Today, I’m talking to Kyle Chayka, a staff writer for The New Yorker, a regular contributor to The Verge, and author of the new book Filterworld: How Algorithms Flattened Culture. Kyle has been writing for years now about how the culture of big social media platforms bleeds into real life, first affecting how things look, and now shaping how and what culture is created and the mechanisms by which that culture spreads all around the world. 

Kyle wrote a fantastic feature for The Verge back in 2016 called “Welcome to AirSpace,” which was an early exploration of this idea. At the time, he noted that the aesthetics of physical spaces, like coffee shops and co-working offices, were being heavily influenced by Airbnb and Instagram, flattening global interior design into one singular and recognizable vibe.

Seven years later, Kyle’s argument is that AirSpace has turned into what he now calls Filterworld, a phrase he uses to describe how algorithmic recommendations have become one of the most dominating forces in culture, and as a result, have pushed society to converge on a kind of soulless sameness in its tastes. 

If you’ve been listening to Decoder, this is all going to sound very familiar. The core thesis of Kyle’s book — that algorithmic recommendations make everything feel the same — hits at an idea that we’ve talked about countless times on the show: that how content is distributed shapes what content is made. Whether it’s Google Search, YouTube, or TikTok, these platforms and their algorithms dictate how videos are shot, how headlines and articles are written and discovered, and now even how food, music, and art become popular. It is a powerful concept, and it’s an inescapable fact of life on the modern internet. 

So I was really excited to sit down with Kyle and dig into Filterworld and his thoughts on how this happened and what we might be able to do about it. You’ll hear us trace the origins of Filterworld back to the rise of modern social media in the 2010s and how this development has been accelerated by the deterioration of the open web, an erosion of trust in our institutions, and the frankly frightening speed and scale of platforms like TikTok. 

That said, it is not all bad news. Kyle and I talked about some of the upsides. You’ll hear us get into the challenges the analog world had. Big magazines, celebrities, and other gatekeepers controlled culture in a major way, and access now feels more bottom-up and democratized than ever before. More people are now able to make art and culture and find big audiences for that stuff — and that seems like a pretty good thing. 

I also made Kyle walk me through what he calls an algorithm cleanse. When he was writing the book, he tried to use the internet without algorithmic recommendations. And he found that it helped really open his eyes to what it means to develop your taste and truly explore new art and ideas.

Kyle is one of the most thoughtful writers thinking about technology today, and his closing thoughts on how we might be able to reconnect with the culture around us and find more organic ways to engage with it really hits if you’re one of those people out there who feels worn down by the apps on their phone. 

Okay, Kyle Chayka, author of Filterworld: How Algorithms Flattened Culture. Here we go.

This transcript has been lightly edited for length and clarity. 

Kyle Chayka, you are a staff writer at The New Yorker, you’re a longtime friend and contributor to The Verge, and you are the author of the new book Filterworld: How Algorithms Flattened Culture. Welcome to Decoder.

Thanks for having me. I’m a big fan of Decoder, so it’s cool to be here.

You have been in and around The Verge for a long time. And this book actually grew out of an article you wrote at The Verge called “Welcome to AirSpace” in 2016, so I applaud your commitment to the concept that algorithms are shaping culture, I’m very interested in it as well, but tell me just about that recognition, that first part of the story that you could see it happening.

It was even 2015, because I remember pitching it to our pal Michael Zelenko and then having a characteristically arduous editing process with him, which was great and resulted in the essay that we all know and love. But I was really observing this kind of flattening effect around then. I was working as a freelance journalist, I was traveling around the world quite a bit and landing in all these different cities, and I would notice that every Airbnb I stayed in had the same kind of aesthetic signature, every coffee shop I went to in Reykjavík or Kyoto or LA or Berlin all had the same stuff in it, and I just started wondering or almost being anxious about why all the sameness was happening, and that’s how I pitched that essay.

That first essay was about how places looked. It was very much: I’m going to all these places and Airbnbs and coffee shops, and the physical space I’m in starts to look the same. As you chased that idea down, did you come to a reason why that was? Was it just familiarity, was it Instagram, which plays a huge role in your book, or was it something else?

The conclusion I came to was that it was Instagram at that time, which was growing and growing in popularity. It was what made visual culture a kind of content that flowed through the internet, and I think it was just the new emergence of these digital platforms that were collecting billions of users around the world. I had the newfound ability to follow a random barista in China or a coffee shop founder in Copenhagen, and we’re all connecting on this very granular level and almost trading back and forth these visual tastes and aesthetics and trends, and that just has a way of letting the trends flow through that entire population around the world.

At that time, describe what AirSpace looked like because it bleeds into what Filterworld looks like now, but there was a specific aesthetic 10 years ago that you were describing.

I think there are different flavors of this generic aesthetic, and circa 2015, 2016, it was very much the white subway tile, reclaimed wood furniture that was sometimes made from rusty plumbing for reasons that I cannot determine. It was the emergence of avocado toast and cappuccinos with latte art. It was vintage posters on the walls. It was a very specific look.

I lived in Brooklyn at the same time you were writing about this. You describe that aesthetic in the book as “high Brooklyn lumberjack,” and there’s no better description of it. There was a moment in Williamsburg, where I lived, and you lived in Bushwick, which is right next door, where it was easier to meet a blacksmith than to go to a bank. It was tremendously easier to be like, “I’ve met several blacksmiths today, but there’s no banking available to me. I have to cross a bridge into Manhattan to go to the bank.” 

That got really commercialized really fast. That’s the thing you’re describing there is there is this weird little art community of people who are like, “We will make our own furniture out of pallets,” and that turned into something else that got commercialized and commodified really fast. Is that the process of things becoming Filterworld or AirSpace? How did the internet light upon, “Okay, this is what it’s going to look like”?

That era was post-financial crisis, people newly moving into Williamsburg and Bushwick, which were these industrial warehouses, and it was making a virtue of necessity where we were adopting hobbies that were super lo-fi, like being a lumberjack or making your own surfboard or making drinks in mason jars. Why that particular style became so popular I think was two factors. Brooklyn, at that point, was seen as the center of culture. It was the cool place to be for a lot of the world, and that has moved on by now, but also it was when social media was becoming more of a thing. There was this intersection of financial crisis austerity aesthetics and the emergence of a digital platform internet, and I think that was what did it.

As I was reading the book, the process of “we’re all going to pick this thing” seems to be the key determining factor and something big about the culture. You describe it as an averaging. There’s a band in your book called Galaxy 500, and their song that sounds the most like all the other songs is the one that becomes popular even though that song is a parody of those songs, and that flattening to me is really fascinating, but you have to come to some first consensus about what will be popular. And that’s the part that I’ve been trying to figure out for a long time. I think your book hits on it — that there’s some sort of statistical averaging of taste that occurs, and then everything becomes that thing. And it’s that first turn that I’m curious if you have perspective on now that you’ve written this book and you’ve talked about it so much.

It’s a problem that plagues me, too. It’s like, “How does the initial quirk happen?” And the metaphor that I’ve come to or that I believe in now is Darwinian evolution. It’s like the first fish that was like, “I’m going to try to crawl out of the water onto the sand.” That fish does that thing, and then all the other fish start doing that thing. What I find is that, in Filterworld, in this world of digital platforms and algorithmic feeds, one quirk goes viral instantly — a new adaptation, a new aesthetic flourish, can go from one person doing it to 100,000 people doing it in a day, whether it’s a TikTok sound or a dance or whatever, and so I think there are these artistic innovations that happen. A stylist chooses to bring back low-hipped jeans or whatever or a musician packs so much sound into the first 10 seconds that no one’s ever thought of it, like 100 gecs or something.

That’s a band, for the Decoder audience I need to... That’s a band.

[It’s] a kind of speedcore or whatever YouTube band. So some artists or stylist or curator or whatever figures out some quirk or innovation that works, and then everyone just piles on faster than has ever happened before. Everyone can just copy that thing, and I think that’s how the flattening happens. It’s like someone figures out a thing that works for everyone’s tastes, and then that thing becomes universal really fast.

You have in the book a meditation on the concept of taste, and you dive into the history of people thinking about taste and what it is and where it comes from, and that dynamic you’re describing is: someone has a little bit different taste or they make something a little bit different, and then suddenly, everyone else has the same taste.

The example that I actually really want you to talk about and help me understand is the Stanley cups right now, where I’ve read a lot of pieces just about what a Stanley cup is and whether they have lead in them and why all these people are buying them. But it feels like it is a Filterworld product that the algorithm lit upon — a cup, literally just a cup — and then everyone was like, “I am the cup. The cup is my lifestyle.” As you think about Filterworld, can you put the Stanley cup in the context of suddenly everything is the same?

I think so. The Stanley cup was interesting in that the chief marketing officer of Crocs moved to Stanley, and Crocs had gone through this viral trend of being adopted by a lot of influencers and TikTok creators and stuff, and so this guy has turned the same process or strategy with Stanley. And I think it’s partly that they seeded the ecosystem, giving Stanley cups to influencers and stuff, and it’s just the fixation that the internet has on one thing at a time. So a Stanley cup starts to become this go-to lifestyle accessory, first for Mormon bloggers — actually, that was an early adoption group — and then it becomes almost a kind of currency on TikTok and on Instagram, where, mimetically, if all my friends have this thing, I also have to have this thing. If all the other influencers are making Stanley cup content, then I also have to make Stanley cup content.

I think it’s almost like [how] on Twitter we could see this happen with discourse subjects. There was one subject of conversation each day, and either you were jumping into that conversation or no one cared what you were talking about, except now, we’re doing that with visual trends and physical objects. It’s like you have to be holding up that Stanley cup or no one’s going to want to watch your content, so you’re forced to participate in the meme, in the trend, or otherwise, you get ignored.

One of the tropes of Decoder is that distribution has an outsize impact on the actual content that people make, which is a very obvious idea, but we just come back to it over and over and over again because we spend so much time talking about platforms. You’re thinking about Filterworld as a broad concept — the algorithms on the internet shape the culture in some big way. But YouTube has a different kind of algorithm. Stanley cups are not happening on YouTube. God forbid, Stanley cups happen on X, what was formerly known as Twitter. That’s a particular kind of TikTok trend that bleeds into everything else. Do you think about the different platforms and their different aesthetics and what they prioritize and what kinds of culture they make?

I think they all have different flavors. My pet theory, I think, is that each algorithmic feed, each platform, generates its own signature culture that fits into it. So we’re familiar with Instagram face, the kind of influencer plastic surgery aesthetic. We’re familiar with TikTok influencer voice, which is the kind of monotone, syncopated, packing as many words into a sentence as possible. So I think there are forms of content that work for each different platform. And on YouTube, my favorite example of YouTube culture is lo-fi chill hip-hop beats to study / relax to, which is this ambient 24/7, never-ending stream of chill drum beats with acoustic instruments and electric synths behind it, and it’s all different artists composing these songs, but they’re just turned into this wash of ambiguous, semi-meaningless music.

And that works for YouTube in a way because you just leave YouTube on. It’s this streaming background. And that’s not how you use TikTok. TikTok, that wouldn’t work because you’re constantly flipping through the feed; you’re going to new videos. A TikTok video, to be successful, has to grab you and throttle your attention immediately, whereas this YouTube content can be ambient and chill and homogenous in a soothing way. So I think there are these quirks or forms that emerge from the structures of the platforms themselves.

Your earliest approach at writing about Filterworld was Instagram and Airbnb, what you were calling AirSpaces. As you expanded the concept into Filterworld, it’s everything. Different social networks take on different levels of prominence in the culture, so Instagram is still huge, but I would say it is not a driver of culture in the way that it once was. That role is now TikTok. How do you see that waxing and waning? Why do you think that change happened? Is it just young people use TikTok, and that’s it, or is there something else going on?

People get bored. I think partly it’s just as fashion trends change, technological trends change, and I think we discount that too often. When we use the same platform for five or six years, we tend to start getting itchy and wanting something else, and I think it’s also been this gradual evolution of the internet from text to more professionalized images, to audio and video, to TikTok, which is this kind of full-featured television, essentially. When we watch it, it’s as if we’re watching television, so I think the multimedia race has gone on and on, and that’s changed things. It’s also just, more and more of culture has moved onto the internet, I think. Digital platforms have absorbed different areas of culture that used to be more offline, whether it’s a television equivalent like TikTok or podcasts that used to be radio, over the past decade, more things have gotten more online, and I think that’s been a major shift.

I’m always curious on the effects that participating in the platforms has on people and creators. The Stanley cup, to me, is actually a really fascinating example. When you are a big distribution network and you say, “Anyone can participate here,” people will naturally gravitate toward exactly what worked for someone else. It’s the first, easiest, most instinctive thing to do, and so you can see why things spread mimetically. A bunch of kids are like, “Well, that worked for them. I will do the same dance and participate in a culture,” and that is a conversation in a way that I think broadcast programming directors who are professionals were like, “Well, we can’t just copy that thing. We have to do something different.”

There’s that element of, “Well, I’m paid to have better ideas than the next person” that a bunch of people working for free are like, “I’m just going to do the easiest thing.” You can see the tension there, but I’m curious if you see it particularly in a different way with TikTok because there’s something about the culture of TikTok that not only rewards that repetition but directly incentivizes it and makes repetition the actual content.

I have... I’m sorry I’m saying “pet theory” all the time, but another framework to use it to cover vocabulary.

That’s a great Decoder word.

A framework that I have is that on TikTok and in the current iteration of the internet, we’re all just middle schoolers running through the hallways, and so it’s like when you see some other kid wearing his hat backward or something, you’re like, “Oh, man, I’m going to wear my hat backward right now.” It all filters out very quickly. The model of culture we have right now is more bottom-up. Trends filter from a grassroots level upward and then get noticed, and I think TikTok rewards that repetition because you rehash someone else’s content in order to participate. It’s like making a new version of the same meme, as you were saying, is how you fill the vacuum of TikTok, and it is how you interact with someone else and have that conversation.

So rather than coming up with something new or trying to make a trend of your own, it’s like the core behavior is replicating a trend that already exists, and that’s incentivized by the algorithmic feed, by the kinds of aesthetic tools that you have, the recommendations of sounds to use, or video editing tricks to use. And I think you can see that replication happen all over the place as X became more algorithmic. You saw a rash of prompt tweets, just people asking for you to list your five favorite breakfast foods or something suddenly. Because that worked for some people, everyone was doing it and being like, “Name the five opinions that everyone else hates that you have.” And it’s not good content. It’s a race to the bottom, I think.

I do often wonder if you could survive a day in real life running around asking real people in person some of those prompts. If you just walked up to someone and was like, “What’s a secret you’ll take to your grave but you’d share with me?” But then, very popular Reddit threads are just that.

Yeah, oh my God. I’m thinking of the cartoon guy with all the swords coming at him, and it’s like, “Name one opinion you have that would make everyone do this,” and it’s like, this form of content only exists in these feeds. It is not real life. It is not a conversation you would ever want to have. I actually don’t care. Does anyone… These things fulfill no function except allowing you to express something. It’s like the person posting it is cultivating engagement, and you are yelling about your opinions. No one else cares.

[Laughs] There’s a deeper idea there, which is, I think generally people are desperate for connection. I think that’s a core piece of the human experience. There is a weird crisis of loneliness in the world, but then people have these full-bodied interactions online, and I do think that does something different to people. You are outside AirSpace, you watched it happen, and you have literally documented it. You’ve watched it turn into Filterworld, but the people who just swim in the water don’t see it, and that does feel like the normal way culture is made and produced. Do you see a difference in the culture of the 2010s when this started happening and the culture that’s being produced now?

Yeah. As I wrote the book, particularly as I was finishing it up, I almost felt like it was a history of the 2010s, very recent history, and it came to feel, for me, like it was a particular era that had a beginning and an end. There was the mainstreaming of social networks, the rise of algorithmic feeds and these massive digital platforms, and then the zenith, which is semi-ongoing, but toward the late 2010s, we’re all thinking that this is worse and worse. We’re less satisfied, the platforms are working in ways that are more exploitative. And then up to now, with the destruction of Twitter and just the general dissatisfaction, I feel like we’re coming to a new phase of things. So I feel like the culture has changed, and I think there’s this dawning sense that these are not doing good things for us.

Generally, being immersed in algorithmic feeds and copying what everyone else is doing and letting TikTok show you what it wants to show you is actually bad for you and is not leading you to a constructive place. So I think, in writing the book and with people’s reactions to it, I think I’m trying to catalyze a little bit of that dissatisfaction and be like, “No, let’s think about what this did to us, and let’s see what we could want that’s better.”

Alright, let me push you on the flip side of that. That’s us on the outside. The Verge came up on the outside. We exist as a publication to document this process. Your book is a document of this process. You talk in the book to creators who are the fish in the water, who started with zero followers on TikTok and are now very popular, and I think, from their perspective, it’s a success. They don’t think the world has been poisoned by their success. One of them is called Nigel Kabvina. You literally met this person when they had zero followers. Tell me about Nigel and where he is now.

Nigel was this fascinating guy. I first came across his TikTok videos. He didn’t have very many followers at all. He was posting vibey videos of him swimming in his condo pool set to Frank Ocean, and I was just like, “These are very cool.” It was 2020, people were going insane online. I thought it was interesting. I wrote this essay about vibes and TikTok that was inspired by him. I was like, “Cool, this guy is making interesting videos.” 

And then I just watched over the next three years as he got more and more and more followers. First it was 100,000, then it was a million, then it was 2 million, then it was 3 million, and he had somehow cracked this code, I think, that just made him succeed wildly on TikTok. And Nigel was a flair bartender in the UK living in Manchester. So he had this kind of facility with making drinks and cutting small garnishes and stuff. And in the pandemic, he just started making really elaborate cooking videos. 

And they would all start the same way. It would be his body, head cut off, apron on, and he would say, “Good morning, let’s begin,” and he would make an incredibly elaborate, weird, tricky brunch thing. In one video, he freezes a bowl made of ice and then pours milk and cereal into it. So it’s these visually elaborate, crazy trick cooking, essentially. This would get him more and more and more followers. He had decided not to show his face because it was too specific. He didn’t say words because that would also alienate some of the international audience on TikTok. And he ruthlessly scanned all of his engagement data to figure out exactly when people were flipping away from his videos.

So rather than seeing people not liking it or scrolling away as a flaw on their part, as bad taste on the viewer’s part, he saw it as a flaw in his own creative process, and so he would change and adapt to whatever held people’s engagement. And all of these tricks together just made him completely blow up to the point that he’s one of the bigger food creators on TikTok, I think, and it was because he thrived in this algorithmic ecosystem.

That’s the tension, right? I don’t think Nigel thinks something bad happened to him. I assume he thinks he’s the winner.

Yeah. Nigel sees this as a success. He has cracked the code, he has gotten millions of followers, he’s now making tons of money from sponsorship deals, and he doesn’t feel curtailed by the algorithmic feed. He likes this data. He likes optimizing for distribution. And so, for him, he succeeded. It works for him. 

What I worry about and what comes up when I talk to other creators is that they feel more anxiety about that process. They don’t want the pressure to optimize their content. The constant feedback from an invisible audience makes them feel like they can’t just be creative and put out what they want to, that they have to follow the vicissitudes of the algorithm, they have to pay too much attention to the feed. So I think that is the tension. Do you embrace the distribution and figure out what works and just go for it? Or do you want to maintain some sense of, “Here’s the direction I want to go” that’s not about what the audience wants?

Isn’t that an old tension? I feel like artists in the ’70s had to make big decisions about whether they wanted FM radio airplay, and when all of the radio stations were owned by big companies or had powerful local DJs, you had to go please the powerful local DJ. There were tastemakers, there was commercial pressure, there were gatekeepers in a way that maybe there aren’t gatekeepers now. They were just people, and those people were perhaps more capricious, less offering of feedback. There’s a real balance there, right?

I mean, I never want to say that the algorithmic feed is a fundamentally horrible technology. I think it helps us a lot, and I think the internet has been really good for getting rid of a lot of gatekeeping. I grew up on that internet where I could publish anything. I can post on Instagram, and that’s how I’ve made my career, too. But I think we have to figure out what that balance is between human gatekeepers and human curators and this completely algorithmic, mathematically defined distribution mechanism that we have. There are advantages to both. The human gatekeeper, on one hand, can highlight a new voice, can bring up something that’s totally unusual, out of left field, and just say, “More people need to see or hear this.”

Whereas, the algorithmic feed is never going to take a thing… You have to get engagement to succeed in the algorithmic feed. You need to please 100 people, then 1,000 people, then 100,000 people, and that iterative process doesn’t work for all forms of culture. So I want to argue for this balance that we need to not just consume things that are algorithmically recommended, and we need to maybe bring back a little bit of the human curation and tastemaking that we’ve lost in this total dominance of algorithmic feeds.

There are two ideas in the book that I think connect deeply to this. One is the concept of algorithmic anxiety. And I think you’ve summed it up in various articles you’ve written as, “The internet isn’t fun anymore.” You open the internet, and you’re like, “What is being shown to me? Who is on top of this? Is someone responsible for this? Or is this algorithm just trying to reprogram my brain?” And then there’s the concept of tastemakers, that there are some institutions and some people who are exercising judgment in some way that you trust or you don’t trust, or at least you can have a human disagreement with and they might surprise you, and those are being reduced. 

So you have anxiety over what you’re being shown, the creators have anxiety over what they’re making, and on the other end of it, you have the institutions that would be opposed to that, that are very human, collapsing. Pitchfork just got folded into GQ, I don’t know what’s going on on Tumblr, you can just see the list. In general, institutions are collapsing. Do you think that’s a natural state of Filterworld?

Oh, man, that’s a tough one. I do think the deconstruction of institutions is a consequence of us giving over a lot of power and authority to algorithmic distribution. Even super early on when Google Search and ad targeting took away revenue from media companies, that was a movement away from the human gatekeepers and tastemakers who were magazine editors to a kind of technological ecosystem. So I do think the way that we’ve given over so much attention to the algorithmic feeds and the way that we’ve let them make decisions for us, that totally destabilizes the institutions that we once relied on. 

But that’s not to say they don’t exist. You can still listen to a great indie radio station, you can still go to an art gallery, you can pick up a book that’s recommended by a human bookstore employee in a bookstore, but I do think the whole cultural ecosystem has been reshaped around feeds and around what succeeds on feeds, and that’s the deadening consequence that I wanted to highlight.

You said earlier people get bored and then things change. You’re saying there’s anxiety, the internet’s not fun for a lot of folks. What do you see as the next turn there? Because I see a next turn coming, I’m not sure what shape it takes, but I’ll give you an example. We’ll put it in the show notes. There’s a post on Threads I saw the other day, and it was someone saying, “I love podcasts, but I don’t have time to listen to them, so I find the podcast on YouTube, I feed the YouTube video into AI, and I have the AI make me an article summarizing the podcast.” And this is true, and at the bottom of it, then there was a pitch for the person’s newsletter, which, it’s like a perfect summation of media.

The whole thing is like, “I’m using AI to steal podcast content so I can make articles, and then I want you to buy my newsletter so you can…” And it was like, “Yep, that’s a PhD thesis in media.” But I’m looking at it from another perspective, and I’m like, “Oh, this person just wants articles.” If you could just deliver an article to this person from the beginning, it might actually be better than this convoluted process where you have to light up GPUs in a data center to tell you what someone said on a podcast. That’s crazy to me.

Articles are still out there. You can read TheVerge.com, and it’s a great summary of many stories going on in tech. People have gotten confused, I think. I think there is this generational shift and consumer shift away from texts, unfortunately. Reading might be over, so a lot of audio and video consumption is going on like this podcast we’re doing right now, and people have—

Just to be clear, we will publish this as a transcript that you can read. It’s a thing we’re very committed to here at Decoder.

… Some people have lost track of articles. And media companies, I think, started to be not very good at delivering texts that people actually wanted to read. There are many examples. But newsletters, I think, are an interesting innovation in that space where they are delivering text in a format that works for people in a subject and editorial approach that works for people, and those don’t look like a New York Times article. They’re delivering it in a new form.

Ultimately, Filterworld is about the distribution shaping the content — and shaping the world, shaping physical spaces around us. If you think of the interior of a coffee shop as content, which is a very fun way to think of the interior of a coffee shop.

Which I think it is. I think everything is content now, unfortunately.

Newsletters are open distribution so that you see this diversity of form in newsletter. There’s no way to succeed at Substack. There’s a way to be a hustle bro on Twitter to try to get people to subscribe to your Substack, but there’s no one saying, “Here’s how to make a Substack newsletter or any newsletter.” Podcasts, I think, are the same way: relatively open distribution, you see a great variety of forms. You mentioned text on the internet. SEO has really shoved all texts on the internet into the same box. It all looks the same because Google dominates that distribution. And then we’ve talked a lot about TikTok. TikTok’s distribution forces everyone to be more the same than different. Do you see any other open distribution that is less Filterworld and more diverse?

Yeah, we’re saying email for newsletters, the open distribution for podcasts—

And just to be clear, there are algorithmic incentives for both of those things, but in general, I think there’s more diversity in email and in podcasts than in TikTok and YouTube and even Spotify, right?

Totally.

You can see the algorithmic platforms push everything toward the sameness, whereas even slightly more open platforms or open distribution methods allow for slightly more diversity.

I want to be optimistic here, but it’s kind of hard. Email is a huge one, and through email, you can distribute whatever content you want. That could be videos, it could be audio, it could be text. I don’t know. What comes to mind is honestly walking around. If you are in a city or if you’re walking down a block of some downtown somewhere, there’s no distribution mechanism there. You can walk down the street and choose what place you’re going into. But on the internet, oh, it’s really tough. I almost think we should bring back those Japanese text message novels. The first person to go from the newsletter to the SMS newsletter or the newsletter that I consume in my text messages—

Oh, it’s coming.

… I feel like that’ll be big.

The example you gave of walking down the street is interesting because one of the things that occurs in your book is you go to a lot of coffee shops, and I’m just going to call this out: in the book, you are constantly going to Filterworld coffee shops. You mention it, it’s like a recurring theme in the book. You’re like, “I was in Tokyo, and I went to a coffee shop that looked like all the other coffee shops.” Why do you choose it? As I was reading the book, I was like, “Is this a running joke that I’m supposed to get, or does Kyle love AirSpace?”

Oh, man. I do think I’m always interrogating my own tastes a little bit. I try to make fun of myself. I do gravitate back to these places over and over again, and in part, it’s to investigate it almost, or that’s my excuse. I need to see how it works and make sure it’s still there in a way. But it’s also the taste that I’ve developed over my 20s over the 2010s, and it’s my generational preference. And it’s a joke. I think this could be a book about coffee shops, and it is a book about coffee shops, and I think that’s really funny. I mean, when you’re writing a book, you have to entertain yourself. And if you go to the very end of the book, to the acknowledgments, I also acknowledge the generic coffee shop near my home in DC where I wrote the book, so it all goes full circle. I’m still trapped in this ecosystem.

Alright. Well, I’m going to have to push you to take this joke a little bit more seriously than maybe it is. How do you feel when you’re in those coffee shops?

I still feel comfortable. I’m like, “This is a space that reinforces my tastes.” In the same way that you go to a bar or a restaurant that reinforces your own tastes, it’s a space that reaffirms your identity in a way. It reminds you of who you are, what you like, the kind of environment you enjoy being in. So, in that way, I find them comfortable. I also feel alienated. I think when you become too sensitive to it or too accustomed to that particular shade of Filterworld or of AirSpace, it’s like the slightest quirks throw you off. So if the decor isn’t quite right or if the table isn’t wide enough or there’s not enough mid-century modern furniture, you’re like, “Oh, this isn’t good enough.” So what I’m trying to do now is actually lose that. I’m trying to train myself out of this total preference and appreciate places that are more different or weird.

How do you feel when you go to a coffee shop that isn’t that aesthetic?

Sometimes annoyed—

Because that’s the one thing that isn’t in the book is a long chapter where you just go to places that are totally not that and just have emotional reactions to that.

Yeah, that should be the sequel. That can be the non-Filterworld—

Leaving Filterworld.

… leaving Filterworld. No, I think what that makes me think of is I have a friend who lives in London, and she is really obsessed with the kinds of London institutions that came up in the post-war period — the caffs, as they call cafes, and the old pubs that are not owned by chains now — and she seeks out this specificity of experience that might be weird, it might be ugly, it might be covered in laminated countertop, the sandwiches might not be that good actually, but they have some kind of stable identity and real-life endurance that is the opposite of a Blank Street Coffee shop. If Blank Street is the ultimate physical spaces content, then these places are the total opposite: textured, real, they have history and reasons that everything exists in there, and I find that increasingly appealing, even as a lot of those places are steamrolled by capitalist gentrification and urbanism and everything else.

Those places are really interesting to me because I see the beginning of a pushback. The example that I will give is the gray millennial floors that are in all the real estate listings. I see people complaining about them more than I see people wanting them now. If you live in New York, everyone’s constantly looking for a new apartment, that’s just a part of the experience here, and everyone’s always like, “All these apartments have been refinished to be exactly the same. And it doesn’t even matter. I just have to pick a price and a location and we’ll be fine.”

But now I see on TikTok younger people saying, “Why did millennials turn everything into these weird gray floors and paint everything white? I want my kitchen to have oak cabinets.” Which I look at and I’m like, “Absolutely not.” But I can see there’s just a pendulum swing back. There’s just a reaction that is 180 degrees in the opposite direction. Is that, do you think, just part of the Filterworld experience, that you just see something enough and you’re like, “Give me the opposite”? Because I don’t see any gradations or evolution. I just see wild pendulum swings.

Right, one side to the other.

And the thing you’re describing with the caffs in London is like, I’m looking for the authentic thing that happened organically here, and I don’t even know what authentic means in a context where there are wild swings.

Right. I think trend cycles happen faster now because we get saturated with them faster. You see it in fashion, certainly, you see it in music and design aesthetics, so I think there is this churn of new things. The switch from coastal grandmother to mob wife or whatever on TikTok, it’s like, “Wait a second. That was two months ago. What happened?” So that has sped up. And the next generation or the next wave of people does crave the opposite thing. The question is, “How do you separate the opposite of millennial minimalism and quote, unquote authenticity?” The seriously authentic thing, the 70-year-old London caff, is not subject to trends. It’s not tailoring its style to anything. It is the way it is because of history, and yet people float in and out of it.

There is this caff in London that my friend really likes, which I won’t name because it’s now gone viral on TikTok, and so she complains that it’s been flooded with new customers who are not there necessarily because they understand the deep quality of it. They just saw it on TikTok and they want to repeat that experience. So the authenticity itself has become a fetishized style, has become a kind of content. And will that pass, too, in six months? I don’t know. That’s the knot that I can’t solve.

I think the question is, 70 years from now, does a coffee shop with subway tile and Edison bulbs and reclaimed wood furniture pass as authentic?

It will be historic. It’ll be like as if you’re in a pub from the ’50s or whatever, and you’re like, “Wow, everything was once covered in wood. That’s weird.” Except it’ll be like, “Wow, everything was white subway tile and small succulents and ceramics. That reminds me of my grandparents.”

There’s a real globalization factor here, right?

Mm-hmm.

Authentic London pub from seven years ago is rooted in a place, and I tend to think almost all great works of art have a sense of place to them. There’s something really real about something coming from a place in a moment, but that’s because I grew up in that aesthetic world. I grew up in the ’80s and ’90s, and there was no choice but for things to be from a place. Literally, East Coast and West Coast rap was a thing that was defined by places and tones of places. I don’t know if there exists an East Coast and a West Coast rap in the same way. I don’t know where Dua Lipa is from. I think her music is great, but there’s not a deep sense of place related to it. AirSpace, Filterworld is all about removing that sense of place, making everything the same everywhere. Seventy years from now, do you think we come back to a sense of place? That’s a big question.

Oh my God, yeah. Well, first, I think platforms are the new places in a way. Instead of East Coast / West Coast rap, we have SoundCloud versus TikTok rap, and those are the opposing styles, and you can claim allegiance to one or the other. And say, like pop music has become so globalized. So, since we’re accelerating down this path of placelessness and total globalization of culture, can we go back? I think I am kind of pessimistic. The book also covers tourism in Iceland, for example, but I think, last year, you had this wave of tourism going to the French Riviera and spots along the Mediterranean.

The classic trendy spot of the French Riviera.

Right. And it’s a kind of timeless, historical, popular destination, very specific, hard to get around, and yet it was like cannons with this wave of 22-year-old influencer tourists, and that has a way of pounding down the specificity of the place, too. It’s like specificity attracts attention, and then attention destroys specificity, and that’s a crazy thing to think about.

One of the things that’s interesting about that idea is the influencers are buffeted by algorithmic pressure, so they have to go pay attention to what everyone else is paying attention to and that loses specificity, but then some of them rise above the others, and you get very powerful individuals who can make different decisions or take brand deals or whatever needs to happen there. 

That’s very commercial. You can go give Stanley cups to a bunch of Mormon influencers and make Stanley cups a thing, which is just a fascinating reality. I don’t think that has been true in the past. Next to that is the decline of media institutions, which they’re not supposed to be collections of individuals in that way. They’re supposed to be brands unto themselves with their own kind of taste. You have the Meryl Streep monologue from The Devil Wears Prada in your book as an example of how people think about these institutions. In that case, Runway, which is a stand-in for Vogue. Vogue still exists and—

For now.

For now. Whatever’s going on with Condé Nast is going on with Condé Nast, but Vogue still exists, and celebrities still want to be on the cover because that institution still has power. In my view, the platforms do not want any institution staff, however. They would rather negotiate with an infinite supply of burned-out individuals that all do the same thing. This is the history of, I think, the 2010s media is the decline of these media institutions. 

Do you see a return to that? Someone else has to play that validating role. Someone else has to provide a celebrity something that feels like a magazine cover, something that rises to that level, and that feels like the antidote to Filterworld. The people seek this validation. People talk about Anna Wintour literally as though she’s in the Illuminati, but that’s not forever, and there needs to be something that replaces it.

Yeah, there needs to be a tastemaking force that works, and there needs to be a way that cultural ideas or people can get distribution that is not just algorithmic, though what you’re saying induces this nightmare for me of TikTok covers. TikTok releases a digital cover for its celebrity of the month and just makes them famous. That’s a scary thought.

But just to be clear, I don’t think they could.

Because it would be so artificial?

I think that would be empty.

Yeah, it would be too empty.

I think the TikTok audience would reject that top-down tastemaking from the platform itself. YouTube famously tried to do YouTube Originals, like, “We’re going to make TV shows now,” and everyone was like, “Why?” And they just disappeared, and PewDiePie went back to making PewDiePie videos. There’s something about the nature of the platforms where they actually can’t do the thing themselves. They need something else to provide that role. And I don’t know what that next thing is. I think it behooves us all to figure it out, but I don’t quite see it yet.

No, no. And no one trusts those platforms enough to give them their tastemaking judgments. So we’re in a weird swing of media institutions are totally crumbling, and I think we are seeing some rebuilding of that. You are a tastemaker. The Verge is a curatorial force that both produces original content and directs attention at specific ideas.

The last website on earth, I keep saying it.

Yes, exactly.

It’s all we have.

But I think newsletters are doing that—

I want to point out that the editor-in-chief on this podcast was like, “I don’t know where Dua Lipa is from.” I do know where she’s from. Just putting that out there as a tastemaker. Continue. Sorry.

… So the rebuilding of those tastemaking forces is happening, I think, in newsletters. You look at Blackbird Spyplane, the menswear newsletter, you look at Magasin, the women’s fashion newsletter. For some reason, it’s happening in fashion very quickly and obviously, but I think those places will build up and grow and hopefully sustain themselves, which they will have to do by hiring more writers, more people. They’ll have to decentralize from the single-person personality cult just as magazines did, just as Anna Wintour has done. And so I think we’ll see them get a little bit bigger and consolidate their power, and YouTube channels will publish articles and make podcasts and everything else, but we are in this rebuilding phase, I think.

I want to end with an exercise you did in the book called an algorithmic cleanse. You divorced yourself from Filterworld. I feel like everyone did a version of this when Elon [Musk] bought Twitter and everyone reconsidered their relationship to Twitter, but you went all the way. Explain what that cleanse was, how you actually executed it, and how you came out of it at the end.

This was toward the end of 2022, so it was as Musk was buying Twitter, and I just hit a point where I felt so saturated by algorithmic feeds, and I’d spent the whole process of writing this book thinking about them, I had to escape. I had to just run from this whole ecosystem, and so I paused all my accounts. I logged out of everything on my computer and on my phone. I deleted Spotify, I deleted Instagram and Twitter and everything else, and I just went cold turkey for about three months. So I was no longer getting any feeds of information, I wasn’t getting recommendations of anything, and I had to figure out new ways of seeking out content.

I had to look at the newspaper, I had to go to a library, I had to point my browser to TheVerge.com and see what was on the homepage. And really what I found was that the internet is no longer built for not being on feeds. Particularly, two years ago even, websites were not thinking so much about their homepages. Newsletters were less of a thing. I feel like we’ve come to rely so much on distribution and broadcast that we, media creators, don’t think enough about just having a place where people go to find things they’re interested in.

There’s a real tail wagging the dog element of this where you can want to have a different media diet. I have set up RSS readers many times in the past two years. I used to read all of my news in RSS. I used to sit in school, my laptop open, and not pay attention, and go through my RSS reader, and I remember saying to some of my friends, “I finished the internet today,” because I’d read everything in the RSS reader, and there was a great diversity in content. No one thinks that way anymore. You open an RSS reader, you plug your favorite websites into it — candidly, even ours — and you get a bunch of stuff, and some of that stuff is obviously made for SEO, and some of that stuff is obviously made for other platforms.

And very rarely do you see, “Oh, there’s an audience here that wants to read every article on this website, and that is a package,” but it’s coming back. People want to do that, right? You can see… You have felt that way. I have felt that way. We write articles about RSS readers, and people read them. There’s demand for it. Do you think that demand is ever going to get filled?

I hope so. I tend to think, “Wasn’t the great promise of Silicon Valley and all these tech startups [that] we are going to give users things that they want?” There’s this thirst for a new form of delivery of content, better curation, more holistic ideas of what we should consume, and I hope that products arise to give us that. I think people are restlessly questing for it right now in RSS, in newsletters, in a parasocial podcast video, whatever ecosystem. But I don’t know. I like internet technology, I like when startups do new stuff. I hope that they take on this challenge and figure it out.

Isn’t the flip side of this that mainstream culture is dominated by filters and algorithms, but niche culture has thrived as well? Because you can be a K-pop fan in the middle of the country and have a community around you that is into whatever extremely esoteric K-pop lore that you need to be into. Or Taylor Swift is another example. I am not sure why numerology is having a moment in the character of Taylor Swift, but numerology is having…

And that just developed out of nothing, organically developed out of millions of people just sharing an experience together and having a good time. Fine. That is not a thing I could do growing up in the Midwest as a teen. I could not find all the other Clash fans in the world. Probably for the best. I had to go interact with other people who did not care about whether The Clash was going to reform. I wish I could, but I couldn’t. Isn’t that the flip side of Filterworld, that you can actually go find these very narrow communities and go participate in them without the pressure of the algorithm because the internet provides those tools as well?

The opportunity for discovery is so big on the internet. You can find almost anything you want to find. You can find the community of K-pop stans of the one tiny band that you’re super into that no one else knows. You can find how to fix your vintage car or build furniture or whatever, and I think that’s really powerful and great. And it still exists within algorithmic feeds. I find niche DIY content on TikTok all the time or vlogs from American expats in Copenhagen praising their healthcare. I’m like, “Wow, I never knew I wanted this content, but I really do.” But it’s hard to build more stable communities around that stuff now. I think there’s this tendency to dip into a form of content and then just let it flow by you and then drift your way to the next thing. There’s a very superficial experience of a feed that is very dominant, and some people will always go deep.

The real fans will always dive down and find the forum or find the Discord or start a literal pen pal thing or something, but I feel like I wish the ecosystem we were in encouraged that more and had more resting places and smaller communities along the lines of a Reddit or a Discord where we could all find our places and develop good relationships in our niches. I think you want the niche to be sustainable. I think the niches you’ve pointed out — K-pop or Taylor Swift — those are tons of people collected into the same big buckets of stuff. Taylor Swift is one huge bucket, K-pop is one huge bucket. I hope there are ways of collecting groups of fans around one new kind of music or a record label, like a Bandcamp in a way, where you are actually giving money directly to artists and sustaining them because you’re fans of them. I guess niches being sustainable is a big concern of mine at this point.

The idea of Filterworld is that algorithmic pressures of the internet bleed out into culture and into the real world in quite literal ways. The opposite view of the internet that we’re touching on here is that the internet has enabled fandoms at scale in a way that dominate culture. We had Kaitlyn Tiffany on the show a while back, she wrote an entire book about how, in particular, the One Direction fandom is responsible for all of internet culture, which is quite a thesis, and I love Kaitlyn for putting it forth. 

But that’s the idea, is that the internet created the opportunities for these fandoms, these large groups of coordinated people to be like, “Nicki Minaj is actually the center of culture today.” And maybe there will be a threat of violence behind that idea given the Barbz’s history, but that is in opposition to, “The algorithms make you all the same” because you have these large groups of people that can override the algorithmic preference and shape culture in whatever way that they choose. Do you feel that tension, or is it actually the same thing?

Well, I wonder, these fandoms are almost military groups that everyone is drafted into. It’s like you might be floating around in the hip-hop community or whatever, and then gradually you get sucked into the Nicki Minaj fandom. Or you might be a young woman who is interested in acoustic guitar or something and then you are orbiting Taylor Swift and then you get sucked into that bucket. One of the forces of fandom that’s not just about consumption is the need to project your fandom at other people where it becomes this homogenizing force, I think, where it’s like, “No, everyone else has to like this thing that I like.”

And I think that’s a Filterworld-era force of, “Not only am I going to be a fan of this thing, not only am I on the fandom, but I’m going to oppose anyone who’s not, and I need to convert, I need to be the zealot and the missionary to go out and convert the heathens and convert them into my group.” I feel like fandom energizes consumption and going deep into an artist or a body of work, but it also becomes this culture as conflict that I don’t think is super healthy.

I think that is a very succinct description of fandom on the internet. When you were doing your algorithmic cleanse, you divorced yourself from that as well, too, by necessity. You can walk away from the algorithms, and you walk away from these apps, but then you walk away from the other animating force and culture, which is fandom. Did you perceive that as well when you were just offline like, “I’m reading the newspaper, and I am confident that some representative of the Star Wars fandom is very mad about this movie review”? There’s something to that where a lot of the culture is in response to these things — even if it’s not explicit in the text, it’s always a subtext. Did you perceive that?

I did. I felt a kind of absence in a way when I was much more offline, when I wasn’t within these feeds. And I knew that what’s presented in a newspaper or shown in a gallery or whatever is subject to the forces of digital platforms, but I was almost missing the community feeling of those feeds and of a fandom where it’s like, I know that other people are consuming the same thing I am, and we’re all chattering about the same stuff, and we’re all leaving comments on the Instagram account, and we’re all clustering in our way around one piece of content online. That’s made very literal on the internet. You can feel the presence of other people in a way you can’t when you’re just reading a newspaper or reading a book.

You might abstractly know that other people are reading this book or reading a magazine article, but you can’t feel them next to you in a way. You can’t feel the gang quality or the fandom around a certain thing. And I miss that because it’s something that I’ve totally grown up with and that I get a lot of energy and support from. When you post something that other people like, when you come up with an idea that other people grab onto, it’s super encouraging and energizing, and so it can be harder to operate without that.

Well, Kyle, I could talk to you all day. As I think everybody can tell, I love having this conversation with you. Sadly, we are out of time. You’ve given us way more than we bargained for, I think. Tell people how to be more thoughtful about living in Filterworld. You’ve got a lot of ideas in the book. We didn’t even get to the “Kyle has thoughts about Section 230” part of your book, which is a whole other hour, but the core thing is these algorithms are designed to make you feel things, and you can be thoughtful about that interaction. Tell people how they should process living in Filterworld.

Being more thoughtful is a good start. I think what I came out of it with was you want to know that you like what you like because you like it, not just because it was recommended to you and exposed to you repeatedly in a feed. Thinking about your personal taste, having a real encounter with a song or a piece of art or a piece of clothing where you don’t think about how many people liked it, where you don’t think about the Instagram account, where you just sit with your own feelings and have an experience of culture that’s in front of you that’s changing your mind or your soul or whatever, that’s truly what I want people to have. I want you to sit and stare at a painting and be like, “What does this make me feel? I don’t care how many likes it has. I don’t care how many followers the artist has. How am I feeling right now?”

Yeah, that’s great. Well, that’s a great place to end it. I hope people feel great about this podcast episode. Stop when you’re done listening to this and just think about it for a while. Kyle, thank you so much for being on Decoder. This was great.

This was really fun. Thank you.