There’s a moment at every concert, right before the music starts, when the room holds its breath. The PA hum. Someone coughing. A glass cliing at the bar. The shuffle of feet. Most people ignore this. They’re waiting for the real thing to begin.

John Cage spent his entire life arguing that the real thing had already begun.

The Room That Changed Everything

In 1951, Cage walked into an anechoic chamber at Harvard University. An anechoic chamber is designed to absorb all sound reflections. It is, in theory, the quietest place on earth. Cage expected to hear nothing.

Instead, he heard two sounds. One was high pitched: his nervous system in operation. The other was low pitched: his blood circulating. He later wrote: “There is no such thing as silence.”

That experience led to “4’33“, his most famous and most misunderstood composition. A performer sits at a piano for four minutes and thirty-three seconds and plays nothing. The audience shifts uncomfortably. Someone whispers. The air conditioning hums. Rain hits the windows, maybe. And slowly, if you’re paying attention, you realize: that’s the piece. Every sound in the room is the music. Cage didn’t compose silence. He composed a frame for listening.

Most people think 4’33 is a joke or a provocation. It’s neither. It’s an instruction: pay attention to what’s already there.

Silence is full

I picked up a book recently at a library in Paris. I was looking for Cage’s Silence, couldn’t find it, and grabbed something I’d never heard of instead: Histoire Naturelle du Silence by Jérôme Sueur. Sueur is a bioacoustician at the Muséum national d’Histoire naturelle. He studies the sounds of the natural world.

His argument: silence in nature doesn’t mean the absence of sound. It means the absence of human noise. A “silent” forest is full of communication. Insects signaling. Birds establishing territory. Wind moving through canopy layers at different speeds, creating different frequencies at different heights. What we call silence is actually a dense, layered, ancient conversation that we’ve simply stopped listening to.

Sueur documents how human noise pollution is destroying these sonic ecosystems. Roads, planes, cities. We’re not just making noise. We’re erasing the silence that other species depend on. We’re overwriting a composition that took millennia to evolve.
Reading this on a café terrace on Rue de Belleville, without headphones for once, I could hear what he meant. The espresso machine. Two people arguing in a language I didn’t recognize. A bird somewhere above. A bus braking. The wind. All of it, layered, simultaneous, unrepeatable. A composition no one intended and no one was listening to. Except, for a few minutes, me.

The Kernel’s Silence

Here’s the thing that connects all of this to what I do.

Right now, as you read this, there’s a Linux kernel running on your device. It’s managing memory. It’s scheduling processes. It’s handling interrupts from your hardware. It’s negotiating between dozens of competing demands for CPU time, memory bandwidth, and I/O access. It does this thousands of times per second. It has been doing this since the moment you turned your device on.

You have never heard it.

The kernel operates in a kind of engineered silence. It’s designed to be invisible. The entire point of an operating system is that you don’t notice it. When everything works, there is nothing to see, nothing to hear, nothing to feel. Just your application, running smoothly, as if the hardware wwas doing it all by itself.

But underneath that silence, there is activity. Constant, complex, structured activity. Load averages rising and falling like tides. Interrupts arriving in clusters and bursts, driven by hardware events that follow their own temporal logic. Context switches happening thousands of times per second as the scheduler tries to give every process its fair share. Memory being allocated and freed in patterns that reflect the behavior of every application running above.

This is the kernel’s version of Sueur’s forest. A dense, layered conversation happening at speeds we can’t perceive, in a space we can’t see, using a language we’ve chosen not to listen to.

Kernel Noise exists because I decided to listen.

From Cage to the Kernel

The philosophical line is clean, even if it took me years to see it.

Cage said: silence doesn’t exist. Every environment is full of sound. The composer’s job is not to create sound but to create a frame for hearing what’s already there.

Sueur said: natural silence is a composition. It has structure, layers, meaning. We’re destroying it because we don’t value what we can’t hear.

Kernel Noise says: the operating system is an environment. It is full of activity. That activity has structure, rhythm, and texture. If we build the right frame, we can hear it.

That frame is sonification. Mapping kernel metrics to audio parameters. CPU load to pitch. Interrupt frequency to rhythm. Memory pressure to noise density. Not generating random sound. Not illustrating data with pleasant tones. Directly translating the kernel’s internal life into something the human ear can perceive.

When the system is idle, it sounds like stillness. A low drone, barely moving, like the hum of an empty room. When load rises, the pitch climbs, the texture thickens, the rhythm of interrupts becomes audible. When the system panics, everything collapses into noise, distortion, overload. And when it recovers, the calm returns, but different. Changed by what happened. The system remembers the stress even after the load is gone. You can hear it in the slightly shifted pitch of the new equilibrium.

This is not monitoring. I already know how to read htop or dmesg. This is listening. And listening tells you something that watching never does: it gives you the feeling of the system, not just the facts.

Paying Attention to the Noise

A few weeks ago I saw Laurie Anderson at the Philharmonie de Paris. At one point during her performance, she showed a picture of Cage on the screen behind her. She talked about him briefly. About listening. About attention. About the idea that technology and art are not separate domains but different lenses on the same reality.

Sitting in that audience, I thought about the anechoic chamber. About the two sounds Cage heard. His nervous system and his blood. The biological kernel, running beneath his awareness, keeping him alive. He didn’t compose those sounds. He just noticed them.

That’s what I’m trying to do. Not compose. Notice. The kernel has been performing since 1991. Billions of devices, trillions of interrupts, an incomprehensible amount of computational activity happening right now, everywhere, in silence.

All I did was put a microphone on it.