Her name is Pixel. She is a cat, she lives in my apartment, and she is — through no fault of her own — the single most important constraint in my entire home-automation design. This is a post about why, and about the small philosophical shift that made my smart home actually pleasant to live in.
The dream sold to you
Every smart-home tutorial sells the same fantasy. You walk into a room, the lights gently fade up. Everyone leaves, the house powers down on its own. Motion sensors and presence detection do the thinking so you never touch a switch again. It's seductive. I built exactly that.
And then I watched a four-kilogram animal demolish all of it.
What actually happens with a cat
A roaming pet is, to a home-automation system, a small chaos generator with paws. Consider:
- "Turn on the lights when there's motion." Now the lights come on at 3 a.m. because Pixel decided the hallway needed patrolling.
- "Turn everything off when everyone leaves." Presence detection sees movement in the apartment — it's the cat — so the house thinks someone's home and never powers down.
- "Trigger the away scene when motion stops for an hour." Good luck. She never stops for an hour.
None of this is the system's fault, and none of it is the cat's fault. It's just a fundamental mismatch: presence-based automation assumes the only things moving in your home are people who want lights. Add an animal and that assumption quietly collapses.
You can buy fancier sensors. You cannot buy a sensor that distinguishes "my human wants ambiance" from "the cat is bored."
The shift: buttons, not guesses
So I threw out the guessing. Every whole-home scene in my apartment is now a manual button on the dashboard. Not a motion trigger. Not a presence rule. A button I tap, on purpose, when I actually want the thing to happen.
There are four of them, and they cover basically every moment of the day:
- 🎬 Movie night — kills the lights, dims the living room, turns on the TV's bias lighting, and launches the media app.
- 🌙 Goodnight — everything off, media stopped, a soft nightlight left on in the bedroom.
- 🚪 Leaving — lights and media off, the whole apartment dark.
- 🏠 Welcome home — the entrance and living areas come up warm.
One tap, the whole apartment responds. The cat can patrol all she wants — nothing fires unless a human decides it should.
The part nobody tells you
Here's what surprised me: the manual version is just… better. Not a compromise — actually better.
Automatic scenes are great until they're wrong, and when they're wrong they're infuriating. Lights snapping on while you're trying to wind down. The house deciding you've left when you're sitting right there. Every false trigger is a tiny betrayal by a thing that's supposed to serve you. A button is never wrong. It does exactly what you asked, exactly when you asked, and the cognitive load of "tap the thing I want" turns out to be approximately zero.
It also draws a clean line: automation handles the boring, deterministic stuff (a light that comes on should match the time of day), and I handle intent (what mood do I want this room in right now). Machines are great at the first. They are genuinely bad at the second, and pretending otherwise just produces a house that fights you.
The takeaway
If you live with a pet — or honestly, with other people whose routines you can't perfectly model — consider building your smart home around deliberate buttons instead of clairvoyant sensors. You lose the magic-trick demo. You gain a home that never surprises you, never locks you out of your own ambiance, and never decides you've gone to bed when you're mid-conversation on the couch.
Pixel, for her part, remains completely unaware that she is load-bearing architecture. She mostly sleeps on the warm spot above the server rack. As she should.
Comments