In September 2024 I typed into an AI a question that had been itching at me for a while. Is there a website where you connect your music profile and it finds the people who listen to exactly what you listen to? The answer was no, not really. That itch is the product, so let me name it. I wanted to know who else is like me, and how many. There is something thrilling about that question. Am I as unique as I secretly think, or do millions of strangers share this exact love? Either answer would be a gift. Rare is flattering, legion is comforting.
Samewave is the machine that answers it. You connect your listening history, and the first screen shows a number on a world map: how many people on Earth share your top artists, and where they are. Thirty-seven humans have your exact top three. Two of them are in your city, and the rest glow softly across a heatmap of countries. That number, your twin count, is the product. Everything else (chat rooms with your twins, country-by-country comparisons, discovering what your Brazilian doppelgänger listens to that you have not found yet) hangs off that one magic moment.
The number, and the arithmetic that threatens it
Let me get the arithmetic out of the way before it flatters anyone. A top-three artist combination is nearly a fingerprint. With millions of artists in the catalog, exact triple matches will be rare, and for listeners with eclectic taste the exact-match answer will often be zero. An app whose magic moment can display “0 people are like you” has designed its own funeral, unless zero is made to feel like a crown. You are the first of your kind, so plant a beacon and the map will alert you when your twin arrives. I like that framing a lot, but it cannot carry the whole experience.
So similarity has to come in tiers, and the tiers are where the matching actually happens. There are exact twins (same top artists), close kin (two of three shared, or the same artists in a different order), and same constellation (heavy overlap across your top twenty, the people whose library rhymes with yours even if the podium differs). Each tier has its own count and its own room. The twin count is the fireworks. The kinship tiers are where the actual humans are.
The wall I hit before I even started
Now the part regular readers of this site will recognize. My September 2024 notes cheerfully specced this on the Spotify API: OAuth login, top artists endpoint, a developer brief ready to send. Two months later, in November 2024, Spotify slammed the door on this kind of project, deprecating a raft of endpoints for new apps and, more fatally, keeping new integrations in a development mode capped at a couple dozen whitelisted users, with extended access now granted about as often as sainthood. A consumer social app that requires Spotify login is, for an independent builder today, structurally unlaunchable. I wrote about the two laws of building on platforms in my EmoTune article, and I will not re-derive them here. This idea got hit by both before a single line of code existed.
This time, though, the pivot improves on the original plan.
The pivot: build it where listening is already public
The scrobbling world (Last.fm and the open ecosystem around it) is the anti-Spotify in every way that matters here. Listening histories are public by default and two decades deep. The community already has the culture this product needs, people who track and share their stats for the love of it. And nobody can revoke the ground it stands on.
One caveat, and it fills the entire first sprint. Last.fm’s API answers questions about a user you can already name. No endpoint hands you the crowd, and scraping your way to one runs straight into the terms of use. The corpus has to come from somewhere else, and it can. ListenBrainz publishes its listening data in bulk, openly licensed, so that people can build on it. Samewave starts from the ListenBrainz dumps and lets anyone connect a Last.fm or ListenBrainz profile to place themselves on the map. The map is not empty on day one, but filling it is the first real piece of work.
And there is a delicious irony to reclaim. Last.fm itself once had a feature called musical neighbours, the people whose taste sat closest to yours. It was, for many of us, the best social feature the music web ever shipped, and it was quietly buried years ago. Samewave robs that grave with pride. The neighbours come back, this time with a world map, a twin count, and rooms to actually talk in.
Homegrown maps where musicians come from. Samewave maps where their listeners are. Same open data, opposite question, and I would happily see one person build both.
The graveyard beyond that is instructive too. Tastebuds matched people by taste and died as a dating app. Spotify’s Blend compares you with friends you already have, which is the reverse of this idea, since Samewave exists to find the strangers. Stats apps show you your own numbers in a mirror, and this one turns the mirror into a window. Two neighbours come closer than anyone else. ListenBrainz already computes similar users from this same data, which is this article’s kinship tier without the map, the count, or the rooms. And Obscurify already tells you how obscure your taste is, which is the uniqueness card without the other humans. Samewave joins two features that I think were never meant to live apart.
Key Features
- The twin counter. One number, front and center: humans on Earth who share your top artists, with the map showing where. Updated live as new people join.
- Kinship tiers. Exact twins, close kin, same constellation, each with its count, its map layer, and its room.
- The uniqueness card. A shareable answer to the founding question: how rare is your taste, as a percentile and a count. It is also the growth engine. Every card shared is an invitation to be counted.
- Twin rooms. Small chats scoped by tier and artist constellation, rather than giant genre forums. What makes them work is the eerie intimacy of strangers who share your exact library.
- Country lenses. Zoom into any country and see its top artists next to yours, and how many of your kin live there. Travel planning for the ears.
- Privacy at country grain. Location is never finer than the country level, visibility on the map is opt-in, and lurking anonymously with just your counts is a fully supported way to use it.
The honest tensions
The pivot trades one problem for a smaller one. The scrobbling community is devoted, but it is a fraction of Spotify’s audience, so Samewave starts deep rather than wide. I think that is the right trade. A live map of a passionate million beats an unlaunchable app for half a billion, and manual import paths can bridge the gap for anyone willing to carry their data over. But the ceiling is real and should be said out loud.
Chat needs moderation from day one, light but present, because “strangers matched by intimacy” is a lovely product and a spam vector. And the twin counter must resist the temptation to inflate. The entire emotional payload of this product is that the number is true, so the moment the numbers feel massaged to avoid zeros, the magic dies.
Monetization
Modest, and it should stay that way. Counting and being counted stays free forever. A small supporter tier unlocks map history and extra lenses. There are no ads, because an app about belonging that sells your attention to advertisers has misread its own soul. This is closer to a beloved community instrument than a venture rocket, and it should be built by someone at peace with that.
Next Steps: How I’d Build It
The MVP is almost embarrassingly direct, and it does not even need accounts. It is a landing page where you type your three favorite artists and get the count, computed from public scrobble data, with the map lighting up behind it. That single page answers the founding question and shares itself. Accounts, tiers, and rooms come only after it has proven the itch is shared.
Then connect the open scrobble APIs properly, compute similarity across top-twenty libraries and not only podiums, and seed the first twin rooms by hand around a few artist constellations with famously devoted listeners. The hard parts are keeping the numbers honest, the rooms warm, and the map alive as it grows. None of that is a technical problem.
Closing
Samewave is an instrument for the question I could not stop asking. Am I one of a kind, or one of millions? Both answers are wonderful, and for the first time a machine could tell you which one is true, with a map, a number, and a room where your twins are already talking. It joins the other music ideas on this site, and this one takes the song as a handshake.
If you build this, tell me, and tell me my number. I have been wondering for years.
