@ YakiHonne

The YakiHonne community is excited about Nostrasia, and they have diligently transcribed Nostrasia's inspiring speeches. Additional speech transcripts will be made available over time. The Japanese and Spanish versions are initially translated and proofread by community members with the help of AI tools. All YakiHonne users are invited to join in the review process. Those who successfully complete the review will be granted a special reward of 3000 Sats. To get started, simply reach out to us (Comment here, DM, or TG) to sign up, and the first person to contact and submit their review will be the lucky recipient of the reward. And hey, if you're up for it, we'd love to have these speeches translated into even more languages. Join us!

🌟中文版: 客户端审查制度和未来的路 🌟日本語: クライアントの検閲と転送パス 🌟Español: Censura de clientes y caminos a seguir

Hello and welcome back to the bitcoin, not just kidding. So today we're go to chat about censorship and the pros and cons of it. I know it sounds like there are only cons, but it's not that simple, so we have Mr. Will from Damus, hello. Milan from Primal. How's it going? Guys are we going to have Vitor from Amethtys in video or something or no Oh, you're looking for, okay, not just get someone else. Yeah, I'm sure we can pick one more client from the crowd, just Kidding up somebody and bring it over. Marty, we need more people. Karen and Marty Karen, are you around? Nope, okay, yeah, good Marty. So we got right by Victor. Victor was going to video link CS and be behind us like fiat did so why don't we get started and sure they can always join us. Midway there is enough on this topic so. First why don't we start with just sort of like your general view on where do you think censorship is useful versus its just a horrible thing? Unfortunately, like we don't have a better word for filtering and censorship is a kind of filtering. So how? How do you go about that? Will?

Yeah, i mean, maybe I gave some background into like, you know what even made me start working on Damus because you know the main reason was seeing all the censorship on all these platforms right. Seeing these centralised institutions having so much power over people that they can easily just completely erase them from the internet, you know they build their following. They build this huge social graph and they spend a lot of time online like and then at a flip of a switch, they can just erase them and all that data. And I realise like, that's just, it's just not, it's not good like it's, it's horrible, so. Yeah, so like I first looked at you know mastodon as a potential solution of this problem and I realised it's like it was worse censorship, resistance was like people were getting banned left and right. It's easily for your count to get your race. So I guess I'm not really answering your Question. But this is. I mean, we all know, mastodon sucks. I mean, like. I was the benevolent dictator on that and I have to cancel some people and i felt I'm like. Oh shit, should I talk bad about cold card because I'm like am I going to get banned? So it's like you have to think about those things. It's like self-censorship, especially when you have someone running in the in incident.

So I felt like nostr was the first time where we have a solution where it's harder for people to be censored. So yeahs. I don't know if yeah didn't answer your question. That's the beginning. Mr. Millian, well. I will echo will's words here. I think the reason the main reason why many of us develop are develop. Developing an nostr is to avoid censorship and to go around it.

Probably the protocol might not even be here if we didn't experience so much censorship out there in the recent years and certainly it wouldn't get the level of passion that it's getting both from developers like Will and myself and many others here as well as the users like this early stage of an nostr where we have like a fairly small user base but very passionate user base,I think is' a product of what actually happened in the last few years. It was a wake-up call for many of us to just like to see how things work Currently and then we definitely need an alternative, the whole notion that there is a handful of companies, maybe one company that owns the kind of global town square.

Yeah, it's just insane, yeah, and I find it interesting because you know, this is why we started our clients, but in some sense it's becoming, it's becoming just like everyone's just used to, and it's expected. And now we're more focused on, you know, client features. And it's becoming like an invisible thing. And it's kind of just accepted, which is interesting. So I think the the counter argument to like our so like we're all fairly self-selecting, you know, free speech, absolutist and all this stuff. But the counter argument to that is that like, you know, there's this great quote that I can't remember who said it, but it essential like people don't want freedom. People want a better sell right. Freedom is a scary thing. There's a lot of responsibility and there's other issues with that too right.

So. A lot of people at least in my view like Twitter or like Facebook or whatever because they can have sort of like a safety space where they can dunk on the opposing view but they don't need to interact with the opposing view when they don't want to because they can just block and they can do whatever they want right so. First is one is how can we provide people also freedom to sensors or what they don't want to see in a way where it doesn't feel like there is some marionette sort of like benevolent dictator on the platform. Yeah, this is like the root of everything which is you know how much should clients be censoring aka like some people call it moderation but. As a client you have a huge responsibility when do you make that decision? Because now you're making a moderation decision on behalf of everyone using your client and yeah, it might remove some spam but it actually might also remove some legit stuff.

You know in some cases like you with, i think the IRS client originally had this web of trust model which is like a really cool way of finding spam. You only see messages from within your friend group which is just fine. But its it's a moderation decision in the sense that you're going to be missing lots of messages. Those affect people are effectively shadow banned right and they might not not necessarily be spam. So I feel like as client authors we have huge responsibility to and when we're making these decisions we need to be very transparent about them because like it's really a really copy experience when they're like, wow, I'm actually missing messages that I'm expecting because they're making some moderation decision on my behalf. So yeah. I mean it's very easy for things to get lost. What's your approach to handling some of these questions?

So when the clients shouldn't be censoring period clients shouldn't be in the business of selfensoring, they should be in the business of offering tools to people that people can choose to use or not. So when we all started kind of holding our keys to our online speech with nostr. I think that's the kind of the magical moment where the balance power shifted Forever and we're just in the early stages of that and we as client developers, probably we can make a bunch of decisions about how we're going to run our clients. It gets even more complicated when we're not only dealing with client apps, but for those companies or organisations that also run services. That's like the next kind of level of complexity. Anyone who is running a service is introduce by service I mean things like an indexing service that offers full text search of the Nosster network or the media hosting service or the caching service for acceleration services. Those types of things they are all running those types of services introduces some centralising factors to the network definitely, and people running those services need to, of course be aware of that. And in my view, the correct way to handle this sonostr is to make it easy.

Stood up as possible so that you don't have one missing service but closedsource toctify that one of this services is dangerous. Many of this services dozens and dozen smoking source software is I mean, so is BAL. This centralization is more efficient by DEF finishing and then for decentralizing. The capital has efficiency. Maybe discoverability would suffer a little and they have issues, but anostr you have can en speech. Cancel your key pair but they can prevent your key pair from participating in their system. We just say that the relays is also their right to not have your stuff right like they may choose not to host your data. But I guess the value here is that you have alternatives and you Can go elsewhere. Yeah I mean relay should be banning as many, many people as they want because like you know it's in the design of the system where relays can make those decisions.

So it's not a big of the deal when the relays do it, but it's when the clients do it because clients are naturally centralising things within nostr because you know you know a lot of people use one client like a lot Of people use domas right. So I feel like I have a lots of responsibility to make sure at the base layer that there's no censorship happening. Even I think that actually happened recently when I was designing a new feature it made me realise I'm like I don't even know if I should turn this on by default. So I added this feature where you can filter NSW Tags hashtags When you're falling hashtags and things like that. And I realised that if I were to turn this on by default which might be like a reasonable default, then that would be the first time where I'm actually like making a moderation decision on behalf of someone. And a lot of people don't go into settings to like turn it off. They might not even know those settings there, so I was like I don't even know, so eventually made decision to not enable it by default. But now it's like you. You have a first time user who like is scrolling the hashtags and they're getting porn in their feed, so it's one of those really as a client author, I'm like I don't know what the right thing to do there is because I don't want. Yeah, how do you guys feel about. Algorithm like fiedalgo right because that's where it gets dicey right. I mean the whole point of the feed Algo is for you to filter and rate and change the way that the content is presented. But as we now know like how Google is and people, it becomes the internet for people, it becomes the only thing they see. So how are you guys see it approaching? Sort of like now that's still small and how do you see this evolving? I like.

How do you handle the algo choices and algo transparencies and all that stuff? I think that Algos will be, i don't want to say, absolutely necessary to use nostr at a certain point. After we get to a very large amount of content, a lot of users, maybe even AI generated content and stuff like that. But in my view that should always be Deliberately selected by the user. So let's say if you're. Probably the correct choice at least for Primal currently is that the default feed doesn't have any algorithmic interventions.

The default feed is just the chronological feed of your follows, and then we'll have, we already have, you know, the trending feed, which is our opinion. Of what's relevant at any a given point. So that's when things get interesting and we noticed that people seem to care a lot about what trend, but but we get 2 two dumbs on that one too. But I guess like my only pushback on that is, you know, if the goal is to sort of onboard new users right, we can't just like dump like the main feet on them right. We have to do some kind of, you know, activity feed sort of like there's going to be a bunch of different feeds and hopefully offered by a bunch of different players, and the user should be able to select through the curated list of feeds. We're kind of starting to work on that at Primal, where things are a little bit more interesting than just like chronological feed of your follows versus maybe what you Primal thinks it's relevant in terms of trending stuff. We're starting to come up with different algos for a bunch of different scenarios, which we think are going to be interesting. But what's important is that every ALGOL is open sourced. So that it's perfectly transparent, how we're coming up with this, and if there are any filter list lists. Those lists should also be transparent and also If not opt- in but at least like opt out. So the user should have the ability to select the types of Algos that they want to see in their feeds or not. Yeah I mean a lot of people think I'm like Antialgo because I have like a chronological feed maxi, but it's not really true. My issue is like I'm against one algol like you know the default ALGOL like you open tiktok and it's just like you scroll mindless.

You know I think that is causing lots of negative effects on society and humanities. But I think nostrs uniquely set up right now to do it much better, which is. I think next year is going to be the year of Argos on nostr. In terms of just you, Damus will eventually have lists different types of ALGOL. We're starting to see the training feed and stuff in Primal. I think even snort. When I open up Snort, there's like a default trending algo now. Yeah. I think it's coming from nostr band, but I'm not sure. So it's like weren't this? We're set up perfectly to start exploring this space and Nasha uniquely is going to do it really well. Just because we have different clients trying different things. There's go to be different ALGOL feed, so I'm really excited for next year in terms of what we do without. That's actually an interesting point, which is. When it comes to content discovery so the notion of offering a content Discovery feature again introduces some centralising factors into the story right. So if you're a nostr client, let's say let's take search. Search is a very good example because everybody agrees that search is useful right, whereas some people might disagree that trending feeds are useful so that's fine. So if you're an nostr client and when it comes to search, you have basically three options you can not offer. You know. Global Search. You can say running indexes introduces centralizing effects during network. I don't want to do that and make a principal choice to not offer this feature. I respect that second choice is to stand up your own indexer and offer the feature and the third choice is to use someone else's indexer. So I mean data. I mean, we have this massive liquidity pool of data. Right. I mean if you're well connected to relays right as a user. You could almost say like you have this universe of almost everything that's on nostr would be available to you except for the stuff that doesn't want to be seen right so like private relays or things like that so. The data is there it is up to. I guess the user to pick a client that will sort of shape the data in a way. The day when I have. I'm curious to get you a take will on the 4th.

Option yeah. I mean so I've been thinking about this a lot. The way that I'm starting to view at least Damus at least my client with nostr DB and things like that is you know. I really want to. I want these features to be in clients so this is why I've been kind of building this database from and so my view is you know you have your note is like or your client is like has a certain view of the network.

Yeah, it's not a global view so we don't have global search but I think you can still get pretty far with just like your view of the network. Search locally on your device. So I think that's gonna open up a lot of cool anyway. I just thought. I agree that's actually a good good take and can't wait to try that. Yeah, it's gona be fun. What about marketplaces for for Feed. Algols and for filters and things like that? Is this like something that's on your radars. It's just something you guys are developing now because I would love to have like you. Here's a bunch of cool things that you could have you like. Maybe shows More cats maybe like yeah. So actually. Primo was the main motivation for creating Notre script because. I noticed their relay had custom like non-pet things you're sending back and forth over the web socket. Well. I was like How could I tap into that and Damus. If if someone wanted to write like a plugin you can Plug, we chat it a little bit like DM. Yeah, even though it's not a stable interface, but it was still cool like, hey, imagine if you could write a small notriscript that talked to the primal relay using their language, and then you could just simply broadcast that plugin as like a note to the network and so the marketplace would just simply you just fetch down all these plugin notes which are just like they could even be embedded in the note. The actual script could be embedded in the note and then you can just install it into Damus. And then now you have a primal training feed in Damus. So I wanted to make it so like you know. I don't have to hard code your relays and the Damus. So I think I think I'm really excited to start doing more with nostrscript. It's pretty powerful. I'm not really doing anything with it yet, but except for unlocking apps, yeah. I know you, you guys are working more on this sorry question.

Yes we are excited about the idea of marketplace for feeds and we're excited about it from both directions. The idea of creating our own feeds that are going to be available in other clients as well as of course making other non-primal feeds available in primal clients. I think it's going to be massive and it's one of those examples where it lifts the whole ecosystem. We can kind of help each other out that way. I think that these things get very like interesting and dangerous right as we move just past only social stuff because you're going to have marketplaces you're going to have in all the dms you're going to have. You know it could be blog posts. It could be whatever so like you imagine when news organizations are posting their long form articles like you The New York Times of such and know you're going to have clients that maybe have state sanctions on them right. So I don know, like Domas China might not be willing to show you know what some newspaper that is against the current regime in some other country, you in that country, we'll learn how exactly how censorship resistant we area. I'm waiting for the day when it's like, you know, some government is going to come to me and say you have to ban this pubkey or this group of terrorist organisations because it'll be really sad where I'm gonna have to. I mean, it's definitely possible where have to hard code centres. I just don't think it's that effective though, because you could just create new keys so easily. I mean Pablo is already censoring BRB.IO, he blocked from NDK. But as a couple of days ago we did two there going oka, so nobody's handling yours. So no, we've just been handling them for so long. We don't have to handle them anymore. So you know, like how? How do we incentivize CRE incentives where all like state actors sort of kind of like give up right because it's guacamole? Anyone can see the same content elsewhere right like that is not true for twit or Facebook right like the content that is there is' stuck in there with nostr. Maybe on the app store in China is not available, but they can go see the head on the web right. I think there's not many like Central places and this is the whole point of building a decentralised network is. There's no pressure points where they can go to and be effective like even you know China or the CCP They banned a lot of the bootstrap relays within Damus, but then I just updated it so that you can change the bootstrapper relays and so the next time you go so it's going to be like this cat mouse game if they want to try to do it. But it's going to be very hard for governments as censor nostr.

Yeah. I think they'll definitely try. If we are successful or when nostr takes off in a major way in terms of users and usage, we will get pressured. If we're successful, we will get pressured and the only way to kind of deal with that Is to have as many apps as many different form factors in platforms running all kinds, as many different possible sets of tradeoffs, as many relays as many caching services and indexing services, all of them based on open source software. Now that's hard to stop. Yeah, it's like there's a reason why email isn't censored. I mean, actually, yeah, email, it complicated. It's complicated, but you know, but it's pretty free. I mean. Yeah, it, it's very hard to really, truly cut it off. It's just it gets tricky. Yeah, i mean, it got a little bit centralized, but for the most part, it's you can. You know, i run my own email server and it works. I'm kind of at my own. Little Independent noted. I wouldn't recommend running an email server because it's a pain in the ass. But yeah, i mean once you build on a protocol, it's just it becomes exponentially harder for them to shut. There's no central place to shut it down, so. Very cool so just like sort of looking forward right. What things do you guys want to see like built that could help clients and relays. And you know indexes sort of coordinate around these things.

I mean so there's one big risk at least especially with like the model that Damus adopts which is you know this relay pool model. So this is the dumber and simpler versions of the gossip model. So there is a risk that's like where there could be very few relays and you know it might just take like one government say all these are now illegal and so all of a sudden you nowhere to connect to. So I think the gossip model is something I really want to get into next year in Damus just to more censorship resistance. Then then basically you could run your own relay and Damus would know to read and write to those relays for each user and that would help a lot for decentralisation censorship resistance. So I definitely at least want to have it as an option you can turn on. But there's many different ways to fetch nose from. But I think the gossip model is something I want to do more.

Exactly the same answer here. So we'll also implement the gossip model probably after the version one is out. So this is more kind of Primal Two. This has been part of our vision from the beginning, so when we launched Primal back in March. I wrote this blog post that nobody read where kind of outlined the roadmap where we're starting with a you in four different steps. We're going to get from this model where we kind of read from our caching service to the model where the ability to work with the caching service is just a competency of the client. But also we're reading from relays directly and specifically implementing gossip. So yeah, this is going to you.

So you know the two of you are known actors right. So you guys are capture points and you know. How how do you see you the kind of energy you're going to have to put in into having some moderation So you guys don't end up in jail right. So there is the unsavory pictures. There is the stuff that it's illegal jurisdiction you're in, and we all know that I mean. Active moderation is an insane job like I mean it's super hard, its costs a lot of money. You see the guys for Facebook that like jump off buildings after doing a day of it.

I mean I'm not a lawyer, so I don't know how effective my theory around this is. But you know I'm basically positioning Damus as a browser as a client. Its like a browser in some sense, so you know Damus doesn't host any images we don't even have like an image cache or anything. So you know a lot of the issues that these centralized services What they run into is that they're hosting the images. And so you know if they need to be taken down, you actually have to go to Twitter and like take down that image where it's like now I just go to the Nashi pi guide can you take that image down? But that's the only kind of the centralizing risk is like I'm pushing a lot of the liability to people who want to do those things. But for the most part Damus it just like we're just plain text.

So more than just like avoiding jail, which is definitely a good thing. Good go, yeah, you're not winning if you're in jail. Beyond that Certain types of filtering can be just useful services, As long as they are opt-in and selectable by the user. So we're planning to implement all kinds of filtering algorithms that people can choose to opt in or opt out of as a service to our users. And I think this as the network grows, this will be more and more useful. And hopefully those types of filters that are fully transparent again based on open source software, so everything by the way that we do it. Primal is open source under the MIT License.

Hopefully those filters are good enough to also keep us out of jail.

I actually want to talk about something really quick about images in general. I don't think images should be censorship resistant because you get into these situations where you accidentally post an image and you realize like you docks your address or something on a photo and then you're like Well, you definitely want that to be censored right. So if you use like nip whatever and try to get upload, get that image to like ten different relays. Now it's much harder to take down the image I don. So I think it's good that images insensible in some sense, it's like.

That's the trade-off that I'm willing to make just for know, it's common reasons. None of these are simple black and white issues. Once you drill into them, i mean, you know. nostr doesn't have a native way of hosting binary data right. So like it's really unsolved problem to begin with, like, where do the pictures go? Where does the video wherever you want them to go? Yes that's right. I mean. I guess we could. That's the other stuff we can do. Bay 64 and just put them in kind S text. But don't do that. Please you listening, hacker out there, dump too fast. Bwy 64. All the things.

So I guess who's doing like a good job at handling this fine balance in the early days where we don't have all the tools yet to have not a lot of knowledge yet, but like who's doing a good job out there? And what are some examples of things that you wish didn't happen.

I don't know what other people are doing so I don't know, like i Don't, if they are censoring people. I wouldn't even know. So it's so early. It's hard to point anything that's being down super well, actually, in any segment of nostr, it's so early, like the whole meme, like nostr's broken. You like, we're assembling the airplane in flight. Stuff's broken for all of you guys out there using our stuff. When when you tell us that stuff's broken, yeah, we know we know we see it like when you the bar is set by Twitter, you like multibillion dollar company. That's been added for ten years and people are like. Oh, why't you? Why does notes do this well, it's just they still haven't figured out. I mean, nobody has right, because the problem keeps on changing too right. Society changes. What's acceptable behavior. What's not what countries expect off the platforms? The laws change. But on the flip side, things are improving so rapidly that I think everyone can agree with because like you just look at where we were. Let's say six months ago. Things are moving quite nicely and I have a good feeling about what's to come back in the coming months.

Yeah. I mean, this year has been insane and just seeing all the progress of all the clients. And yeah, exactly what you said, how far we come, it's just it's been amazing and now look, we're here a year later almost and we're all in Japan and we're trying to figure this stuff out. It's just it's amazing to see this much energy and people just trying to solve these hard problems. So yeah, just appreciate everyone that's here and that are on the on the mission with us. So thank you yeah.

Yeah. I mean we're definitely on a user bear market now for nostr, which is perfect. It's the time to build thank God right because once the users come. I mean we saw this last time right. I mean you're busy trying to make sure servers don't melt like you notice when like a whole country comes on. I noticed I think it was like Saudi Arabia or something they came on. There's a million messages in like ours, something that's when BRP became BRB ye and then they left because they realised there's no delete.

That's right. I mean, that's actually a very interesting thing to bring up, because people have come to expect a capacity to the leach, they don't understand the ST strike and fact. They don't understand that on the internet, everything is there forever because they do remove from the immediate world that they live in, and things can kind of go away. So I mean. Germany has laws the right to be forgotten right, that are kind of like hilarious because the Internet doesn't forget and doesn't forgive. Isn't that anonymous. So I mean you had the feature right for for, I've deleted myself accidentally on demos. Well, yeah, its like, it's not a real delete, it just wipes your profile.

Is that just the past App Store? Yes, no, it was. It was the dumbest thing I ever implemented. It just like wipes your profile. And then it sets a flag on your profile that says Delete. And then when you try to log in with Damus, yeah. I actually. So I actually got rid of that because it's just it was so annoying. It's too confusing was that Paul okay with that they have. They're not checking anymore. Probably like yeah, they accepted it was a check mark first. And then now once you're on the app store, it's okay. Now you can just start to like undo the stuff that you got. I mean, we went through all this right with Apple in the early days of Bitcoin, right like Apple said no bitcoin wallets. I think it was 2014 or so and everybody started shooting their phones on videos on the internet. And then they kind of change course a bit.

I guess eventually the demand is so great that it doesn't matter what they think or want. Right. I feel like that's what we're going to end up seeing for zaps. It's unsustainable for them to want to take 30 percent forever of that stuff. Right, yeah, my goal with zaps is like we just need to put it in everything like and and just make it so such a common operation among in the nas ecosystem. And once the nas ecosystem is so large, it'll be like banning zaps will be like banning the hyperlink in the browser that and then at some points just be it's going to be silly for them to keep maintaining that policy. I agree it should be obvious that ios is just not as much fun as, let's say. Android or PWA because zaps are either not there or they're inferior or they're subject to 30 percent tax or things like that. But having said that. I think we, we should. We should try and do our best to somehow get the zaps back into the App Store even like I think it's more important for them to be there, even in the form where we bend the need, so to speak, and, you know, pay the 30 percent tax. I think that's a better outcome than not having them in the App Store at all, because Zap apps are so integral to the overall experience and enabled so many other use cases. I mean, we see this with Twitter subscription right. I mean. Apple said Elon, we won 30 percent of this, so Elon said. You know what if you buy it on the Internet, you? It's cheaper and let him get get away with it. But they like didn't let anyone else get away course Amazon. But eventually it becomes, you know, like you.

If you repays the cost to the user and they make it clear like this is the Apple Tax that they call it Apple Tax. On the receipt, users themselves are going to sort of like start complaining to Apple. They actually don't allow you to do that, so I think don't no Facebook actually tried it like and so their users if you tried to buy something or something on the marketplace. I forget the exact what what happened, but they actually would show like Oh Apple to be taking a cut and this is what it actually is. They actually got removed from the App Store when they did that because they.

They don't want users coming after them and being unhappy. Yeah, but you know, like people are going notice right that on Damus Android is cheaper to buy the zap that it is on ios right, like that's what does it? I think yeah, and especially if it's the same brand right, that's where things get like obvious. When I had to call with the Apple people about this whole thing, what I thought was really interesting. Was you were allowed to buy things with zaps physical things with Zaps. So this started to get me thinking i was like. What if there's a loophole here? Where can you buy someone a coffee like? What? What if I can buy someone buy someone's lunch? That's a physical thing that you're zapping. So what if we have a button on post? That's like Buy me coffee or buy lunch like I started thinking about this like Is this a potential way to get zaps back in? I don't know. But that was an idea I was thinking about.

Well. I mean, let's say things get worse right. And like the Pixel Store also doesn't want it anymore and like. Eventually we go back to maybe there's a different nip. Now it's just like lightning again. You know what I mean like, it's not lightning called Zaps. It's just, you know, something else, because people are not going to stop sending bitcoin to each other.

You can post a bitcoin address and get paid. Right. Becoin is no permission. Receiving. Yeah. I mean, these are the This is the huge like you know, these are the Gatekeepers, you know, the Google Plays Store and the Apple Store. That's of the biggest risk right now because like so many people use those devices, you know it's much more fun to use nostr on your phone. So I think that's kind of the biggest threat right now for Nossa and we need to.

Maybe we need new app stores. We need more ways to like Side load. Obviously they said they're doing that with Apple. I don't know if they if that's even Don't, I haven't read anything about it. But yeah, we need different ways to like, get things on our phones. It's not controlled by these. You know it economically gets weird right because see like we're not selling some token right like Bitcoin is a traded asset right like that has a price. So like if you give the app ten dollars to buy zaps and you're only getting 7 dollars in zaps to give away like people get crunchy even if you're not showing the Apple tax on it right because the clients are not going to take that loss right, its at 30 percent like it becomes uneconomical for you to have those right.

And we're going to definitely see more on ramps for sets for zapping in the platform too, like there's strike, there's all this a block with the Cash app. You know these apps are going to start getting integrated, and the source of Bitcoin for zaps may not be the one that we have now that Apple is after. How we're doing time minutes? Oh wow, what are we going to talk about? Yeah, let's do some Q and A get a mic around and really get some some deep questions here I see. I see hands there.

Hi. Lottie again. So first of all, it's really cool just to be sitting in front of all of you guys because you build things that are really essential to everything working on. I'm a huge Dumas fan. Use it every day, so thank you. I think one of the things I would like to know is in what ways I know you had said you. You don't necessarily need any more feedback because you guys already know all the things that no keep the feedback coming. I didn't mean don't give us feedback, just know that we probably know about it and maybe you know are going. Need some time to get to it, but definitely keep people complaining is like the main reason why I know certain things are broken. So I love it when people complain at me absolutely. But sometimes it's like Yeah, everyone like when everyone's saying it like. Yeah. I know remember the best form of is a bull request. Okay is so that comes to my question. It's like, what is the most constructive way, not just to give you all feedback, but also make sure that the community is aware of it too. Right because again, if it's like 70 people adding you that this thing is wrong, that's probably not helpful. The most useful thing that you can do as a user if you want to help fix the issues that you're seeing is just provide a way to reproduce it, because a lot of time people, people would say. Oh, this is broken. I'm like, well, what are the steps you went through to get to that state? And I think that once we have that flow, then we can easily fix the problem. So I think if you, yeah, if you want to help client developers just provide ways that you can reproduce the issue. So I consign that short of providing the code that say the actual fix. Yeah.

Thanks

Will you touched on kind of the hi over here? I know social media and how it's been negative for our generations, but you also talked about it earlier. You put out a note saying 99 percent of Damus users leave because it's not sticky. It doesn't have the algorithm right ye. So my Concern is how do we prevent these clients from becoming Twitter again because the algorithms work because it keeps people engaged? And hopefully, if we create the most efficient algorithm is probably going to be something like that, like how do we prevent that from happening again? Yeah, it's really Interesting because we ran through this with drone about our actual retention in the app and we're going through it and we realized it's like Damus is not even gonna be successful if we don't in some sense make it more sticky and add more engagement features. But then it's like Yeah, what you said, we're going down that route of what if we go overboard right? So maybe there's a balance where we can, you know, make an engagement, make it engaging, but in a healthy way, not in a way that's like driving people insane and making bad effects on society. So I'm really looking forward to exploring what that is because I want to build apps that like make people happy and make people that people have fun and not be overly addicted, which is weird like. Yeah, it would be great like, yeah, get everyone on, you know, make da. Like crack and get everyone addicted and everyone's using it. That's like but is that really what we're trying to achieve here. We're trying to do something different. We don't want to just repeat the old system, so I think there's definitely a balance that I'm looking forward to exploring. What that is i think you used an interesting word you said Prevent and the key here is you can't. People will do what they want to do and know. Hopefully. Twitter is going to see the value of nostr right and they're going to run their own version of it so you can't really prevent. It's more like. How do you compete? How do you have such an immersive, interesting, sticky experience that maybe it's like healthier right. But you still can compete against like a war point.

That people are sort of like super glued now on. I think if the question is, how do we prevent from nostr to becoming Twitter? I think the answer is, we just let nature take its course and it's going to happen organically because there are so many of us when I say so many of us so many already only a few months in there are so many different client apps making their own decisions and making different tradeoffs. And we'll get to see what happens when it all plays out and we kind of learn from each other.

Yeah, that's really interesting because we're definitely going to see some that are gonna be like super addictive, like really addictive algorithms. And we're going to have maybe some that are more chill and care about your like wellbeing. But that's the beautiful thing about nostr is, we're going to have all of these different things you can choose from. And depending on your your personality, you'll choose the one that fits best for you.

I don't know if that's your quote or if you stole it from someone. MVK, but I love it, you said.

Everything is allowed so everything will be tried. So that some got

So but this this has another layer right like the data is still there right like you can say For example, you are more interested in having a healthier relationship with your social media like you can still interact with the person who really wants that intense, horrible, sort of unhealthy relationship with it, Right, you still exist in the same network so you actually look completely off like can imagine the guy is super angry, angry tweeting and replies about some war stuff right and then you were watching his replies on the Super Calm interface, Where where is this guy coming from? There is nobody validating his existence in this other sort of way of displaying it. Maybe yeah.

Does that answer your question? Great? Do we have more questions? Hey, guys stage, right over here.

Millon you were talking about how there's three options for basically pulling in algorithms that have to run on a server. You either do it yourself, trust someone to do it or you don't do it at all. And then of course, there's the in-memory kind of like client side version, but we'll discount that. I think data vending machines are really promising approach to this because in any of those three options or I guess two options, you have to address the service that is providing the functionality and so you're coupling your client to that thing. Even if you make it swappable or user configurable or something like that, it's still a centralising thing. And more importantly, the algorithm is not fungible. It's not something that can be swapped out transparently. Do you guys have opinions about data vending machines? Is that something that you've looked into, you know, that seems like a really. There's already a kind for recommending content on the data vending machine. Can you guys first explain? DVM is so people get, Pablo agrees with me? Yes, is that huddle by the way Pablo should be on that talk somewhere. It's in five minutes. Okay, I'm good, so get his talk. I think I completely agree with everything. You just said I can't see who that was, but I co, oh over there, ok.

Yes. These are centralising forces and that's why I agree that the algos should be swappable. And so if we use Primal as an example, we'll make sure that our algos are available in other clients and that we also are able to consume other algols from other services within our clients. Probably dvms are the correct way to implement this, but we'll still, you know, go deeper and kind of figure this out.

I'm a bit of a DVM skeptic what my general philosophy is, and this is what I actually just wait and let veder implement everything and then see if it like is good at ideal or not. And it's actually a great way to just avoid doing stuff that has bad ideas. But dvms could be a great idea. I'm just like I'm gonna sit and see how it plays out before I do anything with them, but Yeah. I mean, I'm more focused on noterscript. I feel like nostrscript specifically for the recommendation or the pluggable Algol thing like I'm more interested in that, but they are competing ideas in that space.

How far are you on that?

It's ready to go. I just need to find time to plug it in. So yeah, once I go back l. Download now two weeks. Yeah, I'm not gon to promise anything. But yeah. I think it's an interesting approach that I just want to see it play out a bit more so.

The way we implement it is almost secondary to the idea that this needs to be done and standardised. Right.

Any any more questions hands up. There's a couple there. Make it difficult for this Guys come on.

I'm here yeah I was wondering what your opinion is about, how likes cost a lot of like people want the tension on line and how zaps are going to cost people to be greedy and they just want money and everything is going to be around getting sat which is cool but it can also have negative.

E-bagging Used to be the term back in the early internet days is, you know, people will have behaviours based on, you know, just trying to like you either ask for money or, you know, just do things for this zap.

Yeah, that's something that we have to worry about or really be concerned or address. It's really interesting cause I ran into this issue when I implemented Only Zaps for the first time. So only ZAps was a feature in Damus where it removed the like button on your posts if you turn it on and people are really upset about it because they're like. Oh, so you know. I can't just like anything anymore. You know, I'm just like I'm just throwing change at you, and that's the only thing that's important to you. So it's like there definitely is like a social a weird social thing. When you turn off likes on someone's post, it's like I only want thats. So yeah, i definitely agree that there that is a challenge, but I don't know I don't know how to solve it. How do you guys feel about one sat means you know, f you. I was at two sats means I can't remember.

Now one sat twice. I feel like one sat you twice. One side is like flinging pennies at people. It's like it's me.

Yeah, luckily, don't have sat, you know values because that would more rude. But tech. Technically supported.

Did we answer your question?

Yes. I was mainly thinking about, for example, young children like six years old. They're going to be using the parent. It's like parents should be the ones sort of dictating that it's the lack of parenting that normally causes those problems, not the technology itself. I mean, at least in my view, like there is only so much you can do if you have kids, you know what? They look at a tablet. The MOF just goes like this. You know their brains cannot like deal with it like it's just too sticky. So you need a human they're prescribed human to sort of direct them in a way that's healthier. At least you can't outsource that to an ALGOL, i think. Not yet.

Next question. All right.

I'm going, give this to Mr. Mbk. So you said this idea of like that the Internet is forever. And maybe that nostr notes are forever and honestly. I just don't fully understand that or like what the mechanisms or incentives are of like, this data continue continuing to persist at scale. Yeah, so data persistence is often like, even if you don't think about it in terms of design, it's just design to do that because you want the data to persist. It's more because we need the data to persist so that we can serve the data to users right. So we have all these redundancies and these redundancies end up never forgetting the data right. You're going to have like, layers of backup that are maybe deeper backups, and the data is always going to exist like it's nearly impossible to erase it.

Yeah, so yeah, a lot of people think that deletion, I We have a deletion spec. It's like. I think it's nip five something. And so the idea is that when you want to delete something, you send a note to the relay and then the relay should remove the note, by choice.

Yeah, they don't have to, but they should. And only the only issue now is that if any of the relays that didn't get that delete note now you still have one relay that has it. So now you have to pull down delete notes for every note that you pull. And so it's actually it's a really tricky thing. It's really and it's it's going to get even harder in Damus now because Damus actually saves all the notes it sees. So now I have to like pull down all the delete notes and make sure that it's deleted. But in reality it's like every Damus phone is going to have the note and you can't delete the note from everyone's phone.

I think the Directional problem here is that like deleting in computer science is a problem. When you have too much aggregated data, it costs more to delete than to store. So I think we're going to essentially disincentivize by design people to actually delete the stuff even though it's part of this SPAC. What you might see is filters based on this deletion right, so you'll hide at the client side. Baby like the stuff that is FLAG as deleted.

Yeah, you pretty much need to do that. Clients need to download delete notes and hide them. If it's deleted, it's an owner system.

It's actually easier to edit notes. I think so maybe we'll see more note editing in the future. So you just like you don't not only pull the note, but you've pulled down potential edits on that note. I don't think anyone's my mind was blown. The first time I saw that in action with long posts on ABLA.news I was like, Wait what?

Yeah, this is actually a huge flaw. I think in the long form notes. So we have this thing called a replaceable note. And so, You know you can edit by replacing the note but then you actually lose edit history because you're actually deleting the old notes so I want to, Me and Semiol have been talking about adding like no like. Instead of like replaceable notes, we just have a version history on notes, so yeah, they are replaced. But you keep the old version and there's a way to fetch old version.

So you're just chaining.

So it's much it's a much better way because we don't want to add deletion as like a core protocol mechanism because it just causes way too many issues. We see that with the contact list like you know, we get the weird sync issues with your with your contact list and it's because it's like there's information destruction happening. So. Any. Not much to it there. I just agree that contact List did that sort of, Did that get your question Answered?

No, no, that's is there any way. We can sort of like rehash that in a way that would answer. Okay,

Alex? Do we have time for more? One more question.

Hi, so going back to one of the previous questions How much of the immersion and user experience do you think? Has to do with the presence of celebrities or experts to have things of interest to say that the algorithm can uncover for the users.

I mean, it's very important, you know, that's probably going to be one of the deciding factors, whether a nostr takes off or not, at least to a point where it's like mainstream adoption is we sometimes we like to harp on influencers and all these types people, but they bring in lots of people because most people are lurkers, most people just want to consume interesting content from people making content. So if we're going to win, we need to actually get more content creators on the network. So we got to find wise. We got to find ways to incentivize content creators, so that's why we said like right when Kly Walker joined, they're like bitcoin influencers in some sense, I'm like everyone should be zapping them like we need more people like that.

I mean, like Twitter edits best right, like if we ignore its other problems, it's like it really is remarkable right that you have, like some of the smartest people and sort of more interesting people, more thoughtful people sort of hashing it out in public. I mean, that's really the public square right there right. So, You know people want to hear from people who maybe are like they're distinguished in their fields and that'd be just somebody who was very good at shuffle dancing all the way to somebody who who is a phd in physics.

Right, the yeah, the problem is a lot of those people don't really care about like censorship resistance and the reasons why we built these clients. So we basically just need, we just need to get a good UX to compete with Twitter at that point.

I mean people don't care about censorship until they get wrecked. Right. I mean so and we're starting to see there that does seem to be lurkers like famous people who are lurkers on nostr. And then there is sort of like people mirroring some of their posts. Right. I think, I think Sailor is is he posting his his bitcoin by is on Twitter because he sorry on nostr because he's on nostr too. Yeah, he just post images, just the pictures, just the pictures.

We need more than pictures right. So okay guys. I mean this was. I think like Super Super Useful Conversation. I think we we've covered a lot so That's fun. Give it up for MBK. Will and Millian Thank you guys.

---

About YakiHonne:

YakiHonne is a Nostr-based decentralized content media protocol, which supports free curation, creation, publishing, and reporting by various media. Try YakiHonne.com Now!

Follow us

• Telegram: http://t.me/YakiHonne_Daily_Featured
• Twitter: @YakiHonne
• Nostr pubkey: npub1yzvxlwp7wawed5vgefwfmugvumtp8c8t0etk3g8sky4n0ndvyxesnxrf8q
• Facebook Profile: https://www.facebook.com/profile.php?id=61551715056704
• Facebook Page: https://www.facebook.com/profile.php?id=61552076811240
• Facebook Group: https://www.facebook.com/groups/720824539860115
• Youtube: https://www.youtube.com/channel/UCyjDDtWFCntGvf4EyFJ7BlA