really three things that I’d need or I’d like to see or to have a direction to go on — and this doesn’t have to be provided for me, I can provide it for myself, but — things like: some kind of solve for discovery is something I’m concerned about — maybe how that works at scale on something Nostr. Data availability: how — when I send my data to a relay — do I know it’s going to be there? Maybe it’s there for 100 days and the 101st day I wake up and it’s not there — what happens? How do I incentivize the relay operator to keep it? And then the other thing is spam. Nostr hasn’t had its Eternal September moment yet, and it will eventually. I don’t want to predict it, but it likely will happen. How does it deal with that? What’s cool about Nostr is every relay operator can have their own approach and the best one will win out, because it’s open. How do we solve that, too, is another thing. So those are three areas where I would begin thinking about first, when I would begin a deeper integration with Nostr. As far as more immediate-term plans: doing a read-only relay and seeing what that’s like, maybe doing some delegation for users and posting their content on Nostr as them might be interesting. But I’d love for Stacker News to be decentralized and relieve me of all the liabilities that we’ve learned from Jack Dorsey about running a centralized service. So those are my broad thoughts. I think for now it’s an interesting thought exercise to remove the halo that Jack Dorsey imparted to Nostr and see what you think about it then — that is biasing people a little bit. It is still kind of early. I listened to a great podcast — Bitcoin Review, with NVK and William and Fiatjaf — and Jack [Dorsey] was like, A lot of people are using Nostr because they’re running away from things. What do we do that makes them run toward it? And I think that’s an interesting question. I think Stacker News similarly has people running away from Reddit. They’re running away from Twitter. We have some people running toward it for Bitcoin, but maybe that’s not super compelling yet. Anyway, that’s my high level thoughts on it.
Max Webster [28:47]: One thing I want to add in there, because I’ve — as DK knows and as you will soon know — I’ve been obsessing over this Nostr stuff: this is the most my Spidey-Sense as an investor has been going off in a long time. And I think it’s for a couple of reasons: 1) the architecture itself does seem decentralized and potentially robust enough that if enough people were running relays then this thing actually could take off, so that’s good. And 2) it actually is having a real adoption rate. Since December 13th — when Jack tweeted about Nostr — to today: at least by some public counts and public keys, which are identities, we’ve gone from fewer than 1,000 users to 250,000 users. And so that’s crazy to see that kind of growth in less than a month. And so there’s real interest, there’s real demand here. I think some of it is because people are running away, but honestly the bigger piece — more than running away — is people having fun because it’s open and they’re experimenting. When you let the nerds tinker, they’re gonna create all kinds of beautiful new things and new experiments. And so when I think about some of the different interesting integration points, there’s a couple of things: 1) the more developers you get hacking on your project, the better — that’s just one of my general things I posit about products in the world. And so if we can do that — Stacker News already has an API, but — if we can somehow get the same guys that are building on the Nostr API to build with Stacker News, that seems like not necessarily an easy one, but the way to get the most experimentation and the most fruit borne. As you publish more content and allow outside users from other Nostr clients to potentially interact with Stacker News posts, you create two different opportunities: 1) those users on Stacker News can potentially earn much more tips, which is really interesting. So if I put out a post — and maybe there’s a couple thousand people tipping on Stacker News — but now I get two orders of magnitude more that are tipping, all of a sudden it’s like Wait a second! If I can increase my revenues by two orders of magnitude, that’s pretty interesting. And then 2) particularly because we would have one the best clients pretty quickly — and maybe this is coming from you running your own relay as part of the client or in addition to the client as well, but — I think that the killer app, really really big opportunity here, is as you said: it’s going to be search and discovery, which is completely untouched right now. And the more data that you can index, especially with this open standard — now that there are people doing Lightning tips, you’ve got the social graph that’s decentralized, the value rank signal is becoming decentralized — it kind of goes back to the Google thing we’ve been talking about for months, which is: whoever can index the most data, find the best signal from social graph and value ranking, I think is probably in the best position. How does Stacker News take advantage of that? I don’t know, but those are the two big areas that I’m thinking a lot about.
DK [31:41]: Yeah and I think the discovery piece that Keyan you mentioned is an important one that the structure of Stacker News could be really helpful to lots of different types of discovery to happen. But I also think the clients in this world and the applications of this world may end up looking a lot different because they’re all in competition with every other client, which is both exciting and then also — if you’re building a client — may be terrifying because it’s like, Okay we thought that was where the value is going to be, but now that there’s a new architecture maybe there’s different places to explore where you can create value? So the same principles that were exciting to begin with like having value as part of communication and publishing and treating the Internet-native currency of Bitcoin as the marker or the unit to indicate quality or commitment to a piece of published content — I think that still remains and is transitive across projects. But I think the relays are a very interesting touch point. I’m happy to hear that you’re exploring that because it’s nice that the architecture allows for it to be very decentralized, but I think some relays will emerge as the really good ones, and they may have some centralizing effects, but probably not monopolistic centralizing effects — as long as they’re of service, they can exist and do well. So I like the idea of experimenting with relays, and one strawman I’ll throw out for brainstorm purposes is: What if the open source software of Stacker News had a version that was a Nostr client — so you can do all of the light integrations with the existing community as is, I think that’s a good first step and you’re really good about sequencing these, but if we take the bizarre, radical thing — what if it’s a community client that I can use to run my own coffee nerds Stacker News-equivalent, but using Nostr for all of the identity and message storing and transport? And if you are the provider of the open source software that runs all of the communities, you may point to your own relay and you might build the relay that’s the best relay that has the best discoverability and knows the most about the quality of people and sees where Bitcoin is being transferred — and that’s another quality signal — so I almost wonder if, in a sense, the client is a little bit of like just a strategic entry point into building better relays and building more relays of deeper service to the broader community? I don’t know, that’s some little strawman brainstorm. What’s your thoughts or reaction on that?
Keyan Kousha [34:55]: I think that’s a really interesting point that if Nostr relays are fairly easy to run, many people can run a Nostr relay that has maybe Stacker News built in as the client for it — that seems interesting to me. I think one of the struggles with Nostr right now is that the relays are intentionally dumb: they try to limit what a relay does — that’s great for a number of things, but it’s perhaps a struggle to do certain types of experiences on Nostr, because you can’t have a client doing so much processing on data that they wait a long time. There’s a huge lag, and there’s some clients you use that it’ll just sit there loading for 20 seconds before you actually get anything.
DK: Is that because there’s just not enough relay interconnect?
Keyan Kousha [36:02]: I think it’s just parsing too much data and then putting it into your feed, in some structure. I don’t believe the protocol — as far as I understand it — can indicate to a relay that you want the stream sorted in any particular way, so it just comes in and the client would sort it, is my naive understanding? But yeah the way I always imagined Stacker News getting decentralized was kind of like a Bitcoin mempool, and I think Nostr’s relays look a lot like a mempool, especially once they can discover each other and communicate with each other — rather than just the clients, which is what they do now. I imagine eventually they’ll have relay-to-relay communication that’s baked into the protocol, and so that becomes kind of interesting and I like that. But even in a world where the relays are somewhat isolated and the clients are the only things that connect them, maybe that’s potentially an interesting area, too.
DK [37:21]: My understanding is that today — from a protocol perspective — there’s no relay-to-relay communication, but let’s say you wanted to run the best relay: you could just say my relay is going to also be a client to every other relay I can know about. It’s legit as far as the protocol is concerned — it’s just not baked into it. So who’s doing this today? Is somebody building the best relay in the world?
Keyan Kousha [37:52]: I don’t think there are as many people building relays as there are clients. From what I’ve heard Fiatjaf say, he seems to think that the relay software is good enough and he just wants really good clients, and so that might be true — he seems to be the authoritative source on the thing he invented, but I know there are people who are experimenting with more performant relays in different architectures for storing messages and sending them out, maybe things that are a little more efficient than a naive approach you might make as a first pass. I know just with my brief experimentation with this NIP-05 stuff, you could literally crawl every person and then query that for their relays, and now you have the entire relay graph, eventually — that can definitely happen. And maybe you don’t need explicit relay-to-relay communication.
Max Webster [39:01]: I think two very basic experiments I would love to see and I think will happen very quickly: 1) paid relays. Olaoluwa has already put out on Twitter how to hook up with Aperture and LSATs — I think someone’s gonna do that pretty quickly. And 2) there’s gonna be a lot of experimentation over how do you charge — is it per post? Per kilobyte or whatever? And then the other thing — to your point, Keyan — is: when you have this giant water hose of data, that’s a lot to sort through, and frankly most people are going to want moderation of some kind, and ideally it’s their own chosen moderation — it’s not forced upon them. But I expect to see essentially caching services or content delivery networks arise here as well. Coming back to the search idea: you look at this this site by Brugeman who’s got realsearch.cc and then launched nostr.band — that’s a very primitive search engine, but I’ve been working on this post about Nostr and I use it non-stop and that’s how I kept finding this post. And what I love about this is this blue ocean moment where Google’s not going to be indexing there for a while, so whoever is there and whoever can build up the best map of those relays — whether they run the relays themselves like you were saying DK, you have the best or you have the best whatever caching service — I think that’s probably the key to the search.
DK [40:19]: Yeah I agree because search has a natural tendency to be somewhat of a centralizing product feature, because it kind of needs to know about global state. I think one of the head fakes that a lot of people got tricked on with trying to decentralize social media is thinking that you need the global state awareness that you do need for something like money — to prevent double spends — but you actually don’t need that same threshold for most of publishing and communication. But for search, you actually do need an understanding of global state. If you can build a relay or a caching service that has the best understanding of global state — and also these don’t live at a given URL, so brb.io has got a service that lets you publish it to a particular URL but then that kind of flattens all the underlying knowledge and information about who published what and what are their relationships with other people on this network — all of that signal, Google basically was invented around discovering that signal that was just hidden in plain sight, which is like backlinks and all the text used in the anchor backlinks. I think all of that information is going to be invisible to Google for quite a long time — because Google’s not going to pay attention to this — but it’s actually super valuable for trying to understand what is the different quality of information. And of course you have sats baked into some of that. The quality of information and the quality of how you do search just seems like it’s ready to be changed. I actually don’t find ChatGPT that exciting from a search perspective, compared to this stuff.
Max Webster [42:06]: And also in addition to the value ranking, again: your web of trust ideas — imagine if that gets applied to the broader Nostr identity system. Like, Who is a trusted Nostr public key versus not? Obviously everyone can decide for themselves around their own software, but I think that’s really important. I’d like to throw out one other crazy idea, even though I love the search stuff — I think there’s a ton to explore there — and DK I really love the idea of like, Hey could Stacker News be the almost Wiki-builder for your community up top? I think that’s really cool — that’s what came to mind to me. One other thing though that I can’t stop thinking about is what you said, Keyan, on this this podcast with Jack and NVK: he keeps saying that the first wave is that when people exit, they’re running away from something. And so as an investor when I think about this I think: in this first wave a company, yeah there’s gonna be some winners, but probably this first wave is not going to be anything new — it’s going to be like taking the newspaper and putting it online versus building the completely new thing like Twitter.
DK: Skeuomorphism.
Max Webster [43:06]: Yeah skeuomorphism, exactly. Now, I literally couldn’t sleep last night — I was writing all of this down — it’s both something that excites me but also terrifies me, because I’ve invested in a lot of these traditional Lightning payments or remittance-type companies, I would say. I think they’re going to be successful — they have a lot of opportunity — but it just struck me as I was going to sleep last night: I was like, Holy shit! What if none of those companies take off because no one uses a “payments app” for remittances in the future? Or a “commerce app” for commerce in the future? What if the killer app is maybe a super app like a decentralized WeChat, or maybe it’s a different kind of social media app? But like, I’m in my social media app and I send money back to Mexico or the Philippines or whatever just as a tweet — and that’s it! You never interact with [specialized apps]. And obviously you still need a way to get the cash and there’s other systems there, or if I’m selling something there is no more Craigslist anymore and I’m just on Stacker News — I’m on the sub where somebody’s like, Yep here’s my service I’ll sell, or, Here’s my thing, and people just pay everything directly from those communication channels. So the thing that I can’t get out of my head is like, Oh man, what if all the stuff that we’re backing for traditional payments is not it, and it’s just gonna be this next wave of social with payments tied in from the base layer?
Keyan Kousha [44:23]: I mean I think that’s how payments evolved outside of the United States, it seems — there’s clearly a path in human desires that leads to that.
Max Webster: Stacker News super app!
DK [44:42]: Yeah the super app idea is kind of interesting. I think the concept of a super app works in a centralized world, but what is a super app is almost like a network service instead of an application service. I still have a tough time: I try so hard to remember this, but when I’m on Damus talking to somebody else, they’re not necessarily using Damus. And as much as I understand it, I still don’t feel it — I have to remind myself constantly, because the architecture is just so new. And so: What is a super app in this world? Is it that caching service? Is it a relay service? Is it a list of Lightning addresses — a directory of people that you can reliably access? There’s probably a bunch of different flavors of what that might look like, but that’s also why it’s such a fun place to be spending time, because it’s just so Wild West! It’s like, Nobody knows. So can we talk a little bit about the NIP-05 integration? I know we were DM’ing about it a few weeks ago and then you were like, Yeah I’ve got a few other priorities but we’ll get to it. And then I think you’ve finally opened it up and said, Oh actually this is pretty easy! I think you told me you did it in a half hour or something — is that right?
Keyan Kousha [46:14]: Yeah, so I think what Nostr is excellent at — and all these other protocols just get wrong for some reason right off the bat — is developer UX: it’s just so dead simple. I mean you could read all the Nostr Improvement Possibilities in half a day, maybe a morning, and have a fairly good grip of what it’s capable of right now. Everything is very high layer web stuff: WebSockets — pretty straightforward, tons of libraries interoperate with them, the wire formatting is JSON. As far as NIP-05, the hard part was mostly recording everything in a database, but everything with regard to the NIP-05 stuff was just displaying a JSON file for other people to read, which is like a very old object syntax.
DK: So, very simple. So once you started NIP-05, that was a pretty easy thing to do once you sort of got your head around those bits. Have you explored other NIPs or other things that catch your attention as potentially interesting? You’ve got your first Nostr-Stacker News integration — is there anything else that grabs your attention of what might be next? Or what kinds of things you might explore and how do you sequence those?
Keyan Kousha [48:17]: Yeah so if I were exploring Nostr as a developer who maybe was only exploring Nostr, I’d just begin building both a relay and a client, is probably what I would do. Because then I would be certain that I understand everything entirely. As far as what is next for Stacker News, it’s probably spinning up an existing relay and interacting with it and seeing: What are the limitations of it? How can I make it compatible with what users want? That’s probably what I’ll end up doing. But yeah the other NIPs are very client and relay-specific. I mean I could do things like connecting to the user’s relays and maybe displaying for them Nostr and Stacker News — we could make Stacker News sort of a Nostr client. There are there are things that can be done like that, too, but I think the first move would probably just be: put Stacker News’s content on a relay, just because one of the struggles I have is not creating too much surface area on the product — it can be a little bit of a struggle for a user’s experience when they have too much choice in terms of where to go on a product. So I think I would just create it as an adjacent thing and see how that goes, and play it by ear.
Max Webster [50:04]: Keyan, have you thought at all or have you seen William’s proposal — I don’t know if he’s actually put it out as a NIP yet or not, but — to have Lightning tipping native to Nostr posts? And if so, to me one of the very big things a lot of users would love is a way for them just to receive tips from outside clients.
Keyan Kousha [50:22]: Yeah I haven’t seen the proposal but I’ve heard him talk about it, and I know that’s why he created Damus to begin with. And so I’m really curious about how he plans to make that work and what’s required on either the Lightning or or Nostr side.
DK [50:43]: I haven’t listened to the whole NVK pod yet — I think I’m halfway through a couple of hours — but I think Fiatjaf was talking about liking to keep Nostr feeling a little more agnostic to Bitcoin. A lot of Bitcoiners are excited because there’s a lot of common principles and themes that run through it, but ultimately Nostr maybe could exist without Bitcoin — we don’t know if it can flourish long-term without some underlying monetary solution, but — I think there’s some hesitation to get the protocol too coupled to Bitcoin too early. Though I did see something from William this morning on Nostr where he was publishing something that was a Lightning payment that looked like it was not Damus-specific but maybe more Nostr — have you guys seen that? Or did you see that post?
Max Webster [51:42]: I didn’t look too closely into it. I do remember maybe two or three days ago he posted a note about exploring it: he mentioned, I might publish this as a NIP or as an LNURL spec. So maybe there’s a way to do this from the Lightning side instead of the Nostr side, to be more protocol-agnostic.
DK [52:00]: Is there currently a roadmap or changes? Has anybody heard from Fiatjaf of what does he plan to do next? Is he gonna add additions to the protocol or just try to evangelize and get more new clients built? Or where his head is?
Keyan Kousha [52:19]: I’m not too sure. I think he’s working actively on a web browser extension right now, is what I thought I saw him doing. But I imagine he’s mostly fielding lots of inbound and trying to make sure the community goes in a direction he thinks is the right path, is probably what I imagine he’s doing. But he has a full-time job at Zebedee, unless that’s changed.
Max Webster [52:47]: And also just listening to that guy I get the impression he’s been prolific — it’s not like he came out of nowhere. He was responsible for lntxbot, he was one of the primary authors with Ben Arc on all the early versions of LNbits, which by the way is doing a lot of cool stuff with Nostr now. So I get the impression this guy just is almost like a Satoshi — throws out this beautiful thing, and then bounces to the next problem. So, who knows? Maybe he’ll give us something brand new?
Keyan Kousha [53:20]: That’d be awesome. He’s definitely super-prolific: I recall when we had him on for an AMA, Jimmy Song’s question to him was: How do you choose what to work on? Like, acknowledging that he is working on so many different things simultaneously. He also did a lot of LNURL work — a lot of things that we on some level take for granted in the Bitcoin space, they’re so ubiquitous.
DK [53:45]: And so we’ve got some of our thoughts and brainstorm on Nostr out now. Any other directional things that come to mind for you around stuff you want to explore? Or do you feel you have in mind your current roadmap or current explorations or prototypes you want to do? How do you think about splitting your time?
Keyan Kousha [54:08]: Yeah I think my time is going to be really hard to split right now, so I’ll probably have to hire — because I had this conversation with my wife last night and I’m like, There’s a Plan A with Stacker News: I keep making it better and it’s on its path. I refine it, I make it better the way it is, and there’s some interesting things I plan to do with sub-communities and other types of formats and stuff like that, but it doesn’t feel all that exciting to me as I said before. Risk is what really gets me going. But then there’s a Plan B: one of my intentions to begin with was to make Stacker News decentralized — that’s something I’ve been talking about since I launched it — and so maybe this is the moment to begin actually focusing on that. But it’s tricky because I also have a strong connection to my existing users and I love the product as it is already and I use it all the time. So I don’t intend for it to go away, but it’s like my baby and now I’m gonna be away from home most of the day or something — how do I manage that with a migration to a new architecture, possibly? So that’s where I’m at right now, and that excites me a lot thinking about that because I might otherwise be kind of bored, to be honest.
DK [56:03]: Yeah I think it’s nice that you’ve said you are that risk-seeking and you don’t want to be bored, because a lot of technology changes under our feet very frequently, and so listening to your own voice and your own energy — what you have energy for — is a great way to find the asymmetries when they pop up. That’s not to say that this one that we’ve talked about is exactly the right one — it looks promising and interesting, but there’s still enough risk that it can be very exciting for a long time.
Max Webster [56:39]: And for what it’s worth, just to echo DK there — I’m reminding myself right now putting on the investor hat as well — I haven’t been this energized in a long time. This feels like to me when I was really learning about Lightning — and over the last few years it’s been really fun and it’s like, Okay I get it: I invest in a Lightning company in Africa, one in India — it’s awesome, they’re great and they’re gonna do a lot of awesome things, but it’s kind of like, All right, where’s the new thing? And I feel like this is such a foundational shift. There’s so much potential to rebuild basic Internet services. All the Web3 crap we’ve heard about for years and years — those things can actually potentially be done now, but in a way that actually scales. So as DK said, technology can shift very quickly, and being able to move with it — obviously it’s a dance because you don’t want to go in too early and maybe it’s the first hype wave and you blow up. But at the same time, going from less than 1,000 users to 250 000 users — that’s not arguable. So one thing that I would be thinking about a lot is: How do we tap and bring some more of those new 250,000 users into this Stacker News ecosystem in some way, shape, or form? Maybe that’s its own subreddit or substackernews around Nostr, or maybe it’s by turning this into a client, I don’t know? But that seems like a first good step is: if you’ve got orders of magnitude of new, interesting people, how do we bring them into the discussion? And then from there, Who knows? One question I did have on that — because I don’t have a good answer this — is: If we saw that much growth — and Damus is capped at 10K users because of TestFlight — where are these people coming on? Like, I created keys in Alby — that’s one place. I know some of the other clients will help you get on. But if Damus is the famous one and that’s 10K users — and there’s 250k new pubkeys — where’s the other 240k coming from? That seems like a good question to answer as we think about where to steer the ship.
DK: I would guess it’s got to be the web, right?
Keyan Kousha [58:36]: Yeah I think a lot of the clients on the web, when you join them it immediately creates a public/private key pair for you. I don’t how these measurements are made, but I know someone who was spamming Nostr as an experiment and they were generating a new pubkey for every message in addition to a new TOR address or onion address, and so there’s probably some spoofing of that going on. But there’s so many clients — as a developer, you’re developing on these things — you’re probably generating a bunch of them as you go, but those numbers are undeniably large.
Max Webster: You’d need to spoof half of it.
Keyan Kousha: You take an order of magnitude away, it’s still a very big number.
DK [59:28]: I wouldn’t worry too much about the numbers as much as the fact that if you go on and look at the fire hose, there’s real new people showing up all the time — you can talk to them. They’re interested to explore. It’s not just spam — people are coming on because they’re genuinely curious. To me, that’s what I really get excited about is just seeing real people show up being curious about it. The actual numbers are hard to really pin down, I’d say, but the human energy around it is palpable.
Max Webster [59:58]: One thing just as we’re spitballing and kind of congealing some of these ideas: DK, you made a great point to me maybe a week or so ago, which is: How does Nostr expand outside of just Bitcoin? Obviously we all love Bitcoin — it’s been a great strategic decision for Stacker News to be the Bitcoin signal spot. Having said that, if we go mainstream there’s only so many people that care about Austrian economics blah blah blah whatever — as fascinating as we all find it to be. But DK you made a great point to me, which is that within the Bitcoiners we’re not just Bitcoiners — we have other interest there’s: Alpine skiers in there, or we’re both extremely interested in this artificial intelligence wave that’s happening, or in my case I’m super interested in music and Hip-Hop. I wonder if there’s some way — and maybe this ties in to you Keyan and you had the referral system you were playing with — where: Is this a moment somehow that we get some of these Bitcoiners to say, Hey can you start onboarding people into other communities? Maybe you create Stacker News, the Wiki — maybe you help the first couple or whatever — where Bitcoin is not the focus, but it’s the other thing that you’re super passionate about. I would love a hip-hop subreddit or substackernews.
Keyan Kousha [1:01:07]: We call them subs because there’s so much of this confusion. It’s the shared pivot point between all the different names. Yeah other communities are right there — I think we’re pretty much ready for them. There’s been a debate over the last couple months about it and we just wanted to really make sure we were ready and had our mind in the right place when we started doing it, and wouldn’t just spread our users too thin across too many sub-communities and create a ghost town, of a kind. So, Nostr would be an interesting one to start with. There are adjacent subtopics like programming generally, or some of these other adjacent areas. But Nostr could be interesting to start. I kind of like the idea of users deciding, so maybe we’ll have some back and forth with them.
DK [1:02:06]: What’s the current vibe from users on the site? Are people talking about Nostr? I’ve been involved in a few threads, but I don’t think I have as much an overall purview feel for what’s really going on.
Keyan Kousha [1:02:20]: It seems like any given day, there’s probably two or three top posts that are Nostr related. We have one top post that’s Nostr related today, which is kind of small, but if I look over the last week or so, I would guess there’s far more. Yeah six of the top 21 posts in the last week were all Nostr posts, so lots of overlap.
DK [1:03:02]: Nice. Another idea that I was gonna float is: I think you’re right to be cautious about launching a second Stacker News, because you don’t want to create that ghost town feel, and I think just having one vibrant town is even itself challenging. So you’ve been a good steward and owner and instigator there, Keyan. I think without your pushing that wouldn’t exist, and so it’s hard to imagine exactly who in the gardening vertical or Alpine skiing vertical is going to take so much ownership. But I do wonder if we were to think about the problem as maybe more in service of a creator who would be their own instigation point. So one example that comes to mind — I don’t know if you follow Anthony Pompliano and the Pomp Podcast and stuff, but — I think about a week or so ago he finally gave up with advertising and he says, I’m not going to run ads anymore. And part of it was maybe he involved with FTX or he had some fallout with some —
Max Webster: There was a couple. Block-Fi.
Keyan Kousha: All of his sponsors.
Max Webster: Exactly. That’s the joke is like: whoever they sponsor goes under.
DK [1:04:24]: But given that state of the world and he doesn’t really know what’s next, maybe he would want to run the Pomp Stacker News sub where it’s still pretty Bitcoin-flavored but maybe he curates who’s allowed to come on and he gives more access to special content that he creates only there, and the whole community is basically: instead of people coming to earn sats, they come to pay him sats for his instigation and creativity work. Then he has the responsibility to make it not a ghost town and we don’t have to find the Alpine skier Bitcoiner overlap to take the ownership that you have.
Keyan Kousha [1:05:09]: Yeah that’s a good point is finding someone like that. Another way I’ve thought about this is: when we thought about these subs, they would be pay to create them. Small in dollar terms — maybe 10,000 sats a month or something — and you would create them in the hopes of earning that back from the usage on it. But that would ensure that they have skin in the game and they actually are really invested in the topic and seeing a community thrive. And then they also wouldn’t get the burnout that you get on other types of platforms where there is no real incentive to be that person in the community. And so that’s maybe another way is actually we just open it up entirely for people, for them to choose. And because we don’t have to really open it up — and floodgates open up — it could just cost a little bit of money to do it.
DK [1:06:09]: And you can also play with the incentives of: should 80% go to the instigator or creator and 20% to the community? There’s probably a bunch of knobs and dials there that would attract different types of people.
Keyan Kousha: Yeah, give them the options.
DK: There’s a service called circle.so, which is not decentralized, not a Nostr client. I don’t know if you’ve seen that but it’s a Slack or a Discord-type thing, but instead of it being more team or community-oriented, it’s really creator-centric. There’s a lot of good ideas that they have explored over the years around how a creator can organize people interested in the creativity and the work that they put out, to organize for special access, community participation, special content. Of course it’s not in any of this world that we’re talking about, but maybe some of the ideas that they’ve explored and learned through would be the things that would be interesting to people in these communities that we’re talking about.
Max Webster [1:07:12]: I’d be interested to see — and related to all this — the Podcasting 2.0, Value for Value stuff. I think that’s actually grown quite a lot over the last year. I think now there’s 11k-12K podcasts that are Value for Value-enabled. I think that’s gonna explode in the coming year. I’ve talked with Oscar and a couple of people about this: maybe it ends up getting tied in Nostr somehow, where you actually can store the audio data. Obviously this would be more expensive, but it’s something interesting. That’d be a sub that would be pretty cool, too.
Keyan Kousha [1:07:40]: Yeah what’s nice about a podcast sub would be that there’s content that we could pull pretty much automatically, and so that has no possibility of being a ghost town. That’s something we’ve thought about — that was one of the first ideas that got floated around for subs, was Value for Value. Oscar and I have even had conversations around how we might be able to do something together there. I’ve been really impressed by the growth in the podcasting space in general. I’ve just been too in my own head to listen to podcasts recently, but I just went and listened to that NVK podcast and then I’m back in the Fountain client for the first time in six months or something and I’m like, Wow! This thing is really polished up — it’s very nice. A bunch of comments and boosts and stuff, so really impressed with everything that’s going on there.
Max Webster [1:08:34]: And again, this all ties back into Stacker News being decentralized and Nostr being decentralized. If the podcast RSS feeds can somehow coordinate with Nostr, that’s an easy way to do interop messaging, which is still something that’s hood today — it sort of works. But again, that’s the beauty of it is: from my Fountain app — I commented on the Value for Value NVK podcast before this one — it was so cool! I sent them my little boostagram, everyone was sending me sats back saying, Oh this is great, glad you liked it so much! People I don’t even know — people that ran the podcast. And then if I went to Breez, I could still read all of that — it’s all interoperable. So again, being able to interact with one global feed of information from a lot of different places — I think we’re just seeing the tip of the iceberg.
Keyan Kousha: Yeah I mean that’s basically the web, if you think about it. It has a very similar type of vibe, I guess.
Max Webster: Fair enough!
DK: Well I know we’re over time — we only really booked an hour ,but we could I’m sure keep going for another two or three hours but I want to respect everybody’s time. So I’ll wrap with a thank you to Keyan for coming and sharing your thoughts and being pretty open-minded and open-ended about our strawmans and brainstorms, and it’s great to hear from a practitioner like you who is in it and figuring out how to balance and weigh and manage these trade-offs. I think it really helps me to think better about the space and I think it hopefully will help other people building to think about those trade-offs too. So, always fun whenever we get a chance to do these. I hope we can do another one soon. And as your thinking is evolving, don’t hesitate to ping us of course.
Keyan Kousha: Yeah, of course. Thank you so much for having me. It was a lot of fun. I learn so much from you guys — I should talk to you more. I don’t know why I don’t, but you guys are great.
Max Webster: Likewise. Yeah, this is a lot of fun.
DK: Awesome.
Outro: Thanks for joining us for this conversation with Keyan and Max. I’m excited to keep having more conversations about Nostr and I hope you can join me on the journey as I keep learning more about this emerging technology. Sign up for my newsletter at www.curiousdk.com to follow along.
