Computer Says Maybe

Alix: [00:00:00] Hey friends. Uh, welcome back. I am very excited about this episode and the next set of episodes, which is a bit of a mini series. This series is called Node Star, which I'll get to what that means in just a second, but I wanted to start with why we're doing this series, and it is connected to why this AI era feels so gross.

Alix: I think for me, that queasiness is caused by the level of. Extreme concentration of like everything of power, like literal power, energy, but also decision making power where basically 10 guys and their like venture bankers have concentrated the whole narrative about the future. Also, the concentration of data where basically AI companies swooped up all of.

Alix: Digital creation and then kind of tried to sell it back to us in this mangled, hegemonic, weird thing that's like wrong and gross and disorienting and confusing. [00:01:00] And then also just this kind of compression. I feel like it, it takes me back at least to that. Bizarre ad that Apple ran unironically, where they showed the compression of like all art, all creativity, all randomness, all joy, and kind of crushed it to this like singular product that you can buy if only you had the money.

Alix: That idea that these companies have taken basically everything I used to like about technology, which is that it was fun and random and you could like make things and break things and talk about making things and breaking things. And I feel like this period has been this rush away from that ethos and that feeling and towards something that's dark and really going in, I think a pretty horrible direction.

Alix: So rather than spend more time talking about that concentration. We wanted to kind of look at the opposite. Um, so what is the opposite of concentration and it's decentralization. And a couple of months ago, Georgia had the idea of doing a deeper dive into the history of decentralization on the [00:02:00] web and then to talk to people that were building things that could help us reimagine what things would look like if they were cool again and fun again.

Alix: And instead of feeling like being on the internet, is this like frustrating self-checkout line at Walmart or something? So that's what we're gonna do, and the series is called Node Star Combining Loadstar. So the idea that, you know, we're looking for guidance, we're looking for navigational assistance and like where are we going with nodes, which is for you nerds out there, or people who are interested in network theory.

Alix: It's this idea that like a constellation of nodes makes up a network and that power is distributed across those nodes and it's not allowed to centralize in any one node. Or the network would break. I'll stop because a lot of the people we're gonna be interviewing speak a lot more eloquently about these topics than I do.

Alix: But this is the series. We're gonna do three episodes with Node star now, but as we've done with other series, we might dip back into this topic to try and connect other people and ideas to it. 'cause I think it's a really important one. The first conversation is with Mike Masnick. He is one of the most important [00:03:00] thinkers on this topic.

Alix: He's done so much stuff. He also like coined things like the Streisand effect, which I think is great. So a lot of the things you think about their kind of core ideas about the internet probably have some connection to Mike. He also wrote this really, really influential piece in 2019, advocating, moving from platforms to protocols, and we'll get into what that means in this conversation.

Alix: He's also a board member of Blue Sky, so is kinda at the forefront of a lot of the decentralization efforts that are popular or gaining popularity today. He lays out the kind of history of the past 30 years of decentralized primordial internet days, and then this period that we've experienced of intense centralization and concentration of power.

Alix: And then we get in a little bit about like the recent green shoots of promising development that might allow for more of the stuff that makes technology like actually good. So with that, my conversation with Mike Masnick.

Mike: [00:04:00] I'm Mike Masnick. I'm the founder and editor of Tech Dirt, which is a publication that covers all sorts of stuff around tech and innovation and policy and. Internet speech and everything, and I'm the executive director of the Copi Institute, which does research and reports and games about all this stuff as well.

Mike: And I am also on the board of Blue Sky.

Alix: Cool. You also have like. Multiple podcasts

Mike: and I have multiple podcasts. I have the Detector podcast, which is an interview podcast. I have Control Alt speech, which is a really fun weekly podcast reviewing all the news related to internet speech. And then I just completed earlier this year, uh, eight part podcast series called Otherwise Objectionable, all about the history of Section two 30, which is sometimes referred to as the law that created the internet.

Alix: I was thinking we could start, I kind of curious if you'd be willing to give us some like throwback context [00:05:00] about when you first started to get into the concept of decentralization. So were there like early exposure to technology where you were like, oh my God, this is this new formation or network structure that's like really exciting for all kinds of reasons?

Alix: Like when did this interest of yours start?

Mike: So some of it comes from the fact that. I'm old and so I was a very early internet person in the early 1990s. I got into the internet and very, very quickly realized like, this is an amazing technology and it, it is incredibly empowering and interesting and. It was gonna change the world.

Mike: And, and that was sort of my view in 1992 or 93 as I sort of first got on the internet, even before the web existed. So, so other pre-web protocols existed and I found it fascinating and curious. And then, you know, we had. 30 years go by and you know, somewhere in the post [00:06:00] 2010 timeframe started to look around and realize that the internet was different than what I had been promised in the early 1990s.

Mike: And spend a lot of time thinking about why and how, and are there ways to get back to that. And a lot of it came down to the fact that in the early days. The power of the internet was the fact that it was decentralized, that anyone could build anything, and that you weren't reliant on big, giant companies who were making all of the decisions.

Mike: There were a few companies, but they weren't as big as, they weren't as giant. They weren't manipulating the experience that you had, that you had control over the experience. And so almost as a thought process. As a thought exercise just kind of went through and it was like, well, where are we and where could we be?

Mike: And just realizing that it felt like all of the. Promise of the power at the ends of the network, the power to the users had effectively died away. Not entirely, and that's a bit of an extreme statement, but you know, getting the basics [00:07:00] there. And so that's when I started to write about how can we bring that back and how can we rethink of this internet and how can we create incentives as well to move everyone back towards something where the power is in the hands of the end users as opposed to.

Mike: Individual powerful entities in the middle, whether it's giant companies, whether it's billionaires, whether it's the government, or those three things at times these days seem to overlap and be one and the same and just sort of say like, how do we find this path that brings us back to an early internet?

Mike: So to some extent you could say that that was my original. Experience with the internet, that it was decentralized and that all of the powers were at the end of the network. And then I just sort of had to rediscover it in the early 2000 tens and then spend the last 15 years trying to figure out ways to move people in that direction.

Alix: I mean, I feel like, 'cause I'm imagining when a OL CDs started getting flung into people's mailboxes. Do you [00:08:00] feel like there's a zero sum tension between access and agency? The more people that got access, the less control they wanted and so the less control they had or kind of, how do you, do you see those two things as intention?

Mike: No, that's a, that's a really interesting framing and I'm thinking about it, but I don't think I agree. I never had a OL, so I sort of got on the internet before a OL became as big and as dominant as it did. So when the AOL CDs were flying around, I was. One of those kids who was too cool to be like, you know, those poor a OL suckers.

Mike: But, but the interesting thing was that a OL lost, right? I mean, so a OL was there and they had the CDs and they had the, what, you know, what was, it was like you got 10 free hours and then they were charging you per hour, and then they finally went to a flat rate and that made everybody go nuts. But part of it was that somewhere.

Mike: Around there. Was that a OL connected to the wider internet? Right. The original a OL experience, like many other early internet services, was this [00:09:00] walled garden experience where you were sort of subject to what AOL's interests were. And then I believe it was. September of 1993 that they sort of flung open the doors to the wider internet.

Mike: And the famous phrase is either the, the everlasting September or the September that never ended because there was this thing, which if you were on Usenet. Which was the early protocol based sort of proto Reddit. You know, you had different news groups, which was all based on an open protocol, and you would connect to a server and there were a bunch of different servers, and some servers would carry some groups and some servers would carry other groups.

Mike: And there were things that were blocked on some and not on others. So what happened was that every September when new students would go to college and get their first internet account, they would descend on Usenet and they wouldn't know the etiquette, right? And they would act like idiots because they didn't know what they were doing.

Mike: And people would get mad at them and they would say stupid [00:10:00] stuff. You would have, every September, just a bunch of newbies would show up. Cause a bunch of trouble and then they would get socialized because people would be like, no, we don't do that here. This is how we act. This is the etiquette of, of using Usenet.

Mike: You don't type in all capitals that's screaming. You know, like all of these like basic concepts that people didn't quite realize. Then in September of 1993, a OL connected to the wider internet and suddenly millions of people flooded Usenet and other parts of the internet. It was called Eternal September, the September.

Mike: That never ended because they overwhelmed the systems that we had before where the ability to sort of socialize and say like, no, this is proper etiquette. This is how you act. They were swamped by all these people entering the space. It was a sign that the wider internet and the freedom that the wider internet provided won in the end, right?

Mike: Because it allowed for more experimentation and cool things to happen. And yes, it was messier and no, it wasn't [00:11:00] as clean and it didn't have the same nice UI that a OL had. It didn't announce you've got mail as you logged in, right? You had to figure all some of that stuff out on your own. But because it was open and decentralized and anyone could build on it, anyone did build on it, and you, we had all these amazing useful services that many people rely on today, but that was not because of a OL and its control, but because of the wider internet.

Mike: So there is this element that went done, right. Even if it's messier and causes more trouble, having a decentralized open system on which anyone can experiment and anyone can build leads to a better situation overall, it's just a question of, you know, how do you do that? And so we had this thing where the pendulum sort of swung from one side to the other.

Mike: We had. Centralized walled gardens of a OL and Prodigy and and whoever else in CompuServe. And then it swung to an open internet and then that was crazy. And then people started [00:12:00] to build these new centralized services on top of the open internet and you begin to get these new walled gardens and. You know the big companies that we know of today, and Facebook obviously being a big one, but Google to some extent too, which sort of shifted from, you know, in their early days, their entire motivation was to send you somewhere else as quickly as possible.

Mike: 10 blue links, we wanna drive you away from our site. They had said that very clearly, but then as the advertising market became more and more important to them and they had incentives, often driven by Wall Street to show. If ever increasing returns, they started to say, well, you know, look, if we can keep people in our platform longer and longer, we can sell more advertisements.

Mike: Or if we can take over the rest of the advertising market and have advertising everywhere else, and track your data everywhere. Everything sort of took us on this path towards more centralization, but with it you lost that freedom and openness to build. And so we're in this [00:13:00] point now where hopefully we're sort of, that pendulum is swinging back from a very centralized system where there's less power at the ends of the network to one that is more users in control.

Alix: That's really interesting. I think I, the reason I ask is 'cause I feel like there's a, um, sometimes an expectation for individuals to engage with a lot more focus and time and attention and energy on technology in order to access that freedom and control. Um, and I think, like I'm thinking about the difference.

Alix: And we're jumping ahead here, but the difference between the rollout of Mastodon when there were these rupture moments where people were ready to escape those walled gardens versus when Blue Sky was available, when there was a rupture with Twitter. And I feel like the, just the familiarity and the ease of the user experience seems to have a really big effect on whether people that just want access, even if they're not as interested in that underlying agency, actually opt into those systems.

Alix: And I just wondered if you had thoughts on like. I don't know. Have your politics changed about what should be expected of [00:14:00] individual users? Uh, should they, should they be able to just be like normie users who happen to benefit from decentralized systems, or do you feel like everyone should experience a bit of pain to, uh, engage?

Alix: Oh,

Mike: no, I'm, I'm definitely not, not in supportive pain. Um, so, so, so ease of use and convenience is really important. And I think we know that, right? So those two things can be intention, but they don't need to be, right? And so I think. The key thing that we discovered was that, yes, when you have a decentralized system, you know, in an open system there can be pain and there can be setups for the people who love that stuff and who want to tinker, can go and tinker.

Mike: But you know, part of the reason why more centralized services came in was that they did provide a better user experience. But I think we're now at the point where we sort of. We've learned from that. Right? And, you know, these systems evolve and, and the, the incentives and the structures and everything evolve.

Mike: So part of Blue Sky's founding ethos [00:15:00] was we're going to build a decentralized system that. It's important that it's decentralized, but it's not important for you to know that it's decentralized. And that's actually really, really important because I think the problem with a lot of the more recent decentralized systems was they wanted you to buy into the ideology.

Mike: And this is unfair, and it is very broad brush, and I don't want to say that like, oh, Mastodon wanted you to buy into the ideology. That's not true. Some people did. Some people didn't. But to really, it's an

Alix: important part of the identity, so I understand why they're like excited about it and want people to engage.

Alix: Like I get, I get where the it comes from. It's not coming from a bad place, I don't think. No, no. When it happens, not at

Mike: all. Not at all. And I, I, I wanna be clear, I'm not implying that it is, but you know what was very important to the people who built Blue Sky originally was that. The decentralization and the openness were critical to it, but not critical for the users to know about.

Mike: Right? So the important thing is that if you want to, if you wanna buy into that [00:16:00] ideology, and if you want to do the things that openness allows, you can, but you don't need to. Part of the thinking behind it is that just having those things there, the ability for it to be decentralized, the ability for other people to build, the ability to exit, which is an important thing to, to control your own data, to have the choice over how you want to experience it.

Mike: Also, just, even if nobody uses it, not nobody, but even if most people don't even know that they're there, it creates much better incentives, right? It creates very strong incentives. For Blue Sky itself, being one operator on the network, the first one, and sort of the biggest one for now, but not to be bad.

Mike: Right? Not to screw it up because the second they do something bad, it's so much easier for people to step outside and say, you know what? I like the people I could talk to here. I like the, the ability to have these conversations, but. I don't like Blue Sky anymore and therefore I wanna be able to step [00:17:00] outside and still be able to talk to people on the network.

Mike: And Blue Sky allows that. But that itself just creates the incentives for Blue Sky, not to screw it up and not to create a mess. And also creates this sort I sort of jokingly refer to as a technological poison pill that, you know, the fear is, well, what if. The evil billionaire steps in and says, oh, we're gonna buy Blue Sky and make it terrible.

Mike: Well now because someone can build Green Sky or whatever, right? The alternative to Blue Sky, and people are trying to build those now, which is great. The impact is that it actually. Disincentivizes, both blue sky from acting bad, but also a malicious actor from coming in and saying, I'm going to buy it and turn it to my own interests.

Mike: Right? Because the situation that we ran into over the last 20 years or so is that with these new centralized systems, they became so powerful that controlling them and controlling [00:18:00] the the dials. In terms of like what was sent to you, what ads did you see, what were you encouraged to do? What were you encouraged to see Becomes really valuable because that, that control that was controlling people's lives, it had media power, it had influence, power, and therefore it had political power.

Mike: So making it so that those systems are not subject to easy control by very powerful interests, I think is a really, really important part of the blue sky approach, but also one that users don't need to know about. They just need to know that they're having a better experience and they don't feel like they're being manipulated because they're not.

Mike: There are a number of different nuances in there, but I think it's important to tease those out.

Alix: Yeah, I agree. And I also think that it doesn't have to be zero sum. Yeah. And I feel like in the nineties that we were kind of digital cave people. Um, and I feel like there's been a maturation of the average person when they engage with technology in a way that I think.[00:19:00]

Alix: Make some of the kind of newbie experience different in a way that makes it possible for people to make all kinds of different experiences that people enjoy in a way that might not have been as likely or possible in the nineties when people were still figuring out what a mouse is and like keyboards and stuff.

Alix: Um, that's super interesting. So I wanna go back to. When you, 'cause you've already started to hint at this, but like when did you start feeling the tide of centralization? Like when people started trying to build wall gardens of services that they could draw people into and kind of not maybe intentionally at the beginning, trap them into, but ultimately that's kind of where it's led.

Mike: It's sort of interesting because it was both gradual and sudden. You had a few different things happening at once. Right. And so. I think at first the rise of successful internet services in the the early two thousands. Was seen as a really good thing and and super empowering. It was because they made it easy to do the things that were empowering to users and people sort of forget, [00:20:00] historically it was very difficult to talk to people.

Mike: Right? Like you could talk to people near you, but to communicate more broadly, right. It was very difficult and find

Alix: information.

Mike: Yes. I mean, all of these things were much, much more difficult and, and had very limited. Analogs to what the internet is today. And one of the key things about finding information and communicating was that to do it on a larger scale, you always ran up against some sort of gatekeeper, right?

Mike: So. In terms of content production, radio, tv, movies, music, newspapers, magazines, all of those were effectively gatekeepers, right? They let some content through, but very, very little. They were very, very selective. The average person had no way, you know, the best you could do was write a letter to the editor of your local newspaper, which they probably would not publish, but in rare cases they would.

Mike: The flip to the modern internet [00:21:00] was. That these companies were still intermediaries. They were still there. They're still middlemen. Like there was this whole thing for like, people were like, oh, down with middlemen. It's like maybe not. The middlemen were still important. But they initially certainly acted in the other way.

Mike: They were very much enablers of connection and information and community and that was incredibly powerful and really wonderful. And you know, the fact that I am still somewhat in touch with people I went to high school with 30 years ago is incredible. Most people lose stuff. I can

Alix: imagine that being kind of a mixed bag, but yeah.

Alix: Yeah, yeah. That's fair. That's fair.

Mike: Very much. Uh, I find it interesting. Yeah. We've gotten gotten over those interesting.

Alix: Sure, yeah. Yes,

Mike: we've gotten over the, the former slights and everything. Yeah. But, but you know, so I think that was really super empowering and I think it's very easy nowadays to sort of forget that part of it and to forget like.

Mike: Wait. You know, the internet was really, really [00:22:00] great and really, really useful for so many different things that a lot of people now are just like, no, it's all terrible. There's nothing good about it, and we should nuke the entire internet. I think that's. Obviously wrong, but then really what happened was a few different things and, and one which I've alluded to is that the successful companies became this sort of winner take all situation, and they were driven in part by Wall Street demands because.

Mike: They got big, they had to return money to their investors and they went public, and then they had Wall Street demands to every quarter that number has to go up. And so they started to look for more and more ways to make money, and that often meant trying to. Limit your freedom. This is the in acidification curve that that Cory Dro has talked about, where the idea is in the beginning, as an internet provider, you are enabling something wonderful for your users.

Mike: You're giving them value, you are providing more and more value to them. And then at some point you sort of [00:23:00] hit that limit to where that comes, and then you start extracting value from your users. And then when you. Can't do that as much anymore. You start extracting value from your partners because you're large enough to command some sort of take from the partners and you're just making the service worse and worse.

Mike: That was sort of the, the issue that came along. So it was this sort of gradual switch from this really, really empowering technology to one that started to take more and more from us. In order to give you that empowerment, and so often that was collecting data. It was locking you into certain things. It was doing slightly creepy things with your data, depending on how you felt about it.

Mike: The deal that people made to get access to these services was feeling worse and worse to a lot of people. So, you know, that was some of the recognition of, we went from these tools, which are still sort of doing the same thing that they were doing 20 years ago to today. It doesn't feel as good to be [00:24:00] using them because you feel like you're being exploited in order to get the power of those tools, and you are.

Mike: And so the question then became like, is there a way to get the value of these tools without the exploitation as part of it?

Alix: Yeah, and I think there were also, along that, along the way these kind of spasms as the, these technologies scaled where there was like positive Arab Spring, for example. Like there are these moments that kind of built up this overall feeling of progress that was independent of any one.

Alix: Company or platform. And that, I think that probably extended the period of time where people were patient. Yeah. With, with some of these dynamics.

Mike: Yeah. I think that was true. I, I would also say that there was an element that contributed, which I think is under discussed, that contributed to the things going downhill, which was amusingly around the same time as the whole Arab Spring stuff happening.

Mike: The companies themselves started to get more aggressive and doing things that. I now view as [00:25:00] incredibly anti-competitive and incredibly anti open internet. One of the clearest ones was there was a company called Power Ventures or power.com that most people have never heard of. Great. URL. Yeah. Yeah. But in the, the sort of late auts power.

Mike: Had come up with this idea of a kind of dashboard for all of your social media. This was at a time where there were still other social media companies out there, many of which people have forgotten their names, but there were a few other players out there. It wasn't just Facebook and Twitter was around, but Twitter was still early.

Mike: And so they created a universal dashboard that you would log into it, you would give it your. Logins to all of these other platforms. They would pull it in, give you a universal interface. You could cross post easily, you could see the content. But the key thing was that you could use Facebook without ever actually going into Facebook.

Alix: Yeah, like an early hoot suite kind of thing. It was like an [00:26:00] early

Mike: hoot suite. But you know, whereas Hootsuite is more like, it was more designed for like the marketing person, right? This was designed for the average user to just have your own dashboard. But one of the clever things that they were clearly leaning towards, so they didn't quite get, there, was this idea that you could separate your experience entirely from the underlying platform.

Mike: And Facebook freaked out about it and Facebook sued them and claimed that they violated the computer Fraud and Abuse Act. Which, if you know anything about the history of that, it's like a 1980s law that is very poorly written and very widely abused. It's in theory, an anti-hacking law. It came about in response to the movie war games, which Ronald Reagan apparently thought was a documentary or something.

Mike: There's like the whole history there of like, and it's basically

Alix: almost exclusively been used by bad actors since I feel

Mike: like, yes, almost exclusively. Now there are people who will, who argue. I'm sure there are some cases. Yeah. Argue at that point. [00:27:00] Um, but yes, it has regularly been used and, and the key part of it is that there's a bunch of parts to the law, but a key question is whether or not you have authorized or unauthorized access to a computer that you are getting access to.

Mike: Now, I would argue and Power did argue that they had authorized access to log into Facebook server because you, as the user had provided them your login, which is your login, and you have said. Power as my agent go and log in for me, but of Yeah. Does. Yeah. Right. Seems seems obvious. Yeah. Seems like that's the way it should work, but Facebook argued, no, this is unauthorized access and therefore it's violating the CFAA and the courts agreed.

Mike: And I think that was kind of the beginning of the end of the concept of the open internet before that. Before that it was very common for people to build services on top of other services without pre-authorization. [00:28:00] Right? But the way that Facebook was saying was like, okay, if you want to access our service, you have to go through our, our API, which means either you have to agree to a specific license.

Mike: Or you have to agree to pay. We reached this stage where the open internet where anyone could just build, started to go away because the threat of, well, we can take you down. Power Ventures went outta business. I mean, before the lawsuit ended, they ran outta money and they had to shut down. The lawsuit continued.

Mike: They still lost in in the end, but that was the beginning of the end where companies and a lot of the big tech companies suddenly realize like, we can use CFAA. As a way to stop people from building services on top of ours that could eventually turn into competitors, and so it became a very anti-competitive, anti open internet.

Mike: Use of the law that stuck around. There've been a few rulings since then that have sort of whittled back the worst version of the Facebook versus power end result, but it's [00:29:00] still really, really limited this concept of the open internet. And I think at that point when the big tech companies sort of recognized like we now have the legal ability to lock users in, we're gonna make use of that.

Mike: So they, they really went hard on. Now it's all about building our own walled garden and keeping people in as much as possible and collecting as much data as we can and then constantly pulling them back into our system. And I think that was the beginning of, of some really bad stuff.

Alix: That's a really interesting example.

Alix: I also feel. I hope that that ruling comes back to haunt them when they start asserting agent ai. Yes. Because I feel like, like it's basically, um, the entire premise of all of these ai Yes. You know, boosters right now that we're gonna have some type of authentication. Yeah, yeah. Anyway,

Mike: yeah. Oh, it's really interesting.

Mike: Yeah. And it's interesting to see how viewpoints change depending on which position you're in, in the market. Yeah.

Alix: Funny that, yeah. Yeah. Um, that's a really interesting example though. And I had not, I hadn't heard anything. I had never heard that before. So it's super interesting [00:30:00] thinking about at Proto and Blue Sky, how those incentive structures have been hopefully kind of redesigned and are significantly different than the kind of other giant players.

Alix: When you think about having one venture backed larger. Instance, on top of a protocol, that organization is gonna have the resources to do the development that changes the sort of technical specifications, documentation, like direction, roadmap of the protocol. How do you think about that power dynamic?

Alix: Vis-a-vis other actors that are gonna be building on top of the protocol or like, have you seen examples where it has been possible to kind of bootstrap a community that has more and healthier governance around who's in control of the direction of the underlying thing?

Mike: Yeah, I think one, it's still early.

Mike: Blue Sky is still in this phase where, you know, they're building stuff. And one of the things that we've seen with other sort of protocols that have come out of different places is like. If you start too early and [00:31:00] you completely free up the protocol way too early, then nobody's able to build on it because it's not ready for it.

Mike: And so there is this balancing act. Blue sky is trying to navigate that where the focus is on how do we. Create a protocol that anyone can build on that doesn't rely on Blue Sky, but also how do we bring in the users who will use it, and how do we encourage greater development? And so far, I think that Blue Sky's doing a good job of balancing those things, and we're very, very encouraging of third party developers technically building competitors to Blue Sky, but that to the team at Blue Sky, that's part of what they're judging themselves on.

Mike: Are we getting other people to build on it? Like we don't want everybody just to be on blue sky hardware and blue sky systems because that actually is a, a risk to blue sky as well. All the, the questions are like, is this a zero sum game or non-zero sum game, right? Like, [00:32:00] do we have to keep everybody to ourselves?

Mike: Like this is the thing that all these other platforms eventually got to was like, oh, we gotta lock people in. We gotta build more walls and we have to build more moats around us. Luckily right now, blue Sky's at a point where they don't believe that and they very strongly mission aligned, saying that one, not only do we not want that, but we want to build up the systems that make it impossible for us to do this in the future.

Mike: That is a risk, and this is why like some people are hesitant to use Blue Sky because they say like, oh, you know, it is the biggest player by far in the space and it currently has control over the protocol, though, that will change their efforts in place to sort of separate the protocol side of it from the company itself, the elements of the system that should be.

Mike: Open to everyone. The company has to get them to that point, but they're working on that. And so we're in this sort of unique time period where [00:33:00] yes, there is this element that people have to trust that blue sky is going to let go of the different elements that it controls. But it is actively working to do that.

Mike: And it has done a whole bunch of things to make it impossible for it to, you know, build up those walls and to create this walled garden. And we're seeing now, especially a ton of development of third party apps and services and it's amazing. I mean, literally last night somebody was showing off this totally new.

Mike: Totally unique UI with which you can engage in ad protocol content, and it's sort of like a game-like card system. You can like slide cards away, but they added in this ability to, as you're going through your feeds and swiping cards away suddenly. Things will appear that you can grab like a game, like you'll find something on the ground and take, I have no idea where it's going with that.

Mike: People are just building unique things and people have built [00:34:00] YouTube clones and TikTok clones and Instagram clones. But I like the thing that I found that I saw this last night and I'm suddenly blanking on the name. Because that's something totally unique and new. I think the first step that everybody does is build the clone, right?

Mike: Blue Sky's a Twitter clone. Sky Tube is a YouTube clone. You know, flashes is an I Instagram

Alix: with a terrible name.

Mike: Yes, yes. So you have all these people, but that's like the first step, right? That's the first step of vetting new technology. But then the next step is like, how do we take this and build something that is unique to it?

Mike: That the old system couldn't build. And we're seeing that now and then as that happens and as more of those take off and become more interesting and more successful, and then, you know, blue Sky itself is one entity in the wider atmosphere as they call it. Has less and less control over where the wider system goes.

Mike: That's where the exciting part is. And Blue Sky as a company needs to prove that they're willing to let go. And I think they've taken steps so far that [00:35:00] shows that they are, and you know, there's still more to be done, but you know, the company is 100% focused on doing everything it needs to do those things in ways that can't be clawed back.

Alix: How is it gonna make money? I know you're, you're a board member. I know that I'm probably not supposed to ask you that question, but I'm just curious 'cause I feel like, um, like I've been. Occasionally following what's happening with WordPress and WP Engine. And it seems like there's a different dynamic that emerges when there's like, it becomes infrastructure that businesses sit the top and like there's money involved and there's like B2B relationships and partnerships and licensing and da da da da da da.

Alix: And I feel like understanding if there's a vision around how Blue Sky's gonna make money, it then I imagine makes it easier to kind of see a path so that others could make money on the protocol without. Having to Yeah. Navigate some of that trickier stuff. So is there any indication of where it's gonna come from?

Mike: Yeah. And so Jay, who's the CEO, has talked about this a few times. We'll start to see some of these things relatively soon. Uh, it's [00:36:00] definitely a part of the plan, the ideas that have been discussed publicly. So I'm, I'm not relating any, any sort of private, no breaking news here. Yeah, that's okay. Uh, n nothing private or behind the scenes.

Mike: And, and to be honest, like all of the behind the scenes stuff is the same as what the public stuff has been. The company has always been fairly open about these things, so there's no secretive plan here, but there are a few different things. And so one is building in some kinds of. Subscription features for premium features and the general analog of thinking about it is the way that Discord has done Nitro.

Mike: It is not designed to be a really sort of pushy and annoying upsell, but. You know, there are certain things that Blue Sky does that cost the company a fair bit of money, like hosting videos. So if you wanna host longer and longer videos or post more and more videos, maybe there's a subscription that that gives you more ability to do more [00:37:00] things along those lines.

Mike: But the bigger thing and, and the more interesting opportunity, and Jay sort of refers to this as. Money chasing value, which is sort of seeing where we're providing value to users and then putting in ways that those can be monetized in helpful ways. And so the most obvious of these is allowing users and creators on the.

Mike: System to monetize. So creating subscriptions or support for creators on the platform. And then if Blue Sky is enabling that, it takes a small cut and everyone hopefully in theory, benefits from that. I think that's a very sort of simplistic idea, but you know, there are ideas to sort of take that concept even further.

Mike: But if we're seeing where the value is accumulating on the platform and we as a company can sort of help the people who are using the [00:38:00] platform, and again, you know. Unlike this sort of inify services where it's how do we extract more value? The idea here is how do we enable more value? And if we're enabling more value, we can take a small cut of that value along the way.

Mike: That's basically the way the company's thinking about it. The actual rollouts of those different services and how they work are things that are still being worked on, but we should start to see some of those fairly soon.

Alix: Cool. Okay. That makes sense to me. I'm glad you didn't say. Ads, obviously.

Mike: Yeah. I mean, and, and like Jay has also been pretty clear on this, where it's just like, there are, in theory I'm adding the, in theory there are in theory ways to do ads that are non.

Mike: Horrible. Almost everybody doesn't do that. Almost everybody does ads in ways that are horrible and extractive and privacy destroying and all of those things. And the company is pretty well aligned on like, we're not gonna do that, but we're not against. The concept of advertising in ways that it makes sense and, and we are [00:39:00] seeing some experiments from third parties on the network that are not Blue sky to do interesting things.

Mike: So, you know, one example is Grays. I don't know if you've seen Grays Social, but it's a tool for helping people build custom feeds, which is one of the very, very cool features on Blue Sky. There are a few different companies that allow you to build custom feeds. Graze Social allows you to put ads in your feed, so you could create a custom feed.

Mike: You know, one of the ones is like a, there's a news feed that somebody created and they've monetized it so that you can buy ads in that feed. So if you use the newsfeed that was created by Graze, graze allows people to buy ads within that feed, and you'll see the ads and Graze gets a small cut and the creator of the feed gets a cut.

Mike: It's an experiment that people are working on, and that's got nothing to do with Blue Sky itself, but it is showing there are places where ads could work, right? And you could also see things where, because we have the custom feeds and we have starter packs, as you know, ways [00:40:00] to help people find stuff. You know, are there opportunities, and I'm not saying that the company's gonna do this, but are there opportunities to say like, here are promoted feeds, or here are promoted?

Mike: Starter packs. There are ways that you could do that. That is not like we're going to spy on everything that you do online and tell you sell you

Alix: athletic greens. Exactly.

Mike: Right. Mattresses and athletic greens, right? Yeah. So, so like, I think there are, there are interesting ideas in ways to use sort of like more contextual.

Mike: Types of advertising as opposed to intrusive we're, you know, building a profile on you kinds of advertising. And so I think that there's possibilities that, that those will turn up at some point. But, you know, again, very, very careful and conscious of how do we do it and doing it in a way that doesn't harm the user and doesn't do bad things for the user.

Mike: And again, the incentive structure sort of. Requires that the company think those things through. [00:41:00] Because unlike other platforms, we have no lock-in. By design, we've designed it so that if Blue Sky does something and starts putting ads everywhere and you hate it, you can go somewhere else and still talk to everyone on Blue Sky.

Mike: And so like the company sort of deliberately put itself in this position where it says we can't do something that makes the experience worse for users. And so that hopefully. It keeps the company from, from doing anything bad like that. And then like, you know, you have people like me on the board right now, at least who, who will yell at them if they do something bad like that.

Mike: But, but you know, the whole idea here is that we shouldn't have to rely on the board who is very aligned on this doing that,

Alix: but you're like the big bird in the, in the coal mine instead of the canary in the komanya.

Mike: Yeah, but, but like, you know, look, we can always be replaced and, yeah, sure. And so the idea is like we have to build up the entire system so that the incentives are in place, that if we are replaced and if, you know, the evil billionaire comes along and takes over the company, that [00:42:00] even, so the incentive structure, you know, hopefully the incentive structure is such that the evil billionaire can't take over the company.

Mike: But even if they can. The fact that everyone can then move to an alternative, continue to communicate. It's not a situation like Twitter where you had to like go off into this, rebuild everything and rebuild everything. With this, we should be able to just move everything to a different server in a click or so, and.

Mike: Still have all the, all your community, all your content, all your contacts, and just remove it entirely from Blue Sky, the company.

Alix: Yeah. Which I think you're, I've heard you comparing at proto to like email protocols and I think that's a really, really great way of understanding it, that you can move email clients and like go somewhere else with your email quite easily.

Alix: I also think that the more people that build on this. The more people are gonna be specialists in building in net proto. And I think people don't realize it in like, that means that in like three years time, there's gonna be people who are so proficient that if that were to happen and people were to get pissed, [00:43:00] they're gonna build something overnight that you could easily move to.

Alix: And that piece I think isn't the case. It doesn't exist in other ecosystems,

Mike: and we're reaching that point, right? I mean, and people complained early on that like, oh, you know, you say that there's a right of exit, but there aren't third parties providing federated services, or something along those lines that yet.

Mike: And that's not really true anymore. Like, you know, we're seeing, I mean, black Sky is a great example. I mean, that started as just originally it was just sort of a custom feed. Then it became a custom moderation service, and now he's got his own entire, uh, personal data server where he's moving people over.

Mike: So users, you know, I think that this week

Alix: I keep watching. Yes. I keep watching everyone being like, see you.

Mike: Yeah, exactly. But, you know, but it's not right. They're just moving over to Black Sky, but they're still. On Blue Sky in terms of like being able to communicate with Blue Sky. They don't lose anything, but they're no longer on on Blue Sky hardware.

Mike: And like in a normal world, you know, a company, we should be upset by that. Like, oh, it's sort of

Alix: you'd be losing users. Yeah, we're

Mike: losing users. Like, no, but we're like, this is [00:44:00] exactly what this is designed to do. This is. Proving out the general concept and when we sort of believe, again, in the non-zero sum world that this, it's a rising tide for the, the entire concept of the atmosphere as, as it's referred to.

Mike: We wanna see all of these services. It shouldn't just be on blue sky to build it. We wanna see. People building these alternative services and we think that we all benefit from that. Everybody benefits when there's a variety of different services, even if they're sort of directly competing with each other.

Mike: But I think the network effects overall are really positive and, and you know, we, we think that's really important.

Alix: Okay. Well, as a final thought, with these sort of new shoots, these green shoots of decentralization, both in social and in other places, like what are you most excited about in the next few years with this space?

Mike: A bunch of different things. Um, I'm trying to think which one comes top of mind. First of all, I think we're starting to see the cool experiments, right? I may have mentioned this earlier, but like the original thing is just sort of [00:45:00] recreating the old stuff and seeing what we could do. The next stage is like figuring out what is native to this system to decentralized.

Mike: Systems and you know, the atmosphere in particular and beginning to look for ways to do things that are unique to it. I think we're entering that moment and that on its own is very exciting. I think that there are some really interesting opportunities, again involving AI to do some really cool things as well.

Mike: I'll give one quick example of that, which is. There's this small company that was started by a guy named Dave Wilner, um, called Centropy, which I don't know if you're familiar with it at all. I know Dave Wilner, but I'm not familiar with his new

Alix: company. Yeah.

Mike: Centropy is his new company, and he spent a long time originally using AI tools to try to do content moderation online, and the idea was could you give it a plain English?

Mike: Policy, content [00:46:00] policy and have it judge whether or not the content violates or not. And he spent all this time writing to show that he had to write content policies. 'cause the one thing he discovered is that most people are not good at writing their own content policies. And one thing that Dave has tremendous experience, decades of experience in doing is writing content policies.

Mike: So he was writing like dozens of different content policies with different levels of strictness and all this stuff. And the idea was like you could adopt it. The thing that they've now done, which is absolutely fascinating, is do the reverse, which is take a corpus of data, let's say a thousand, 10,000 pieces of content, say which ones.

Mike: Should be allowed and which ones shouldn't be allowed. And then have it write the policy for you and in some cases, improve your policy. So you could have, this is what my content policy is, and then here's the content. Here's a bunch of content. These are the ones that I say are okay, these are the ones that I say are not.

Mike: And it will look at your policy and say, your policy's not good because your policy would [00:47:00] say this one should be allowed, but you're saying it shouldn't be allowed. Here's how to write a better policy.

Alix: So it's like reverse engineering, inter coder reliability.

Mike: Yes. Okay. But, but now, like you can begin to take that a step further, right?

Mike: Like, so you can see the, the advantage of that for like a service provider who wants to write better policies or a trust and safety team that wants to create better policies, but why limit it to the platforms themselves? Why not have individuals be able to do that? So why can't. I take this system and say, here's the content I want to see on Blue Sky.

Mike: Here's the content I don't want to see on Blue Sky. Write me a policy that creates the perfect algorithm for me, not the perfect algorithm for whoever controls the algorithms, but you know, if I can just do that in plain English and say, this is what I wanna see, this is what I don't wanna see, and give it examples.

Mike: And so like every time I come across content that I don't wanna see again, take that out or like, you know, you can start to put in flexibility to it where it's like. [00:48:00] On Mondays, I'm okay with political content, but on Tuesdays I'm done with it. You know, like, you know, who knows what it is. But you can begin to use these tools to actually provide yourself an overall better experience rather than relying on somebody else to to choose what you can do.

Mike: And so I don't know exactly where that goes, but I'm beginning to see interesting things coming out of that.

Alix: That's super interesting. I mean, I think I, I fundamentally have, um, deep distrust of the quality of transformer architecture producing accurate, uh, consistently accurate things. But that said, I think that there are new challenges of scale that if we're not going to just break up these giant platforms, or basically say scale is not a possible thing to manage, we're gonna continue to have these systems operating at scale.

Alix: We're gonna have to come up with innovations on the side of making. Governance choices because it's just not working the way things are are happening. And I think that it would be great if there were efficiencies we could find that reduce the amount of strain on individual people who are having to make these decisions and like [00:49:00] going through a bunch of muck that they shouldn't have to see.

Alix: And also anything that we can do to improve the efficiency of like. Dying trust and safety teams. I dunno. I feel like, yeah, I don't wanna, I mean, I feel like it's good. Um, yeah.

Mike: Yeah. But also, like the powerful thing here, I hope is putting more control again in the hands of the, the users, which, you know, I think.

Mike: Is potentially very, very empowering. You know, it all comes down to, again, like who's turning the knobs, right? Who's turning the dials and, and determining what it is you can and can't see. And you know, there's this concept of that that I've heard in a, a few different versions of it, but the one that I like the most is there's a difference between what you want to see and what you want to want to see, which is like.

Mike: If you ask people what their intentions are, they may tell you one thing, but then what they actually do is different. I wanna have

Alix: a salad for lunch.

Mike: Exactly. You know, you, you've got this concept right? And [00:50:00] so like the more that we can do to say like, well maybe, you know, it doesn't have to be perfect and it doesn't have to fully rely on like the salad for lunch approach, but like if you can tell a system, like, my intent here what I would really like to get out of this.

Mike: What would be most valuable to me is this, and at least take that into consideration that like, yes, please try to feed me a salad most days. Occasionally you can give me, you know, a slice of pizza or whatever, but most days I really should be having a salad and, and sort of lean it in that direction. The fact that help me help

Alix: myself.

Alix: Yeah, exactly

Mike: right. As opposed to like. Some billionaire who has to satisfy Wall Street, who knows that it's much better for him if you eat pizza every day. And I think, you know, a system where you have at least some more ability to control the experience, I think that leads to a better place.

Alix: Yeah. I think my takeaway from this conversation is that you want a choose your own adventure internet.

Alix: Um, truly choose your own adventure. I feel like we can get back there. I don't know. Yeah. Maybe I'm [00:51:00] naive. Um, I feel like, yeah, I, I feel like we should,

Mike: and again, like to, to be clear, just to put emphasis on this, like I don't want everybody to have to like sit there and go through the pain of doing that.

Mike: Like, I think defaults still will matter quite a bit, but the more that we can do to enable this possibility, just also then creates better incentives so that even if you just accept the defaults, that the defaults themselves will be better. Because if they're not, then people will leave.

Alix: Well, I think that is the best place to end it.

Alix: Um, thank you so much. This has been wide ranging, uh, and, and wonderful. So thank you.

Mike: Yeah, thanks for having me. This was a really fun discussion.

Alix: All right, so that was the first episode of No Star. I hope you learned something with that whistle stop tour of sort of where we've been and where we're headed. Up next week is Rudy Frazier and he's actually building stuff on at Proto Rudy runs Black Sky Algorithms, the company that developed Black Sky, the app.

Alix: It's kind of like another instance, another Twitter clone [00:52:00] on at Pro. So it runs on the same protocol as Blue Sky, and it's just super interesting. Rudy essentially wanted to construct a community moderated space on at Proto that would allow black users to have more control over what they see and how they connect with other people.

Alix: That was kind of the beginning of his journey. Of coming up with super innovative solutions and conceptions of what at Proto might be, and we get into that next week with Rudy. So thank you to Georgia Iacovou for coming up with the idea to do this series, and also as ever helping produce these episodes and to Sarah Myles who makes everything sound so wonderful.

Alix: But with that, we will see you next week.