I have been thinking lately about the refrigerator. Not about any particular refrigerator — though the proliferation of internet-connected refrigerators offers plenty to think about — but about the refrigerator as a category of object, and what it represents in the history of the domestic cyborg.

A kitchen counter with an array of small appliances, each with its own cord and indicator light
The modern kitchen is dense with interfaces. Most of them were designed in isolation from each other, and from the humans who use them.

The modern kitchen is one of the most technology-dense environments that most people encounter in daily life. A typical kitchen contains a dozen or more networked objects — not networked with each other, necessarily, but networked with the humans who use them through the feedback loops of heat, sound, light, and smell that tell us what is happening and what we should do. The timer beeps. The oil begins to smoke. The bread darkens. The coffee finishes with a gurgle. These are signals, and the skilled cook has learned to read them.

What strikes me is how much of this network is managed peripherally. You do not stand at the stove staring at the burner, monitoring it directly. You move around the kitchen, doing other things, while keeping the stove at the edge of your attention — available when it needs you, absent when it does not. This is calm technology working exactly as it should. The kitchen, at its best, is a masterclass in ambient awareness.

The glowing element of an electric stove seen from the side, the coil shape visible in the dark kitchen
The glow of a burner is information. It does not demand your full attention — it is simply there, available at the periphery, legible when you need it.

Then consider the smart kitchen appliance — the refrigerator with a touchscreen, the oven that connects to your phone, the coffee maker that requires an app. These devices take what was peripheral and make it central. They demand your full attention in order to deliver what the old appliances delivered at the edge of awareness. They have not improved on the old feedback loops; they have replaced them with worse ones, dressed in the appearance of sophistication.

This is not an argument for the past. It is an argument for understanding what the past was doing well, and carrying that knowledge forward into new designs. The human organism has been developing domestic intelligence for hundreds of thousands of years. The least we can do is design technology that respects what we already know.

← PreviousDesigning Calm AI Products: Real-World Engineering Principles to Create Safer LLMsNext →The Invisible Light That's Harming Our Health — And How We Can Light Things Better