← All episodes

An Internet Worth Living In

With Jason S. Canon & Aaron Ross Powell

We spend a substantial part of our lives online. The design of those spaces shapes how we feel, how we treat one another, and what kinds of communities we can build.

Listen in your podcast app

Show notes

We spend a substantial part of our lives online. The design of those spaces shapes how we feel, how we treat one another, and what kinds of communities we can build.

Jason S. Canon and Aaron Ross Powell explore what made early internet forums feel welcoming, why social media so often feels exhausting, and what we can learn from the way we inhabit cities and neighborhoods. We discuss Bluesky, Reddit, and the possibilities of AT Protocol, including the prospect of participating in the same conversations through very different online environments.

The conversation also turns to the relationship between our online and offline lives: how platforms can reward anger and doomerism, how those habits might carry into our relationships and politics, and whether better design could help create healthier patterns in both directions.

Audio produced by Landry Ayres.

Transcript

Jason S. Canon: Aaron, it was a busy summer for both of us. One of the interesting connections between our travels and our work that took us all over the place might be design problems. Or at least that's what I heard in a lot of the stuff you were thinking about over the summer. So catch us up.

I know you were thinking a lot about platform design and the state of things like Bluesky and distributed protocols. What's interesting in that world, and what have you been thinking about?

Aaron Ross Powell: Everything is interesting in that world. Part of the reason it's interesting—and this is a preamble to how it ties into the ideas we talk about on this podcast—is that part of what we explore is how we live in the environments we find ourselves in and share with others: the attitudes, perspectives, practices, and behaviors we bring into those, and what those mean for our delight, enjoyment, satisfaction, happiness, and so on. But also, in the aggregate, what sorts of societies those create, enable, support, or undermine.

We've talked about that a lot in terms of the geography of the real world and the institutions that exist within that geography: towns, cities, governments, organizations, and so on. But all of us spend a lot of our lives, probably too much of our lives, in digital spaces, in virtual geographies of sorts. The chat groups we're in, the friends we're texting with, the social media platforms we're on are all digital parallels to the physical and social environments we exist in in the meat world, in real life.

It's not just that the issues in one also exist in the other. There's this extra feature of the digital lifeworld, so to speak, which is that it is incredibly dynamic in a way the physical world just can't be. You can create a new social media platform in a matter of weeks. I'm not a developer. It takes a while, but it's a relatively fast process. You can create this thing, deploy it into the world, and lots of people can join it, explore it, and spend a lot of time in it, in a way that you just can't create a new town or a new city. You can try, but it's a much more involved, longer, and more expensive process.

The structures, the laws, whatever they happen to be, of those digital spaces are built in code and applied through that code to the people who use it. You have an interface for using this thing, and if the interface changes tomorrow, that changes how I use it, and therefore how I interact with other people who are using it.

You get similar sorts of design problems. One of the things we talk about is: what are the rules and institutions that shape how we live our lives in societies and governments? You can experiment with that kind of institutional and structural design very quickly, very inexpensively, in a way that's much more immediately applicable. You change a law in the real world and nobody follows it, or whatever; the code enforces it.

It's not just that we spend a lot of time in these spaces and should think about them the way we would think about real-world spaces. There are also opportunities for experimentation and reform, and for thinking through institutional and social-norms design. That's the background of why it's not just that I think the tech is cool, but that I think the tech is a really interesting way to explore the kinds of questions we talk about on this show.

Jason S. Canon: One of the experiments I know you've been thinking a lot about—and not just thinking about, but doing actual work in—is the Atmosphere. Can I call it the Atmosphere?

Aaron Ross Powell: Yes.

Jason S. Canon: Okay. Tell us what that is, why we should care about it, and why it's sexy.

Aaron Ross Powell: The internet isn't great right now. One of the reasons it's not great, or at least one of the reasons I'm nostalgic for the internet I experienced in high school or college in the late '90s and early 2000s, is the rise of platforms. These are Facebook—before that it was MySpace and Friendster—but now it's Facebook, X, Instagram, and so on. Those didn't exist when I was in high school and college. I think I got a Facebook account when I was in law school, so I'm dating myself here.

Those are very different in terms of how we interact with the internet than the way things used to be. A lot of that difference is centralization and a kind of homogeneity, I suppose. Right now, when you interact with a platform—and most people's interactions online come through these platforms—you're using a service that looks the same to you as it does to everyone else. You're posting into it in a way where it owns everything you're doing and displays it in line with everything else.

This brings benefits because it's very easy to connect with people, to get what you write, the videos you create, or the photos you take in front of all sorts of people very quickly and at low cost, or free to you. There are cool things; there are reasons these platforms dominate. But that centralization also brings real problems. One is control by bad actors.

Elon Musk buying Twitter, turning it into X, drove a lot of people off that platform and caused them to lose what's called their social graph, which is essentially the people you're following, the people following you, and those connections. You had to start over somewhere else if you didn't want to be there.

Or take the economic incentives: they give this stuff to you for free. You're not paying for it. They have to extract some sort of economic gain from it. So they start doing things to extract value from you—selling your data, increasing ads, whatever it is—that make the experience of being on these things worse.

The internet didn't used to be that way. It used to be weirder, and it used to be decentralized. People would build websites they hosted on various hosting providers all over the place. Those websites were very strange, very obsessive, very niche. You had smaller communities, web forums, and so on. It wasn't everybody shoved into one place where everything looked the same and everybody was performing for each other because that's how you maximize reach and engagement.

The Atmosphere is an interesting way, in my mind, to get us back to the better world there used to be while taking advantage of a quarter-century of technological development. Very simply, this is the technology behind Bluesky, but it does a lot beyond Bluesky. It's what's called a decentralized open protocol for connecting to each other and sharing content around.

I won't go into all the technical details, but the main idea is that you own your own place on the internet—call it a personal data store, data server, a PDS. When you create an account, it's owned by you. You have the cryptographic key, which is just—I think you just log in; it's your password—but you own this space. It's like having your own website on your own server.

Everything you create—all your posts, likes, connections, anything else you might write, all the photos—lives in that, belongs to you, and is under your control. Then this network can tie those together into something that looks like Bluesky, where it's a social media platform you can use like any other. But my personal blog at aaronrosspowell.com is powered by this, so all my posts and essays live in it. The webpage for The Inner Life of Liberalism, liberalpractice.org, is powered by this, so all our posts and episodes live in this thing.

The main idea is that it's all interconnected in really interesting ways, but it's not owned or controlled by anyone. If you sign up on Bluesky and then Bluesky turns evil or collapses, you don't lose anything. You can move somewhere else, maintain all your connections and everything you've created, not lose any friends, and interact transparently.

It's a way to do a couple of things. In a political context, it radically decentralizes power. Facebook, Twitter, and so on are authoritarian regimes. The regime owns everything: the infrastructure, the rules. Your only option if you don't like it is exit, but exit is very costly because you essentially lose all your friends and everything. You can't take anything with you when you leave—your social relationships, the content you've created, and so on. It's authoritarian in that sense. This is not; the technology prevents it from being authoritarian.

But it also means—and this gets to some of the themes in our work here—that you can experiment very quickly. If I want to build a new interface for using social media that, say, ranks posts differently, doesn't show engagement metrics, or is structured around building conversations, I can just build it. I don't need anyone else's permission.

Anyone who logs into it uses the same account they use for everything else in the Atmosphere, which is the name for all these things built on this protocol. They automatically bring all their network connections and all the content they've already created. You can experiment very quickly with new ways of organizing and interacting online, and do it in a permissionless way, where you have total control and that control can't be taken away from you. Monetarily, it's very cheap to do.

I'm excited about it because there are a lot of people in this space—millions and millions using this thing—and it enables us to think about how we build better, healthier spaces online and then just do it. But without the tremendous startup costs of needing to start something new and convince a whole lot of people to use it instead of the other stuff. People can try it out, keep using all the other things they're using, and everything interacts.

It's almost like you start a new city, and people can live in it but also stay in the old cities they're in, if that metaphor makes any sense at all. There's a very good China Miéville book called The City & the City that's kind of about this, although this is a more hopeful vision than what he articulates. If you care about how we design really interesting spaces online, and whether we can do it right now, this is, I think, the best playground for doing it.

Jason S. Canon: Two problems: one I think what you're describing can really help with, and one my time on Bluesky has inspired, and I'd love to hear your thoughts on. Both are about the power of decentralization.

To make this a little more concrete for listeners, both you and I, in our younger days, in the younger days of the web, started websites. You had a forum about tabletop RPG games. I started a website about hacking the Mac operating system to work on PC hardware when that was starting to be possible in 2006.

I don't know what your experience was like, but my experience was that there was something unique about that period of online conversational life, when hobby-ish things and niche interests tended to be based in these fora. What made it special? It felt pretty good. This wasn't the case for all the fora you could subscribe to; some were really nitpicky, and the vibe wasn't great on all of them, depending on the hobby. But a lot were really great, very welcoming, especially welcoming of newcomers. You really got the sense there was a community there.

I would spend hours on these fora and get to know people in a funny sort of way. There'd be a handful of people whose handles you knew, and occasionally you'd meet up with them in real life. It was as close to a virtual community as I've ever experienced. At least my experience is that all of that died with the rise of the platforms. Basically, Reddit cannibalized all the best parts of forum life. Many of those fora still exist, and they still have pretty good SEO from decades of thinking about this stuff. But for the most part, they're just a shell of what they used to be.

Facebook groups are very strange and detached, and everything about Facebook is terrible: the interface and the clothes Zuckerberg is wearing these days. Just up and down. It feels like you and I experienced something in those adolescent days of the internet. We won't go all the way back to IRC—that sort of predates me a little bit. I still used to use that, but anyway. We got to experience something very cool that encouraged genuine connections in a way we've lost.

First problem: do you think something like the Atmosphere can help us reproduce that, or get it back? I know Bluesky itself is experimenting with some of these things right now.

But Bluesky is an interesting test case for what I take to be the second problem of decentralization. Unlike many of those fora, the general-access nature of Bluesky, and the way it took off as a lot of people exiled themselves from X, has not reproduced that welcoming atmosphere for newcomers. I think even those of us who are big fans of Bluesky as a platform appreciate that it can be a tough environment, depending on the day, the issue, and where you're intervening.

I'm curious what you think about that feature of the platform as it currently exists. Is that an artifact of how it came about, where we are, and the political ideologies that make up the soup of people on Bluesky? Is it something we should be concerned about as platforms continue to distribute themselves in terms of where people hang out?

Will these communities become increasingly—not necessarily hostile to outsiders and outside views—but less welcoming than perhaps things were when it was just one platform everybody had to use and you could create your own little microgroups as you went? How do you think about the ways decentralization, using Bluesky as a test case, will affect the quality of the connections we build going forward?

Aaron Ross Powell: Man, there's a lot there. Let me think out loud for a little bit, because there's an interesting way these questions interact, too. It's interesting that you raise Reddit because Reddit is centralized and authoritarian in the ways I described. It's run by a single company. It's a closed platform in the sense that you can't fork it, take parts with you, or modify it. You're using their service as they built it.

At the same time, of the various platforms online, I think Reddit is probably the healthiest—or at least the one where, if you want to enjoy yourself, you can figure out how to enjoy yourself and connect with people in a way that works a lot better than other platforms. It's because it enables these smaller communities. You can join a bunch of subreddits on topics that are very large and of broad interest, like the NFL, or very niche, fussy, obscure, and pedantic. Maybe that's what you want, but you can have those communities and put yourself into them.

I think one of the reasons that works better, and why the forums in the past were largely a better experience—I got you beat on IRC. I was doing BBSs. I was dialing into TAG and Renegade BBSs and using FidoNet to talk in forums. They weren't even web forums; the BBSs would dial into each other in the middle of the night and exchange messages. But it was smaller and interest-based, community-based.

This gets to one of the things I think about. If you live in a big city and you're the kind of person who enjoys being in a big city—which I think is all right and proper people; big cities are awesome—the experience is extremely diverse, extremely dynamic, bustling, almost overwhelming. Pluralism turned up to eleven.

But it's a comfortable kind of pluralism. You basically experience it as background activity, the way music or other conversations in a coffeehouse might be. You're not directly interacting with every last bit of that pluralism at all times, and there are social norms around that.

If you walk through the Loop in Chicago, it's very diverse. You're seeing all sorts of things: weird people doing weird things, people who are very different from you doing things very different from what you do, all happening at once in this visual and auditory field. But you can be alone with yourself through that. You can have your headphones on or sit on the train, and those people aren't in your face. They're background.

Then you can go into your home and invite your friends over, or get together with your friends at the park, and have your little community within this broader space. You can drift between those little communities. That's how you experience it. You're not interacting with everyone in the city of Chicago all at once.

In a digital space like Bluesky or Twitter, people post publicly, and any post can be seen by anyone and retweeted out to thousands of followers. danah boyd says you can run into what's called context collapse, where communications internal to the norms of one community get picked up by another that doesn't understand them and gets [unclear]. There's all of that.

To analogize it to a city, it's like walking through the Loop in Chicago with a friend and having a conversation, and everybody in the city could hear it. Not just listening, but popping in with their own thoughts, getting up in your face and yelling that you were wrong. It would be miserable. I would give up my liking for cities and move to a rural community where fewer people would yell at me. It would be awful. But that's kind of how the norms function in these spaces where everyone is packed in.

There are millions of people. It's like being in a web forum with millions of people, where any post can be seen by all those millions and you become the main character for the day, or whatever it is. It's a different set of norms. You could have different norms there. You could say, "Yes, I can read all these things, but I'm not going to reply to people who I don't [unclear]."

That might make these spaces a lot more pleasant. There can be settings on your posts, like "Only my friends can reply," and that can certainly help. But the structure leads to a different set of norms and incentivizes a different kind of behavior that, if we looked at it in the real world, we'd say would be miserable.

It's not that this is unique to these platforms, that they're producing something different where people are meaner. I think there are different interaction norms playing out. If you go smaller, if you join the fountain pen community on Reddit, other people don't just pop in and start talking about the NFL or yelling at the fountain pen people that they ought to buy ballpoint pens. We recognize that we're in a smaller community, the people share this interest, and that's the topic of conversation. It feels better, richer, and deeper.

I think Reddit's a good example of how to do it better. The question is: can you build Reddit in a way that's not authoritarian, that's decentralized and open, but preserves the interesting ability to be in multiple communities at once and see all their things in one place? The answer is yes. These technologies can enable that.

One way to get healthier, I think, is to reconstruct in digital spaces something that feels more like the interaction patterns of a big city, where pluralism is happening around you and you can participate, but it's not being forced into your face at all times in this interactive way.

The other thing, when we talk specifically about Bluesky—because yes, there are very toxic people there, and the community can have these prickly edges—is that part of it is a lesson about ideological spaces. Most of the people who came to Bluesky did so for ideological reasons. Elon Musk bought Twitter. They didn't like his politics or the social and cultural views he started privileging on his website, so they left. They viewed being on Bluesky as effectively an ideological project.

It was a cause. You were here as part of that cause, and there was an ideological alignment that came out of it. People on the far right didn't mind Elon Musk. People on the far left did, very much, so they're the ones who came over. There's an ideological narrowing. People view themselves as participants in an ideological space versus, "I just want to hang out and talk about knitting or football or fountain pens or guitar pedals or politics or whatever else."

It's almost like there's an attitude that you need to be engaged in this political project, which is the reason we're all here. That leads to really awful behavior because people start carving out ideological enemies. All the things that happen in radical political groups, where they splinter and turn into factionalism and infighting, you see happening.

If we look at Bluesky specifically, there is this additional issue: spaces that were intensely ideological to begin with can fall prey to a set of corrosive norms you're less likely to find in other spaces. It certainly exists elsewhere. Fandom cultures can turn very toxic in these ways. But the notion that we're all in a project, and therefore if you're doing anything that's not part of that project, you're not paying attention or doing something wrong, is, I think, very corrosive to healthy environments.

Jason S. Canon: I'm also pretty optimistic that we can fix those going forward.

Aaron Ross Powell: I agree.

Jason S. Canon: There have been a handful of cases in this period we're talking about—the past ten or fifteen years—where an environment was pretty toxic and then became considerably less so. Reddit was one of these. There was a span of several years when a lot of bad communities found their home on Reddit, a lot of pretty extreme folks in every direction, and fixing that was a real challenge.

The one that came to mind, and you and I mentioned this the other day—and we should figure out who to talk to about this—is the YouTube comments section. It used to be the armpit of the internet. All the worst people were saying all the worst things there. Then suddenly, sometime, they changed the comment algorithm, and now it's pretty good. I actually enjoy reading many of the top comments because they're often witty or uplifting, or teach me something about the video I may have missed.

That's a design choice that has considerably improved the lives of billions of people. You don't have to be a utilitarian to think it's pretty cool that just a few lines of code written differently can make vast swaths of humanity better on a minute-by-minute basis.

I'm pretty optimistic that we can figure out how to take a platform like Bluesky, which is great right now most of the time for most everybody, and accommodate all the ideological conversations people want, and also make it a happier place to be most of the time.

I love your city analogy. We do this all the time with cities, too. I was in Frankfurt over the summer. Germany did a ton of rebuilding after the war and made a lot of terrible choices about how they rebuilt their cities. The center of Frankfurt was hideous. It was the worst sort of—not even good brutalism, but brutal brutalism.

They came to a point fifteen or twenty years ago when that building was too old to keep around. They tore it down and rebuilt the city in a beautiful, attractive way, such that the people who live in Frankfurt love to go there. Tourists now love to go there in a way they didn't before. They transformed the experience of what the space was like, such that it now creates all these beautiful positive externalities for everybody who experiences it. I'm pretty hopeful about our ability to identify the right sets of design choices and do more of them.

The problem is that we don't really know what the design problem is until a new technology comes along that allows us to create something new, or there are external shocks. Bluesky itself was a huge advance over what came before. How do we improve Bluesky? Not just Bluesky, but how do we use that protocol to make better spaces?

This seems to me like the Wild West of social science in a really interesting way. It's the intersection of the social sciences, the humanities, and technology, and it's all brand new.

Aaron Ross Powell: Something worth mentioning—and this is where the city metaphor breaks down, or becomes very weird if we try to think it through into digital spaces—is that decentralization enables interesting and pluralistic ways to solve the problems you just acknowledged that cannot exist in the real world.

You talk about Frankfurt. It revitalizes its downtown in a way that improves it. But there is one downtown. If you're in downtown Frankfurt, you're experiencing the same downtown as the other people there. It's the same buildings, streets, weather, and so on. Even if you're having different conversations, the bulk of the lived environment is necessarily one thing.

We could say it's better than what came before, but it's not necessarily the best for everyone there. Some people might prefer taller buildings and a cloudier sky that gives a neat Blade Runner atmosphere: that's where they feel comfortable. Someone else might like solarpunk, happy stuff with plants everywhere. Neither one of us is necessarily right, but what are the vibes where we feel comfortable in the environment, even if we're experiencing the same conversations?

You and I could have a conversation, and you'd be happiest in the back corner of a library, while I'd be happiest at a coffee shop. But we can't have a conversation with each other in both places. In digital spaces, with decentralized tech, you can.

Again, I don't want to get too technical. If what this protocol does is have all the words Jason says live on Jason's server, and all the words Aaron says live on Aaron's server, and a protocol knits them together into ongoing conversations, we have lots of ways to have windows into that.

Right now, there's the network of short-form text posting, social media that most people view through Bluesky, the app. You download Bluesky on your phone and use it. But other people can write radically different clients to access that. There are some that look very much like Bluesky and function similarly—Blacksky, Eurosky, and so on—but they're completely distinct. They're not running on any of the same infrastructure, not owned by the same people. They're totally independent. If one went away, the others keep working.

You can switch between them, install both on your phone, and move between them, seeing the same posts and interactions. But you can radically experiment in that space. I want to have these conversations with these people. They really like the Bluesky lived environment. I prefer this other one that looks very different and has different features, but shows me the same conversations.

It's like I get to have the conversation with you sitting in a coffee shop, with you across the table from me, but from your perspective, we're in that library. We get exactly the environment we want. The conversations around us can be filtered differently based on our preferences.

It's not just that we can find the better thing; there doesn't need to be a better thing. Each of us can find what works best for us, but we're all in the same conversation with each other. We don't have to isolate ourselves to live in an environment that's exactly what we hope for.

In the same way, right now, if I like the aesthetics of macOS but you have a weird Linux installation, we can still interact over Riverside Recording or view the same blogs or websites, but from a space that's more our own and fits our preferred vibes and structure.

This takes that promise of a liberal society as an open society where we can each pursue our own bliss. In the real world, so much of our different blisses, because they have to happen in physical space, have to be separate from each other. In digital spaces, especially with decentralized technologies like AT Protocol, that doesn't have to be the case anymore. We can construct the most comfortable environments for ourselves without cutting ourselves off from others, which seems extraordinary to me.

Going back to your point about a simple change in YouTube's algorithm making life better for lots of people, that is really critical. We tend to think these things are just silly. The YouTube comments section is this silly thing. Who cares? But people spend time in there, and that time is actual time in their lives, the same as any other time they're spending. If that time is better, that's an actual, real improvement in the quality of their lives and social interactions. The more time we spend in these spaces, the more that's the case.

If we suddenly have this opportunity to optimize what makes each of us happier, without meaning that others have to take something that makes them less happy because there's only one option, that's an extraordinary gain in happiness. I think that happiness has knock-on effects for the broader social environment. If each of us is interacting in this network in a way that makes us happier than we would be if we were jammed into one form factor, then we're happier when interacting with each other.

That carries into the interactions, everything becomes healthier, and that reinforces norms of positive interaction. It's this feedback loop, all based on the idea that we can find what makes each of us happiest without cutting ourselves off from each other.

Jason S. Canon: It's exciting stuff.

Aaron Ross Powell: Yeah, I'm optimistic. This is another part of the reason I'm so excited about this stuff right now. You look at the political world, and there are some reasons for optimism in the long term, but it's grim. It's awful. I don't like reading the news. It's not good. It's lots of people who seem much more interested in hurting each other than helping each other, and fixing it feels so uphill.

It's an important project, and there's lots of work that needs to be done. Each of us should work to try to make it better. But it's not thrilling and hopeful in the way a lot of what's happening in these other spaces can be.

Jason S. Canon: It's so depressing.

I hadn't thought about it until what you were saying just now, but it very well could be the case that the real world, the meat world, is so depressing right now in part, and potentially even in large part, because our online spaces are so toxic.

I think a lot of us have thought about that in the context of worrying about misinformation and disinformation, the way populists and far-right figures use online spaces, and the terrible ways Elon Musk has turned X into his platform for white supremacy. All of those clearly have real-world political effects.

But one of the reasons—and this is true everywhere you go right now—so many people feel like the world is getting worse and things are really bad is that the way they consume their online worlds makes them feel bad. By feeling bad, I mean anxious, angry, fearful, and all the rest.

This may be obvious to people who have reached this thought before, but it's a new thought to me. I think most of us have drawn the causal arrow of online spaces getting worse because things in the real world are getting worse, like the rise of populism. But there's probably also a causal arrow running in the opposite direction: the shittier we feel because online spaces are bad, the worse our real-world interactions become. At a micro level, on the subway and in the elevator.

Does that jibe with your sense of how this sits?

Aaron Ross Powell: I want to say yes, but I also want to say this is an empirical question there's probably actual research on. I'm wary of saying, "That fits my priors, so yes, I think that's true."

It does seem to be the case that we are ruled by the very online right now, and that the cultures that sprang up in 2chan, then moved into 4chan, 8chan, Twitter, X, and so on—it often feels like the political environment is performing for subcultures online, and those subcultures are incredibly wretched. There is a causal direction from the online directly to the political environment being inflicted on all of us.

I think the causes run in both directions, interacting with each other. You see this in—I wrote a piece about doomerism a while back, about Bluesky and these people who, if you'd write anything about reasons for hope, would be like, "Well, they're just going to steal the election." If you pushed back, they were like, "You're just being naive. They're going to steal the election."

Those were beliefs about what will happen in the world, and those beliefs carry on. Maybe these people don't participate in the political process because what's the point? Or maybe they're critical of, mean to, or dismissive of people trying to make the world better because they've internalized this doomer mindset. There are real-world effects of it.

But a lot seems to be driven by performative doomerism getting the clicks. There's research going way back about how cynicism is often viewed as a more sophisticated position than optimism. If you have an online platform with metrics, you can immediately see—you're not just trying to read the faces of the people in the crowd when you talk—in this discrete, quantified way whether they liked it. At a granular level, this got twenty-five percent more likes than the other thing I said that wasn't as doomerist. It incentivizes more and more of that.

I'm always struck by how, every now and then, you'll come across someone who's known for being an absolute jerk online. Someone who knows them offline will say, "They're not like that. That's not really who they are. I know them in real life."

There's something to that because each of us can perform online. We're not exactly in real life who we are online, but we're not exactly in real life who we are in real life for different people in all sorts of contexts. You can't really compartmentalize the person that way. It's the thing of, "He's a really nice guy, but he's mean to waiters." Well, he's not a really nice guy. You don't get to be mean to waiters and be a nice guy. It doesn't work.

I think they're deeply integrated with each other. The fact that we spend so much time in these environments and online, and increasingly so, inevitably affects how we are in the world and how we interact with it. The way we interact with the world gets carried into—a lot of the doomerism is people feeling powerless and hurt by what's happening in the world. They express that and get attention and positive reinforcement in online spaces, so they do more and more of it.

I don't know what to do with that, but this is, I think, a core project of The Inner Life of Liberalism. What I just described is a series of feedback loops, behavioral incentives, and so on that, over time, makes the individual's life worse and society worse. If you can describe feedback loops, structures, and norms that have that result, there are ones that do the opposite. What are those?

We can talk about them in the real world, but they also exist in the digital world. There isn't a distinction between those two. If we can make things better in the digital world, we can make things better in the real world, and vice versa.

Jason S. Canon: It's the upward spiral that Maria talked with us about.

Aaron Ross Powell: Yes. That's episode seven of this show, for listeners who haven't listened to it yet.

Next in the archive: Friendship on the Knife’s Edge