Latest / Computer Says Maybe / Nodestar: Building Blacksky w/ Rudy Fraser

Transcript

Nodestar: Building Blacksky w/ Rudy Fraser · Computer Says Maybe

• Alix: [00:00:00] Welcome back to Computer. Says maybe this is your host, Alix Dunn. And in this episode we are gonna go further into our wild galaxy of decentralization, and we are gonna be talking to Rudy Fraser, who runs Black Sky Algorithms, which is the company that has built the Black Sky App. But Rudy is so much more, uh, than the creator of Black Sky. Alix: He is, uh. Super interesting thinker who has been navigating the possibilities that at proto presents to people who want to build interesting things for their communities. And we talk about a wide range of possible applications that he sees for these technologies. What makes him excited about it, um, what he's been up to, and kind of how it all fits together. Alix: So onwards to my conversation with Rudy Frazier.[00:01:00] Rudy: I am Rudy Fraser. I'm the founder and CEO of Black Sky Algorithms. I'm also a fellow at the Applied Social Media Lab at the Berkman Kline Center. Alix: Do you wanna say a little bit about how you came to set up Black Sky and kind of what the maybe political motivations or technical motivations were for you to begin that work? Rudy: To be honest, why I started was really outta just curiosity. I was already working in the decentralized tech space. The first couple ideas that I tried, I was trying to follow the normal startup advice, and it just didn't feel right. It didn't feel like my life's work. So I ended up taking some time and reflecting on what my life's work should or would or could be, and all the ideas kind of centered around forms of community and community broadly. Rudy: Like not even just as like a shorthand for like the black community, but just community building. The really like starting factor for wanting to, to move quicker on making my own [00:02:00] thing was that we're in an executive offsite and there was a conversation about. Creating a diverse pipeline and there were people kind of questioning that as a, a valuable thing to do. Rudy: And to me, I was getting very tired of that kind of pushback in the tech industry. I always felt out of place in tech spaces. I still due to a large degree. It seemed like the best thing for me was to kind of, if I kept feeling like an out of place puzzle piece, was to build the puzzle around myself. Rudy: After I left that company, did a couple ideas. The project I started working on was this project called Paper Tree. I was inspired. By what I was seeing in 2020 around mutual aid. There was a big mutual aid movement that I'm still a part of in New York that kicked off in 2020. It really started with this idea around community fridges. Rudy: There are these fridges all around Brooklyn that have signs to say, share what you can, take what you need, and so people can just fill them [00:03:00] up with food and groceries and produce, and then. They're plugged into like a local bodega or local storefront. They work it out. These mutual aid groups, it comes to find out, worked it out with these folks who own these stores to supply electricity to the fridges, and then anyone in the neighborhood can come and get the resources that they need. Rudy: And I started it off with like, why aren't those everywhere by now? It's two, three years later. And I just did a deep dive on that. That's how I started learning about mutual aid organizing. And how these fridges are maintained by volunteers. I then got really interested in this new type of organization for me, like a, a non-hierarchical, horizontal organization, and I found that with a lot of that's, this still rings true, that with a lot of organizers and a lot of groups, there's a kind of resistance to technology. Rudy: Some of these groups hit a level of scale, like they were getting $20,000 in donations per month that they. Kind of had to use technology, or at least [00:04:00] some of them did. And those are the ones that I was really interested in. Like they had GitHub repos, they were using Airtable, they had all kinds of automation set up. Rudy: They were organizing via signal. They were doing hybrid meetings. When tech companies were struggling to figure out how to do hybrid meetings, there were people who had such an operational skill that was. Even better than what I was seeing at, at a lot of tech companies. There's a lot of talented folks who do mutual aid organizing, and so I was just really impressed with that and just wanted to be a part of it and wanted to find my space in it. Rudy: So I ended up creating paper tree, which was the goal was to make like a digital community fridge. For the neighborhood of Bedstuy and Papert Tree still lives on, it's used by a nonprofit named Gather for to do these suzu uh, community giving circles. But there was a point when I was trying to do paper Tree just for the neighborhood of Bedstuy and my bank account got shut down. Rudy: Banks don't like folks mingling money together, and that was kind of where the idea from Paper Tree really stemmed from that. The only kinds of bank accounts you can [00:05:00] have are. Your own bank account or like a joint bank account with a spouse, but you can't ever have like a family bank account. You can't share it with your group of friends. Rudy: I was looking at what unincorporated groups like, you know, like we, the people, NYC, which I'm an organizer for, Bushwick Island Mutual do is they get fiscally hosted by a nonprofit so they can get donations. So like Open Collective really took off because they were not only this like platform for people to do these kinds of transparent accounting. Rudy: Their groups. It was also doubled as a fiscal host for a lot of these groups. That's how other folks got around it. But I was trying to do it in a really scrappy way. Then I was like, I still want to do community work of some kind. Paper tree had taken me into these like Web3 for social good kind of spaces. Rudy: And so that's how I heard about Blue Sky and I signed up for, I thought it was really cool, their pitch for not Blue Sky, the app, but the protocol underlying it, which promised algorithmic choice. So choose how you see content on [00:06:00] these social media apps, democratic content moderation. So you can choose your moderator and you could own your data. Rudy: Those were all very appealing to me just from being in the decentralized web space. So that Cipher Punk kind of own your own stuff. It always just like appealed to me. And also for the algorithmic choice side of it, I kind of do that hacking myself. Lots of folks actually to this day, like you know when there's a early big wave for Black Sky people were joining and they're like posting certain things, like just trying to get the algorithm to show me the right content. Rudy: 'cause that's what we're used to. Not having direct control over the algorithm, but you wanna nudge it in the direction of what you like to interact with online. I have two different YouTube accounts, one that I watch music videos on, and one that I watch tech documentaries on because the algorithms can't understand that I'm this like. Rudy: Seemingly complex individual who like just has these two different, not usually combined personalities. There's lots of folks who do things like that. So I was like, oh. One of the things that I do when I show up at a social media [00:07:00] site that a lot of folks do is try to find the black content on the network. Rudy: It seemed like a really cool idea to just, it was just kind of a side, like Black Sky is really just a side project that became my main. Focus and yeah, so about a month after I joined Blue Sky, they launched the custom feeds feature and I was able to launch even before custom feeds were visible in the app, I had launched the Black Sky Feed. Rudy: You had to go to a whole third party thing to get it to work. The kind of jankis of the network was always also appealing to me. There's a personality trait that enjoys doing things that are hard for other people to do, and I'm very much that. So I launched the first version of the Black Sky Feed May 24th. Rudy: What? 2023. You know, there's lots of talk about traction in the. Tech startup space. All the while that I was ever building anything, I probably never got more than like 200 people to use anything that I built, especially when you're doing like it at a company, there's 400 people at a company. You'd probably be lucky if 50 people use the thing that you built, but right away, [00:08:00] as soon as custom feeds were visible in the Blue Sky app, there was like 900 views of the feed within minutes. Rudy: So I was like, oh, in big social media space, that's a small number, but like for an indie hacker, that's like, whoa, okay. I should, you know, keep focusing on this. Alix: I find it really ironic. Maybe I shouldn't find it ironic that you entered the space via Web3, like finding out about Blue Sky at a Web3 event, because I feel like there's some interest, especially since it seems like you're really interested in this combination of like community mission, money and decentralization. Alix: But I oftentimes think of like Web3. There's like a layer of community that feels. Hypey and like part of this like scammy thing happening, but then there's like the libertarian loner who I feel like wouldn't spend time like curating a community space in this thoughtful of a way. When you think about like combining mission and money and decentralization is like, has it like always been you kind of. Alix: Trying to find this space for yourself, or does it feel like you're just always kind of maintaining [00:09:00] these different sets of interests and occasionally you see this like mission oriented, interesting project and you'll just like keep doing that as you keep working? Rudy: I would say that is a fair assessment of Web3, especially like Web3 as it relates to crypto. Rudy: The underlying protocol behind Blue Sky is very, like some people have written articles directly calling it Web3 and saying that like it would be the most successful Web3 project besides Bitcoin because it has cryptography. Not cryptocurrency, but cryptography underlying it, it has decentralized identity underlying it. Rudy: So those are the parts of it that appealed to me on the community building side of it. I was recently at Harvard Business School as a part of this social media and tech solidarity workshop, as well as this ethical tech showcase presenting on Black Sky. Dr. James Riley, he was, I'm kind of gonna paraphrase, but he was pointing out how, 'cause there was folks there that were working on ai, working on vr. Rudy: Me working on decentralized web. There's a common phrase of like, you [00:10:00] can't take down the master's house using the master's tools. He made this quip of the biggest trick is like. Convincing you that the tools are the masters. These are just tools in and of themselves. I never really thought of myself as like a Web3 person or a D web person or any of those things. Rudy: I just always thought of it like community is very fundamental to me. I talk a lot about how like I experienced like food and housing insecurity. I didn't finish college because I was going through financial insecurity. Community is what has kept me alive. Even in like professional settings. Like after I left my job to pay the bills, I was doing Salesforce consulting and it was all referrals. Rudy: It was all people who I had built relationships with, who had known me, that supported me. And so the aspect of like what is a community need is more so like what I feel like I've honed in on over the course of time and like what technical infrastructure contributes to community. I'll have this like quip of the revolution isn't free. Rudy: You need funding. Money is the lifeblood of organizations and the traditional financial system just doesn't [00:11:00] work the way, like for nothing really works for horizontal organizations, right? There's not a single like CEO and person in charge. Nathan Snyder has a really good writeup about implicit feudalism and how like the default for the web is that there is an admin and then there are people with lesser permissions than that. Rudy: It always starts with an admin. The internet itself is hierarchical by design, and so any opportunity to make things fit for a non-hierarchical structure is what really interests me. And you can do that when you can have full control over it. So crypto kind of has that. It doesn't play out that way and how people actually use it. Rudy: But it's really interesting on the D side because. There is no financial incentive, and so you can just kind of structure a community how you'd like it to be structured. It's kind of been the appeal. Alix: I think a lot of the reason that people are drawn to decentralized technologies is there's an intellectual and kind of abstract. Alix: Sense that there's something [00:12:00] possible there that isn't currently possible. But there's very few examples where it's like, ah, like this. This is what we were seeking to do with the thing. And I feel like what you're working on is a great example of that where essentially you've asked this question of like, what does a community actually need to be able to be non-hierarchical and like have what it needs to do, what it wants to do. Alix: And it's just nice to see. Specific examples of those values being embedded in a technical implementation in that way. When you say that, you're trying to answer the question of like, what does a community need and then you think about Black Sky, like what is the answer to that question For the community that you're curating for? Alix: How do you think about. Its needs. And how has that evolved for you? Rudy: First, I guess if I were to just repeat it in how we built things and the real reason why we built things like the feed is so you can have a place. So if you have a community, you need a space to gather around. So there's a difference between social media and social networks. Rudy: And at one point I was like, articulating Black Sky is the last black social [00:13:00] network. Not because it'll be like an app. That will be the last app someone use. But because it will be this infrastructure layer that no matter which app you're using, as long as it integrates with the app protocol, you can participate and be a part of Black Sky. Rudy: And that's starting to become true when people are building these like, like flashes, which is an Instagram alternative or skylight or spark, which are TikTok alternatives, all on the app protocol. When flashes launched, black Sky was already there and that was something we were able to really easily coordinate with the founder of Flashes. Rudy: But yeah, so the custom feed portion allowed us to have a sense of place and space and there's more work that we wanna keep doing there. There's aspect of a dotted lines community, which is different from like the hard borders of like a subreddit. You can join Black Sky, you can leave Black Sky. You can not necessarily have your content shown, but you can be observing it and interacting with it. Rudy: This a sense of place in some way I think is important. And we have a public. Space. It's like the jumbotron and there's a lot of people all across, you know, at least [00:14:00] in the United States who. As a result of having a distrust for mainstream media and large social media platforms, they want more private spaces. Rudy: People are going back to group chats or discords or social media sites that are not subject to ad driven, attention based revenue models. I also think that's super important, and so I'd like to. A private space for a black sky as well. I think that's also key, being able to have your own rules that someone else can't override. Rudy: So that takes its effect in our curation as moderation. So the rules that remove certain content from the feed so that we aren't amplifying certain things. It goes into our automated labeling processes that labels anything on the network. If it uses like the N word that goes into how. We have our, our, a team of eight content moderators who are applying like our definition, interpretation of anti-blackness and misogyny, war to content that is reported to us. Rudy: And so, yeah, I think it's very important for us to [00:15:00] be able to establish our own community norms and then have our own rules and a team that maintains and, and manages and enforces that. Some folks try to word that as like censorship, and I just think it's like establishing the norms of. Because ultimately we can't take down someone else's post. Rudy: You always have freedom of speech on the protocol, but we believe like freedom of speech, not freedom of reach within the dotted line borders of our community. We have a specific set of rules that we're able to apply and you don't really get that level of ownership. I think the other piece is like, uh, important for sovereignty is to be able to just stand on your own two feet and not have to depend financially or technically on another organization. Rudy: That was why we built the relay, which is what enables the global social media aspect of a blue sky. Let's look at Blue Sky as the main example, right? Blue sky is decentralized, but you log in and you're able to interact with any of the 35 million users, you can hit the search bar and find any post. You can't do that on [00:16:00] the fedi verse or activity pub, and we think it's important to be able to have global access. Rudy: While still maintaining our own space and our own norms, but the only running technical infrastructure, the server that makes that global search and global interaction possible was mainly ran by blue sky up until about a couple weeks ago when we launched our own independent relay. And so the story there was just a, like if blue sky went down, black sky should be able to keep running 'cause a court. Rudy: Way that I'm trying to build Black Sky is that Black Sky, the social network, the community can outlive Black Sky the company and can also outlive Blue Sky. The company, black communities tie our identity to these corporations so much like calling it black Twitter, calling it Black TikTok. It's just. Rudy: Black folks interacting online. You can interact with the same people on different apps. You may meet those people in real life. All of this is all part of the same kind of social network and relationship. And [00:17:00] so we built the infrastructure though so that if Blue Sky, the app or Blue Sky, the company goes down, we can keep going. Rudy: There's a bunch of tools that we keep building on with that theme of like, the community should outlive the company. Alix: It also fits with the theme of other work you've been doing and I think there's like, there's something about like. How Twitter views a user and how a bank views a customer in an account. Alix: And like not being able to have the architecture even to have shared spaces that where there's autonomy over the governance of the thing with the group of people that are participating in it. So I feel like there's a, there's a through line there that I feel like is really interesting. How does the governance work around the decisions about the bounds of content that you all curate for the community? Rudy: I was actually. I was kind of talking about this a little bit earlier. We don't have the kind of like, or I guess I don't necessarily personally believe in the, like voting on every decision kind of model that some folks try to implement, especially on something like [00:18:00] moderation decisions. We've made over a hundred thousand moderation decisions since November. Rudy: We get about a hundred to 200 moderation reports every day, and so. I think someone has done a writeup about like the tyranny of lessness and I, I haven't necessarily read that as I've said these thoughts. People have repeated that back to me, but just from what I've seen in mutual aid groups, there is no hierarchy. Rudy: But the people who do the most work and are willing to put in the most work tend to be the ones who have the biggest sway and influence on decisions. And they're usually like, even if there is a full vote. They're kind of the defacto ones that folks listen to. And so I think we take that model for the moderation decisions. Rudy: The moderation team makes those decisions. But anyone can join and become a moderator. But I do try hard to like let folks know that moderation is traumatic work. Whatever the violist thing that you're, you've seen on the internet, that's kind of what you'll see every single day. [00:19:00] And. We are a black moderation team moderating anti-blackness. Rudy: And so it'll, you know, hit you a bit differently than if you've even had trust and safety experience in the past. So the team right now, they've all done moderation in some capacity in other spaces. Most folks think we're just handling like the N word slur, for example. There's that. There's also kind of a less explicit form of anti-blackness, like cheering on Daniel Penny or who had like murdered Jordan Neely, but you know, got a non-guilty verdict. Rudy: There's that kind. There's also in community issues, right? One of those examples was there's a conversation that generally is happening right now, I think since the election, which is what, you know, blue sky spiked and black sky spiked. Following the election. There's conversations about like how black folks should show up politically in this moment. Rudy: Are folks protesting? Are folks sitting this out? Or whatever the case may be. And there was someone who had made a post who was like, we should all be protesting right now and dah, dah, dah. And then that got [00:20:00] 200 replies, 200 quote posts, 50 reports to the Black Sky Mod Service. And we had to just do a quick huddle and be like. Rudy: I don't think this is anti-black. Maybe everyone disagrees with this statement. Maybe this wasn't the right thing to say, or whatever the case may be. I have no opinions on it, but we're pretty all clear that it's not anti-blackness. And so yeah, we, whenever we're unsure 'cause we're, we're dealing with so many different kinds of. Rudy: Content we're dealing with, you know, text-based content, images, videos, and all the different nuances of anti-blackness, internalized, anti-blackness, all this stuff is going on, and so I don't expect anyone, any one person to always feel super confident. We want to try to empower them to make these decisions, but we also come together as a team when there's things that are like, oh, this is, this is unsure. Rudy: We should talk about this. You know, Alix: I really respect the. Not trying to be like liquid democracy style, every single decision. Something that, uh, everyone reviews. 'cause I feel like it's like [00:21:00] there's a purity ambition there that I feel like doesn't actually serve the underlying like, practical challenge of managing these kinds of projects. Alix: So I feel like, although I do recommend Tyranny of the Structurelessness, that's a really good read about how. Basically, when you don't assign hierarchy, which it sounds like you already have assigned a little bit of it, or if you don't make power explicit in a network, it manifests in ways that are harder for people to grapple with because it's basically less explicit, which makes it hard. Alix: Also like walling it off. 'cause I feel like content moderation, like I have friends that are content moderators and it is such a specific experience and I think also skillset to be able to deal with the psychological impacts of it. And I feel like encouraging people that are amateurs to participate in that, or people that might not understand the implications, like it can really, really affect people permanently experiencing that kind of thing. Rudy: The first mod I brought on burnt out, and so that's when I was like, okay, not everyone can do this, even though everyone may want to participate. And yeah, the, the power structure that just forms, if you [00:22:00] don't make the, the structure explicit is like what I've witnessed in different groups. And I was like, no, there should be like a clear, especially we're dealing with stuff where people, it's very public decisions. Rudy: We've had to label celebrities, right? It's moderation decisions are loud and controversial, especially if you're trying to like. Cultivate a certain set of community and norms. There's people who are just gonna butt up against those norms, right? There are black folks who like join Blue Sky and they're like, I don't wanna be a part of Black Sky. Rudy: Like, I don't believe in this like, collective thing. And you know, that's fine, but like, don't tag us, don't tag Black Sky as you do that. Um, everyone can have their opinion and stuff, but so there's just, there's just natural tensions that occur and, and yeah. So, and, and people can appeal things, right? And so, and so it's a complex. Rudy: How Alix: does that work? What, what is, what's the process? Rudy: So on Blue Sky, the app that you're using sets some default moderation services. So the main Blue Sky app that most folks would use defaults to using the Blue Sky [00:23:00] Moderation service. And then you can go search and find, you can search Black sky.app. Rudy: It's a special type of account. So there, there'll be a button that says Subscribe to labeler. And then if you're subscribed, you can then see the things that we label. And so if you've been labeled by us and you're subscribed to the labeler, you can label accounts, posts, profiles. Really any record in the app protocol ecosystem, right? Rudy: So if you see that label on your account, you can click into it and submit an appeal and write up a text of your appeal message essentially. And then we review it on our end. And usually for me, if it's something that we're gonna overturn and especially for intra community issues, I'll reach out to the person and have a conversation with them and see like what making it right. Rudy: Looks like We try to take like an a kind of a, a re reparative approach. There's some folks who have offered to like. Help us flesh that out a little bit more. But there's basically situations where things got heated, words were said, we can kind of walk some things back and there's cases when that happens and there's cases where we reject [00:24:00] an appeal because the thing that they did before was blatant and it almost feels like an appeal is kind of a troll. Rudy: Anything can be a vector of abuse. You know, when there's an open text Alix: box, words never spoken. Yeah. Um, that's so ing So is there, I mean, 'cause a lot of this feels very, um, hands-on, bespoke, done with care. How is scale affecting your ability to do things in like the ideal version in your head? Is it as things get bigger, like is it making you rethink some of these more deliberate. Alix: Intentional. Lovely. It sounds like interventions or is it more like we've reached a level where we think we can like maintain, or like how are you thinking about scale and care? Rudy: We're on a decentralized protocol with the explicit intentions that there will be billions of users on it eventually. That's kind of, that's the goal, right? Rudy: That's Blue Sky's goal. They want to build a decentralized version of like a Facebook scale. Network. And you know, the Fedi verse was very concerned with much smaller [00:25:00] communities and just didn't reach this level of scale. And so what I find really interesting is that although the, the broader network can be billions of users, you're able to make this dotted line community for your. Rudy: A couple millions of folks. I think there's like an upper bound of what I would expect to be a part of Black Sky, because it's not meant to be every black person, right? It's like whoever just opts into this community. And there's other communities that have followed along in our path, so we're not the only one. Rudy: There's people who've like seen our model, their depths from other communities. Are contributing to our code base and giving us credit of like forging this path forward. North Sky's an example of this. I think that there in this couple million user space, that's where Black Sky is at, right? We have, our feed gets viewed by 370,000 people per month. Rudy: We've had over 2 million. Users in the last couple years interact with our feed one point something million just in the last few months. And I think this is a very manageable area for [00:26:00] us. And I, I think that when you're a platform, there's a distinction I find between platform moderation and community moderation and platform moderation is very much concerned with kind of insurance for a social media company, make sure you don't end up in. Rudy: Law enforcement's bad graces or in like a bad article about you, right? It's protection against cs, a copyright infringement, that kind of thing. And community moderation is concerned with the. Wellbeing of a certain set of community and establishing their norms and, and guidelines and things. And there's tons of tools for the platform moderation. Rudy: They get bucket loads of money from the mega corporations, and there's virtually nothing for community moderation. Usually it's like, you know, Reddit, mods Reddit because they know they, they depend on their moderators, even though it's a very like one-sided relationship. They give them tools that they're able to use. Rudy: Protocol. We had to build everything from scratch virtually, and so our tooling is in place [00:27:00] between our tooling and the intentional scope and size of Black Sky. As a community, we're able to meet the needs of our community. I grew up on the social media that was not the global newsfeed, not the Twitter algorithmic feed. Rudy: When I joined Facebook and Twitter, they didn't have either of those things I grew up on like AIM and internet forms. You see this still with like subreddits that you can just make up really wacky rules. My favorite example of this is, uh, there's a very popular subreddit with millions of users and I, I almost forget its name, but they had a rule where you have to curse in your post, or it'll be taken down because they found that there is some crew of folks that are farming karma, which is the reputation system on Reddit. Rudy: They're farming these accounts by making posts in this community and flooding it with like nonsense posts and then selling those accounts. And so they found that if you have to use a, a curse word in your post, that would kind of be like a poison pill for these accounts. And so those accounts stop showing [00:28:00] up. Rudy: There's tons of examples of those. You can scale that to millions of like, you know, have these rules for your community. Essentially. That's different from if you're trying to be a billion person platform and you like have to use AI to like enforce and you're really just concerned with the platform moderation kind of things. Alix: Do you want, do you want it to get gigantic? Like is that an ambition? Rudy: I think the goal with Black Sky is to create, because we've built all this infrastructure for ourselves is to. Sell this infrastructure to other communities as a way for us to have a our business model. And so each component of our stack, that's the other cool thing with that protocol. Rudy: All this stuff is composable. As a user, you can use our custom feed and not use the mod service. You can use the mod service and be looking at other feeds and the mod service still works. You could have your data hosted on our server. You could have it hosted somewhere else as a dev, you could interact with our relay or the blue sky relay. Rudy: All this stuff is composable. And then if another community wants to use our infrastructure, maybe you don't use our [00:29:00] custom feed that has this feature of like community membership hashtags and the the band from TV feature that we have that stops people from viewing the feed. Maybe you don't need that, but you lack our auto reporter service. Rudy: Which scans the entire network and then labels content or reports it to your moderation service based on a certain rule that you apply so you can take different pieces of our stock and use it yourself, you know, ultimately what I think about is I'm a weird case of a person who is technical enough to write code and rust and do DevOps and like spin up a virtual private server. Rudy: Then also have a very strong opinions on community norms and, and needs and like can build a moderation team. But like typically those are two different, maybe three, four different people. And so I think two things, a barrier to community builders is the technical stuff. So can we make a one click solution? Rudy: The sell here is not run your own server. The sell here is create your own rules. To create your own rules, you need [00:30:00] to run your own server, so how can we make it easy for you to just get started doing that? That's one goal that we have Also. The relay that we built. We got funding from Skye to hire a dev and, and build that out. Rudy: And in the conversations with Peter Wang from Skye, he's very much interested in like, how can we replace the centralized networks with a decentralized service? Like how do we get there? Maybe there needs to be new apps besides just the blue sky. You need like, you need the TikTok app, you need the Reddit alternative, all that. Rudy: Agreed. You know, in part what's stopping folks from building those apps, they're all scared to do content moderation because on App Protocol, you no longer have the problem of the Cold Start problem where you won't get any users, and now you're spending a bunch of time like having people sign up for wait lists and doing marketing. Rudy: Essentially, you probably will get users. Are you ready to have like 50,000 users sign up for your thing when you launch it? Do the content moderation for all those folks, like I think Skylight especially 'cause it's a TikTok alternative. When TikTok got [00:31:00] banned, I think they got like a hundred thousand users to sign up and they like went number one on the one of the app stores. Rudy: I think they're doing something that a lot of folks are doing where they're still relying on Blue Sky to do the content moderation because if you have a quarter million folks sign up for your app. And it's videos. Now you're doing concept moderation for, you know, maybe hundreds of thousands of videos. Alix: It's hard. Yeah. Yeah. I hadn't thought about that in that way, that like essentially at Protocol, just as one example, you have a pre-built in user base. Potential user base but without the lock-in. Yeah, that's so interesting. Like lowering the barrier of risk associated with trying. Rudy: That's why I tell people, so like just try to build weird stuff. Rudy: 'cause now you don't have to try to think of like getting billions of users or like you, you don't have to go that far. You can just like have a app that's just used by like a couple hundred people that really like it and maybe they'll pay for it. You know, you can. Especially Alix: the half, the top half of the funnel is completed for you. Alix: Like there's a Yeah, [00:32:00] exactly. Which is like really convenient. I also really like this idea of like B2B, white label portfolio of technologies that allow people to basically like not. Again, the Cold Start problem, like not have to, I don't know. It's like watching every single startup that seems surprised that even though their entire goal was to scale beyond their wildest dreams, when they do, they haven't done any planning around governance and moderation. Alix: Like it's like they like Columbus the same problem like over and over again. And it's like, but didn't you see every single other company along this path have this exact same moment of being like, oh, I guess we need a trust and safety team. It's like, yeah, like you needed that from from the beginning. Alix: But that's such an interesting, so are you thinking. For financial sustainability that you all would kind of be like a, I guess instead of B2B, it would be like C to C, like community to community, um mm-hmm. Uh, like portfolio offer of this, like social infrastructure rather than technical. Yeah, yeah, Rudy: yeah. Rudy: And it's all open source, so if anyone wants to just run it themselves, they [00:33:00] can also do that. Sure. Um, yeah. But yeah. And yeah, for other communities, for sure. An example of how I've been thinking about this is just like to make it as easy to launch a community on app protocol as it is to launch one in a Facebook group. Rudy: But then you have like way more control over that way. More ownership over it. Uh, same thing with subreddits, especially 'cause like subreddits, they can't even monetize. They're not allowed to take money. Alix: If we go down to the very bottom of like thinking about at the protocol level, like are there any concerns you have about like, you've essentially chosen this and to invest a lot of time, which obviously you could pick up and maybe move somewhere else, which is the beauty of these systems. Alix: But how are you feeling about the trajectory of Blue Sky as a company and an entity of the protocol? Like how do you feel about the trajectory of that base layer, decentralized network staying the course as something you wanna be like a top of? Are you imagining there being other. Things like at Protocol emerging, like I know Activity Pub wasn't ideal for certain use cases and then at protocols feels like it solves some of the problems of Activity Pub. Alix: But then there's also like people with Activity Pub they're not maybe happy with [00:34:00] at protocol being the thing that everybody's obsessed with now. Um, like how do you think about that like trajectory of decentralized infrastructure that you all build a top of? Like are you anticipating change? How are you feeling about it? Rudy: I think there's a lot to learn from other protocols and the histories. There's also like a whole cultural aspect to. Activity Pub and Mastodon that are always, you know, I always just start off with like the whole thing that people always bring up with Activity Pub. And that protocol is Activity Pub, just the first. Rudy: And I'm always like, so why wasn't there a black sky on there? And there are examples of like what happened when someone tried to make a black sky on Activity pub. And so I've learned a lot from that and had tons of conversation with folks. So from a technical perspective though, I think there's stuff to learn from Noster. Rudy: I think there's stuff to learn from forecaster. There's lots of ideas there that could be interesting to bring into the protocol. I think I really like the, the dev ecosystem around a protocol, like whenever someone tries to say like, oh, a protocol is this or that, and the third, we just go and we [00:35:00] prove it. Rudy: Like we just go build something and then show folks that like it's a very show, don't tell culture, and that resonates a lot with me. I think the, the, my one gripe with app protocol, there's all the aspects that folks from the outside think are centralized or something that's not my concern. Those aren't things I care about. Rudy: So I don't really argue about those. I do think about the future state of things. So like, you know, blue Sky is very explicit about decisions about what the protocol becomes are in, in their control. So for example, I have this kind of. Hope that one day paper tree and black sky can kind of integrate or become one in a new thing. Rudy: And you really have this financially independent, self-sovereign online community that has its own like local economy and blue sky is planning to include. What they call did to, did payments or basically account to account peer-to-peer payments. Alix: Interesting. Okay. So you're feeling generally [00:36:00] pretty good, I feel like about like the, yeah, yeah. Alix: The way the infrastructure and like the direction of travel and like, which is great to hear. Rudy: Yeah. My biggest concerns are around the wider ecosystem, right? Understanding like in the press from journalists, from funders, understanding that app protocol is more than just blue sky. That's kind of been my primary interest in what, when I'm going to all these different schools and doing these public conversations, is trying to get folks to understand that blue sky really is just the, the entry point for this ecosystem. Rudy: And there's tons of folks doing really cool things elsewhere. Another example is like the Blue Check system. So I have a blue check on on Blue Sky that was given to me by Blue Sky. I was making the case in a recent article that like. We should have community. I want Black Sky to give out checks to verified community members and somebody forked the Blue Sky app. Rudy: Uh, it's called Dear Social. They enable you to choose who can verify who you respect the verifications of. So you can elect Black Sky as [00:37:00] someone who can give out checks, and then you can see on those accounts if Black Sky gave that person a check or not. And so this is like taking the underlying protocol stuff and putting it in new app experiences so that you can then create your own norms elsewhere. Rudy: I think all that stuff is really cool. So lots of exciting stuff happening in the ecosystem, so very excited for the protocol and the devs in the ecosystem, but. Need the broader world to start to understand what we got going on. Alix: It's pretty complicated, but I feel like even just that rhetorical suggestion of not saying that it's built on top of, I think is really helpful. Alix: 'cause I feel like people's mental models, as you mentioned, are so stuck in the like app store headspace that I think it's really. Important to find ways of like visually representing or just communicating more effectively about how it works and also how it's different. And also like helping people understand that we're so trained to think about hierarchy in these spaces and like centralized control within, within companies that I think it's gonna take practice almost for people to [00:38:00] feel able to even like let themselves believe that this could be the way that something is run. Alix: Um, okay. Well I feel like my last. Question is literally just what do you think this is gonna evolve into, which you kind of partly answered, but like Black Sky in particular, like let's say to be more specific, let's say in the next year, what other things are you excited about, either rolling out or considering or experimenting with, or kind of what do you see in the future? Rudy: I'm excited to kind of prove out a business model for. Hosting these services for other communities. I'm excited about that. The devs are doing a ton of work right now to make that turnkey solution of press a button and get your own server really easy. There's other people doing it in the ecosystem, but we're doing it for our specific code base that we've developed. Rudy: I'm also excited about having an app. That black sky can go to that is kind of personalized. All the stuff that I've built for Black Sky so far is like over a hundred something thousand lines of code. It's all backend services though, and so I'm kind of excited about building stuff on the front end and doing that in [00:39:00] collaboration with folks and what I mentioned with Dear Social. Rudy: Like that was really inspiring seeing someone take that and start to make it their own. I'm like, okay, so you know, it's, it's within reach and we just have to prioritize it. So I'm very, very excited about that. Yeah, I think we're bringing on someone Zaria Geelong to on the, the Black Sky Team, and she's, she used to work at Google Jigsaw helping to develop the, uh, perspective API, which is like a free tool for community moderators. Rudy: And now she's at MIT in the MITB. Studying communication, doing research on communication online. And so really interested in working with her perspective and some, some of these AI tools don't really work for community moderation. They're good at detecting illegal content for platform moderation, but they can't really adapt to a community's rules and norms. Rudy: And so I really want to get something like that deployed to make it easier for the moderators. Like I wanna be able to scale out our moderation team. And then be able to [00:40:00] demonstrate that, hey, this is a way of making moderation easier for the rest of the network. Alix: Thank you. This was great. Thank you for taking the time. Alix: I'm sure you do this like a million times a day. Uh, uh uh, especially given that's, yeah, that's all I do. It's all you do. Um, I mean, I hope you do other things. I really hope that for you. Um, but thank you. And thank you for all your work doing this. 'cause I know that, um. It's, it's a risk doing this kind of thing and investing your heart and soul in something, and also trying to find a path for yourself career-wise that makes sense and like fits together with mission. Alix: And I just always respect people that, that take those leaps. Um, so thank you for doing it. Rudy: Oh yeah. The entrepreneurship journey has been kind of a spiritual journey too. Alix: All right. I hope that inspired you to go explore all of the apps that might be interesting to you on at Proto, and also to potentially learn more about Rudy's work. Next week we have Andrew Trask, the CEO of Open Mind, and we are gonna dig into what it takes to anticipate and build the [00:41:00] protocols that allow for decentralization of training in ai, which might sound super wonky. Alix: It is, but Andrew's an exceptional communicator. In fact, I think machine learning education is one of the things that he's most passionate about. Um, so don't feel intimidated. It's a wide ranging conversation that I think is super accessible. And the problems he's working to solve are directly related to what I set up top in this series, this sort of dramatic concentration of power. Alix: So he's been doing. A lot of interesting work to think about and build the protocols that we need to decentralize the next AI era. I'll leave it there. Thank you to Georgia Iacovou and Sarah Myles for producing this episode and, uh, Georgia for coming up with the idea for the series, and I'll see you next week.