Creating for an Audience of One

From the years 1986 until 1997, eight different videos by Juliana Hatfield or The Juliana Hatfield Three were played on 120 Minutes a total of 38 different times. Those songs were: “Everybody Loves Me But You” (five times), “Feed Me,” “For the Birds” (three times), “I See You” (four times), “My Sister” (nine times), “Spin the Bottle,” (five times) “Universal Heartbeat” (eight times), and “What a Life” (three times).

I know this because I did some work. I did some work because I have an obsession, apparently, with the idea of the radio, in its most base construction: a place where you go to listen to music without having to actually choose that music.

March 2026: Creating for an Audience of OneListen on Spotify. Listen on Apple Music.

  • “A Comfort You Borrow” — Chalk Hands
  • “The Corner” — Common (w/ Mos Def & Scarface)
  • “Nobody Knows You (When You’re Down and Out)” — Otis Redding
  • ”99.9F°” — Suzanne Vega
  • “Water’s Edge” — Tsunami
  • “Universal Heart-Beat” — Juliana Hatfield
  • “Good Rap Music” — Bahamadia
  • “Sometimes” — my bloody valentine
  • “Waydown” — Catherine Wheel
  • ”Unsatisfied” — The Replacements
  • “Hey Jupiter” — Tori Amos
  • “Duvalier’s Dream” — Kris Kristofferson
  • “Your Flag Decal Won’t Get You Into Heaven Anymore” — John Prine
  • “9th Wonder (Blackiotlism)” — Digable Planets
  • “You Can’t Stop the Prophet” — Jeru the Damaja
  • “Helpless” — Sugar
  • “Myage” — Descendents

The paralysis of streaming choice pushed me back toward the randomness and lack of control that radio and the music-focused era of MTV once provided. After a few nights of watching uninterrupted block of MTV’s indie/underground show 120 Minutes on YouTube, I decided I wanted to make my own version of a 90s alt “radio station.”

(Station is a bit of a stretch. Really, all I did was create a very large playlist — 1,149 songs, 3.1 days of music — but the process itself was the key.)

My issue with any existing 90s playlist is that they were all made by someone else — like, for example, someone who sees Spin Doctors and thinks “Yeah, that’s alternative!” (It’s not!). I wanted my own 90s playlist to reflect what I might have been happy to see at the time and also what I have come to appreciate in hindsight. And, while this entire hand-crafted playlist task might have felt fussy and pointless, I found a lot of peace going down my list of artists, poking around at what was considered a part of the alt-rock canon at the time, and placing them within the context of that time. It was mesmerizing, and I found by the end that it was a deep and personal project that I am kind of proud of, in the way that anyone can be proud of a playlist.

I call it KCWV 91.7 because I am a huge dork. More to the point: I did all of this because I could not help myself. Because the act of digging — of cross-referencing, of finding the gaps, of arguing with myself about what counts — is the actual thing I wanted to do. The playlist was just the excuse.


A playlist is an act of creative expression.

When I create a playlist, I’m not just selecting songs. I’m feeling things out — poking at the corners of the package to see where the seams are, trying to figure out what will fit within the guides I’ve given myself. For the monthly mixtapes I put in here each week, I’m playing with expectations: balancing moods and genres and genders and topics. For a playlist like these “radio stations,” I’m having a conversation with my own taste. I’m worrying about inclusion (Gin Blossoms are corny, but I love them) and exclusion (Live are ultra-corny, and while I used to love them I now kind of hate them).

I’m working with friction. The glue to a playlist is friction — how to create a responsive flow when the list is shuffled and no one controls the transitions, and how to create purposeful tension for something designed to be listened to front-to-back. This friction helps me make decisions. It determines whether a transition is smooth, or if an artist sounds out of step, or if any of this will be interesting in the first place.

The friction is where I turn a playlist into an act of self-expression. It’s what makes the work personal, what gives it identity. And it insists that the final playlist itself is not the actual point of any of it. The act of creation is. Which is to day, no algorithms or AI playlist generators were harmed (or consulted) in any way.

A personally-curated playlist reminds me a bit of zine culture. I went to punk shows in high school, which means I was exposed to the zine at a time when it felt like the most DIY thing in the world. I’ve read a lot of zines. Hundreds? Thousands? A lot of them were: not good. But, a the same time, they felt like a creative release: an explosion of fan mail and love letters to whatever cause was top of mind. And while a lot of them were, again, not good, I read all of them, because they were — and still are — the most accurate and genuine representation of someone’s mind. They’re physically awkward, and no one is really asking for them. Yet, when you hold one, it feels real. You can feel the decisions in a zine. The layout, the handwriting, the cut-and-paste edges.

Someone is in there.


The opposite of that is also true.

The one thing that AI-generated work cannot fake is texture. You never get the idea that someone spent any time with it.

I used a little AI during this project, briefly, for two things: analyzing the raw data from the site that catalogs every video ever played on 120 Minutes (this is how I know the Juliana Hatfield numbers), and filling in gaps in my knowledge of artists I didn’t fully recognize. This is how I realized that Suzanne Vega, who I thought was just the woman who sang “Luka,” also wrote “99.9F°,” which is a wonderful little song I’d have totally forgotten about. For this kind of mechanical work — data analysis, discovery prompts — it’s a useful tool. I love it for what search could eventually become.

But I do not love it for creative work, because it is not good at creative work. To be clear, it is VERY good at APPROXIMATING what creative work looks like. And it is bad for creative workers, because it offers a shortcut, and the worst thing for the creative process is a shortcut.

The efficiencies of a tool that automates file organization or synthesizes data do not translate to the creative process — and we can see that because when AI does creative things, it’s boring. It’s out of touch. It’s Steve Buscemi holding a skateboard, “hello fellow kids”-ing the ideas we have in our heads.

AI can absolutely build me a playlist. It can fill a fictional radio station in the time it would take me to research Juliana Hatfield. But it doesn’t matter, because I was not there for it. I didn’t argue with myself at 11PM about how much Pavement makes sense, or whether Melvins land too far on the heavy side, or whether anyone will give me shit if I put the one Blues Traveler song I kind of liked in high school on there. (I did not put that song on there for that exact reason.)

The output might be identical to the untrained eye. But the untrained eye isn’t the only audience. We don’t just make things for other people. In fact, the biggest reason we make things is for ourselves — to feel the small satisfactions, to claim the experience, to put our minds at ease. To outsource that is harmful to ourselves and to the work.

This is not a screed against AI. This is a question about what we actually want from creative work.


I make these playlists, ostensibly, for you. But in reality … I make them for me. I have a very limited readership and get maybe a single comment each month. This is not an ego-boosting exercise. So when I ask myself who I’m doing this for, the answer is always myself. A playlist is for myself. A newsletter is for myself. It might be for a future me or a present me; sometimes it’s so clearly for a past me that I question if even I care anymore. But none of the external response actually needs to happen, because this process, this creation, is for me and me alone. Making something complete and coherent, even if no one is paying attention, is a way of taking yourself seriously. Of acknowledging you have something worth creating, no matter how small. With your hands. With your mind. With you for you.


The best part of this weekend’s No Kings protests was the handmade signs. A beautiful sea of signs, created by a beautiful sea of people fighting against, among other things, the exact thing most AI tech bosses would have you embrace: the increasing normality of selfishly uniformed thinking. The more we get used to a lower standard, the more we’re okay with every other standard lowering alongside it.

Every person with a sign made that sign because they thought that message was important. The real benefit of a protest is seeing yourself — recognizing that the individual feelings and anxieties and fears are not yours alone, that you are not on an island raging against a storm, but part of a movement. Protests are for visibility, but more than that they are for ourselves. They help us understand who we are among all of it.

And you can tell, because the signs are all handmade. Some are beautiful and ornate, and some are ugly and utilitarian. They’re someone’s Sharpie marks on someone’s cardboard. Someone’s actual handwriting and someone’s actual jokes. Someone’s actual anger. Someone’s actual hope.

There’s something happening right now where being visibly, messily human is the only credential that holds up. Algorithms can fake polish but they cannot fake presence. The playlists, the zine, the scribbled letter. The garden, the painting, the casserole. They’re all the same gesture as the handwritten sign.

I was here. I made this. This is what I think.


This was lovingly handwritten on March 31st, 2026