I've been listening to articles more than reading them lately. I'm not sure why that is. But at some point the reading mode started feeling like more effort than it needed to be.
When I read, I can see the shape of an argument before I'm inside it. Scan the length, catch a phrase in the third paragraph, decide how much attention it deserves. It feels like being on the bank, watching the current. With voice I'm in the water from the first sentence. No pre-evaluation. I committed the moment it started.
That felt like the key difference, and also somewhat a loss of control.
If screens eventually disappear (this is a thought experiment, not a prediction) voice becomes the interface. That's already imaginable. But then there's smell. A device that emits scent doesn't compete for my visual attention; it bypasses evaluation entirely. That's like walking past the bakery and I'm already hungry before I've decided anything. Translating that to the imaginary device, a travel piece could carry ocean salt, and a wine review could emit that particular wine scent. Content that works on me before I've consented to engage with it.
Reading gave me control over pace and a fixed object to disagree with. Listening puts me in the current. Smell removes the pause between stimulus and response entirly.
Then I thought about conversation. An actual back-and-forth, fluid, interruptible, responsive. I can push back mid-sentence, redirect the whole thing. And the content doesn't exist in fixed form without me in it.
We're already partway there. When I ask an AI to edit a paragraph, I'm not making a direct change; I'm describing an intention and watching something regenerate around it. With images it goes further: describe what I want and it rebuilds the whole thing, often with twists I didn't ask for. The kind of control I have in a word processor, where I change exactly this word and move exactly that element, is already loosening. I'm describing direction, not making edits.
That's where the static comparison breaks down completely. Everything we know about publishing assumes a document: something with a shape, a beginning and end, a version we can share. Conversation isn't that. It's a different.
For anyone making content for a living, that's strange to sit with. You're not issuing articles anymore. You're training a voice that participates in conversations on your behalf, responding differently each time, to each person. The visual toolkit, logo, typography, photography, all of it collapses when there's no screen. But the fixed document collapses too. What remains is something more like character. A way of reasoning. A set of positions.
This post argued that clean, structured content outlasts whatever interface gets built around it. RSS proved it. Agents are proving it again. But that assumed content was still a fixed thing underneath, just more portable.
I can't answer what happens when content stops being a fixed thing.
I find it strange and genuinely interesting. I don't know where it goes from here.