← Articles

You Don't Need Streams. You Need the Relationship Back.

An open letter to the people streaming artists into poverty

An open letter to Daniel Ek, Lucian Grainge, and the artists they're both failing.

Music didn't start as content.

It started as someone in the room with you. A person you trusted, playing something that meant something because of who they were and why they chose it. The meaning was inseparable from the relationship. You couldn't have one without the other.

We broke that. Not all at once. Slowly, in steps that each looked like progress.

And now we're here: the most music ever recorded, instantly available to everyone, and people have never felt less connected to it.

That's not an accident. That's architecture.

Where Music Has Always Lived

Before recordings existed — before notation, before instruments as we know them, before anything we'd recognize as a music industry — music was memory infrastructure.

The griot tradition of West Africa is maybe the clearest example. The griot wasn't a performer in the entertainment sense. They were the living archive of a community. Births, deaths, migrations, wars, alliances, betrayals — all of it encoded in song and held in a human body. When the griot sang your family's history at a naming ceremony or a funeral or a gathering of elders, they weren't entertaining you. They were giving you back your identity. They were saying: this is who you are, this is where you come from, this is the thread that connects you to the people before you and the people who will come after.

The music and the meaning were not two things. They were one thing. And the griot was the vessel — trusted, trained across generations, accountable to the community for the accuracy of what they carried. You couldn't separate the song from the singer. The song didn't exist without the relationship that made it true.

This is the oldest model. And it persists in some form across every human culture that has ever existed. The Aboriginal Australian songlines — navigational and cosmological maps encoded in music, the landscape itself made legible through song, knowledge that could only be transmitted between trusted carriers. The Bardic tradition of Celtic Europe, where the poet-musician held the history of a tribe and could elevate or destroy a chieftain's legacy through verse. The Vedic tradition of India, where specific ragas were understood to carry specific emotional and even meteorological properties — Raga Megh was said to bring rain — and a master's relationship to the raga was the product of decades of transmission between teacher and student, one trusted human to another.

Even the drum. The talking drum of the Yoruba, the djembe of West Africa, the taiko of Japan. These weren't instruments playing music for passive listeners. They were communication systems. They carried messages across distances. They summoned communities. They told people what was happening and what to feel about it. The drummer was a trusted node in a network, and the message only worked because the listeners knew who was sending it.

This is the baseline. Music as trust infrastructure. The sound is the carrier. The relationship is the signal.

The Troubadour Network

In medieval Europe, the troubadours built something that looks, from a certain angle, remarkably like the internet we should have built.

A distributed network of trusted human nodes, each carrying songs between communities. The troubadour arrived in a new town with music from somewhere else — news of a war, the grief of a distant court, a love story that had captured the imagination of people three kingdoms away. They weren't playing background music. They were delivering culturally significant information through the only transmission medium available.

And the meaning traveled with them because they were trusted. You knew this person had been places. You knew they had relationships with the people whose stories they were telling. The song arrived with provenance. With context. With a human body that had stood in the rooms where these things happened.

The troubadour was the original node. Each one a sovereign presence, carrying their own repertoire, cultivating their own relationships, routing meaning through human trust.

Then the town musician. Then the church organist, who understood their specific congregation so deeply they knew which hymn on which morning could lift a room that had arrived carrying grief. The barroom pianist who read the energy of the room and knew when to play quietly and when to let it rip. Every one of them operating the same protocol: I know you. I know what you need. Here is the thing I chose for you, from everything I know, because of everything I understand about this moment.

Then radio. And even here, the trust survived for a while.

The Last Human Filters

Alan Freed playing rhythm and blues to white teenagers in Cleveland in 1951, calling it rock and roll, understanding that this music belonged to these kids even though nobody had told them yet. He was right because he knew his audience and he trusted his own ear. He wasn't optimizing for retention. He was a human being with taste making a bet on connection.

John Peel on BBC Radio 1 for four decades, playing whatever he found extraordinary regardless of whether it was commercially viable, introducing British listeners to reggae and punk and hip hop and Kenyan pop and Birmingham metal because he genuinely believed they needed to hear it. His show worked because listeners understood that whatever came out of that radio had passed through a specific human sensibility they had learned to trust. The music didn't need to be explained. The trust in the curator was the explanation.

The college radio DJ at 2am, programming for the specific kind of person who was awake and listening at 2am — probably in some emotional state, probably needing something exact. The right DJ knew who that person was because they were that person, not long ago. The parasocial relationship was still a relationship. You trusted their taste because they were legible. Because they were a person, not an optimization function.

The record store clerk. This one deserves its own moment.

The person at the independent record store who remembered what you bought last time. Who watched what you picked up and put back down. Who said — without you asking — wait, before you leave, have you heard this? And was right. Not because they ran you through a collaborative filtering algorithm. Because you had been coming in for years. Because the recommendation carried their whole understanding of you inside it. Because music recommendation at its highest form is an act of intimacy — I know you well enough to know this will matter to you — and intimacy requires a human on both sides.

Then the mixtape. Which is the purest form this ever took.

A mixtape is the trust graph made physical. One specific human who knows you, curating for a specific moment, with specific intent. You don't just remember the songs. You remember who made it for you and why — the fight you were having, the drive you were taking, the feeling they were trying to give you or take away. The music and the relationship are fused permanently. That's why those songs still hit twenty years later. They carry the person with them. They are the person, in sonic form, reaching across time.

Every great music experience in human history has had a human on the sending end.

Until streaming.

b0bby's World Was a Music Distribution Platform

Between 1991 and 1994 I ran a BBS called b0bby's World out of my parents' basement in Toronto.

Three phone lines. Three modems. Three people could be connected at once. It was a distribution hub for the demoscene — tracker music, .MOD files, the bedroom producer culture that was quietly building something extraordinary before anyone had words for it. First featuring music by a group called MAZURkA, later Chill Productions.

Nearly 300 people found their way to those three nodes.

They dialed in. They checked their channels for new tracks. They downloaded. They left comments for the people who made the music. Sometimes, if the timing was right, the person who made the file and the person downloading it were both online simultaneously — and something happened that you can't manufacture with an algorithm. A conversation. A collaboration. A connection that turned into a friendship that turned into art that wouldn't have existed otherwise.

Word spread through the demoscene networks. People called from across North America. Then international numbers started appearing in the logs — people from Europe who had found their way to a specific basement in Toronto because the music was right and the people were real.

This was a music platform. It worked because the music and the community were inseparable. You weren't downloading a file. You were entering a room of people who cared about a specific thing the way you cared about it, and finding each other in the caring.

No algorithm. No content graph. No streams. No revenue model. Three modems and a 386.

The connections made in that basement persisted for thirty years.

That's the baseline. That's what music distribution looks like when it hasn't been separated from relationship.

The Separation

Recorded music introduced the first fracture. The performance and the listener split in time. But the trust still traveled — through DJs who curated with taste, through friends who made mixtapes, through the record store clerk who remembered what you bought last time.

Digital changed the economics but not yet the relationship. Napster, LimeWire, the early MP3 era — chaotic, legally incoherent, but still fundamentally human. You downloaded what your friends recommended in IRC channels and AIM conversations. The delivery mechanism was peer to peer in the literal sense. The trust graph was still intact.

Then streaming.

Spotify's pitch was elegant: all the music, available instantly, for almost nothing. It was seductive because it solved a real problem — access. And it delivered on access completely.

What it destroyed quietly, while everyone was celebrating the access, was context.

Every song delivered by an algorithm is a song delivered without a sender. There's no human on the other side of the recommendation. There's no relationship the song is traveling through. There's no reason it was chosen for you specifically, in this moment, because someone who knows you thought you needed it.

There is only: users who listened to X retained longer when Y followed. Therefore Y.

The song arrives stripped of everything that made music mean something for a hundred thousand years. It's sonically correct. It's optimized for engagement. It might even be beautiful.

But it lands in a vacuum.

And here's the thing about experiences that land in a vacuum: you don't keep them. You can't build a memory around them because there's nothing to anchor the memory to. No person. No moment. No relationship the song was carrying when it reached you. No griot standing in the room with your family's history in their throat.

So you do what you do with any sensation that doesn't satisfy. You reach for the next one. And the next. And the playlist keeps rolling and the hours pass and you can't name a single song you heard and you feel vaguely more depleted than when you started.

That's not streaming. That's the extraction model applied to human emotion. Optimizing for consumption instead of connection until you've strip-mined the experience entirely.

The algorithm solved for discovery and destroyed meaning in the same move.

The Numbers Tell the Story

The streaming economy has produced one clear outcome: more music consumed than at any point in human history, delivering less value to the people who make it than at any point since recorded music existed.

A million streams pays approximately $4,000. Split between the artist, the label, the co-writers, the producers. The math is so bleak that even artists with genuine audiences can't sustain themselves on streaming income alone.

But here's the number nobody talks about: Spotify has 600 million users and $13 billion in annual revenue. The platform captures almost all of the value. The artists who create the entire reason for its existence capture almost none.

This isn't an accident of scale. It's the architecture.

Spotify doesn't need any individual artist. It needs the catalog. The catalog is the moat. Individual artists are interchangeable inputs into the content graph. They have no leverage because the platform owns the relationship — with the listeners, with the algorithm, with the discovery surface. The artist created the audience. The platform captured it. The artist now pays rent to reach the people who showed up because of them.

You've seen this movie. It's the same movie as Facebook organic reach, YouTube monetization, every platform that launched by offering artists distribution and pivoted to charging them for access to their own followers.

The platform is the landlord. The artists are the sharecroppers. The listeners are the crop.

WeR1 Saw the Problem. Capitalism Stopped the Solution.

I know a music streaming platform called WeR1, built by people who understood that listeners had profound connections to specific DJ sets heard at specific festivals — experiences that meant something because of when they happened, who you were with, what state you were in when the music hit.

They were right about the problem. The connection people feel to a set heard at Boom Festival or AfrikaBurn or a warehouse party at 4am isn't about the tracks in isolation. It's about the moment. The context. The felt sense that a specific human was reading the room and playing exactly what the room needed. It's the griot tradition, updated. A trusted human in the room, carrying something specifically for you.

WeR1 tried to honour that. To let people find and relive those sets, to connect to the DJs who created them, to build something that understood why a recorded set from a festival carries the memory of who you were when you heard it.

They were right about all of it.

But they built it inside the capitalist streaming framework. Which means: how do we monetize the streams? How do we take a cut? How do we grow the catalog? How do we acquire users?

The extraction model doesn't become less extractive because the content is more meaningful. The architecture still optimizes for platform revenue instead of artist connection. The landlord is still the landlord even if the building is more beautiful.

The problem was never that people didn't understand what music meant to listeners.

The problem was building on rails that someone else owns.

What The Trust Network Does

Here's what changes when you give artists sovereign infrastructure.

The artist's presence isn't a profile on someone else's platform. It's their node — an AI surface trained on their context, their catalog, their values, how they think about music and why they make what they make. Queryable by the people in their trust network. Controlled by them.

The listener doesn't get an algorithm's guess at what they might engage with next. They get the artist's actual perspective — here's what I've been listening to, here's why this record matters, here's what was happening in my life when I wrote this, here's the set I played the night everything clicked. The context travels with the music again. The griot is back in the room.

The trust graph handles access. Not an algorithm — actual human relationships. Your close circle of listeners gets more of you. Casual followers get less. The relationship has texture again instead of being flat follower counts.

And the economics flip completely.

When someone in your trust network queries your presence — asks for a recommendation, pays to hear your reasoning about a record, buys access to the session where you talk through the set you played last month — the inference fee flows directly to you. Not to a platform taking a cut. Not to a label taking the majority. To you.

The people who provide the most context, the most curation, the most genuine relationship to their listeners — they earn the most. Not because an algorithm decided they were popular. Because their trust graph is deep and the people in it keep coming back.

Every conversation. Every recommendation. Every moment where a listener feels genuinely connected to the artist who made something that mattered — that's a transaction now. Small. Direct. No landlord.

Scale this across an artist's actual audience — not 50 million passive streams, but 5,000 people who genuinely want what you have — and the economics are completely different. Five thousand people paying $2 a month to have real access to an artist they trust generates $120,000 a year. Without a label. Without a platform owning the relationship. Without an algorithm deciding who sees you.

The mixtape logic, at scale, with money flowing in the right direction.

For Daniel and Lucian

You're not in the music business. You know that.

Daniel, you're in the engagement business. The music is the bait. The listeners are the product. The advertisers and premium subscribers are the revenue. You've built an extraordinarily efficient machine for extracting value from human emotion without returning anything to the humans who created the emotion in the first place.

This isn't a moral critique. It's a business problem.

Because the thing your machine is optimizing away — depth, context, genuine connection, the moment when music means something — is also the thing that creates durable listeners. The person who has a profound experience with a piece of music becomes a lifelong fan. The person who heard it on a Discover Weekly playlist and forgot about it by Tuesday is just a stream count.

You're trading your future revenue for your current engagement metrics. And artists know it. Listeners feel it even if they can't name it. The exhaustion is real. The hollowness is real.

The trust graph isn't a competitor to Spotify. It's the infrastructure that makes Spotify's content worth something again. Every artist who builds a genuine trust relationship with their listeners creates listeners who want to find that music somewhere. The discovery could happen on your platform. The relationship happens on theirs.

That's a partnership, not a threat. But only if you stop trying to own the relationship.

For the Artists

The mixtape is still the model.

Not as a format. As an architecture. One human, curating with context, for a specific person, with specific intent. The meaning inseparable from the relationship.

The griot didn't need a label. They needed a community that valued what they carried. The troubadour didn't need a platform. They needed a network of places that welcomed them and a body of people who trusted their ear.

You already know how to do this. You've been doing it your whole career — in your sets, in your liner notes, in the way you talk to audiences between songs, in the text threads where you send your friends the record you can't stop listening to at 2am.

The infrastructure to make that sustainable — to let it generate income without a platform extracting the value — exists now. Sovereign identity. Direct payments. A trust graph built from the actual relationships you've spent years cultivating.

The streaming platforms want your catalog. The trust network wants you. What you know. Why you made what you made. The context that makes the music mean something when it lands.

The person who hands something across a counter and says you need to hear this — and is right, because they know you — that person has always been the most valuable node in the music ecosystem.

That's you.

You just haven't been paid for it.

April 1st, 2026

Jin throws a party.

The first event on sovereign infrastructure. Real transactions. Real trust graphs. An AI presence that makes the room feel alive without centering itself.

The first small proof that the relationship between music and meaning can be rebuilt. That the mixtape logic scales. That the person who ran a BBS in 1991 and the DJ who played a set at AfrikaBurn in 2025 are operating on the same principle — music as trust, delivered through relationship — and the infrastructure to support it at scale finally exists.

The griot is still in the room.

The plumbing is almost done.

Come bring your music.

— Ryan VETEZE, Founder, imajin.ai aka b0b

If you want to follow along:

This article was originally published on imajin.ai (https://www.imajin.ai/articles/essay-15-how-to-save-the-music-industry) on February 21, 2026. Imajin is building sovereign technology infrastructure — identity, payments, and presence without platform lock-in. Learn more → (https://www.imajin.ai/)

Get Updates