I bought a ring with physical buttons. It is not smart. It is the smartest ring I bought.
The ring that cost me the most has no health monitoring. No app. No AI. No sleep tracking, no heart rate, no stress score, no subscription.
It has two buttons.
I test rings. Not as a hobby — as a research commitment. I have been tracking what physical wearables can and cannot do for long enough to know that the gap between the pitch and the object is widest exactly where the price is highest. The more features a ring claims, the more I expect to be disappointed by at least one of them.
The mouse ring was different from the listing. It does one thing: it connects to your device over BLE and turns your finger into a mouse. Glide sensor on the band. Two physical buttons. Top button opens an app. Bottom button closes it. Five-second press powers on. Five-second press powers off.
That is the complete feature set.
I paid more for it than for any other ring I own. I waited months to catch it at a price I was willing to pay. When it arrived I read the manual — short enough to read in under five minutes — and had it paired within the time it took to finish a glass of water. Found it in the BLE system tray. Clicked pair. The cursor appeared on screen.
No account. No firmware update prompt. No permission screen asking for my location. It just worked.
I wore it on a walk. Left the tablet it usually connect to at home. When I came back, the ring had no response. I don't remember powering it off, maybe it had powered itself down while out of range. Five-second press, the indicator lights came on, and control returned.
I am willing to call this a smart sleep mode. The ring noticed the connection was gone and drew the right conclusion. One autonomous decision, one legible recovery. That is the only kind of self-management a device like this earns: a predictable state, a button answer. The contract held.
The real test came after.
I had been out for an hour. My hands were sweaty — not unusually so, the ordinary result of movement and heat. When I tried to use the glide sensor, it triggered continuously. The sweat was registering as input. The cursor moved on its own. I call it the phantom trigger. It can drive you nuts.
I pressed and held for five seconds.
The ring powered off.
I wiped my hands. Five seconds again. It powered back on.
Problem solved.
I had been testing rings with capacitive touch surfaces for a while. Every one of them has this problem. Sweat, rain, incidental contact — anything that conducts registers as a gesture. There is no interrupt. There is no override. The surface decides what is input, and you are not part of that decision.
The mouse ring has a physical button. You are always part of that decision.
This is not a small thing. It is the entire thing.
A physical button is not a limitation. It is a commitment. It means the device acknowledges that you have a body, that your body is not always in the same condition, and that the interaction model needs to accommodate that. A capacitive surface assumes ideal conditions. A button works in the rain.
From less than half a day with this ring, I already mapped seven functions it could carry with its two buttons and the gestures available:
- Launch music
- Increase volume
- Decrease volume
- Next track
- Previous track
- Change playlist
- Camera shutter
That is seven functions from two buttons and a glide surface, none of which require an app, a subscription, or a permissions screen. The hardware already supports them. The only missing piece is firmware that exposes the gesture options — double press, long press, the combinations that experienced button designers know how to use.
The manufacturer did not add them. That is a shame. But I can name seven things I would do with this ring, which is more than I can say for most of the rings that came with five features built in and worked reliably for none of them.
Smart is not a feature count. It is not a sensor array or an AI integration or a companion app that requires an account to function. Smart is planning for what your user will do and stay out of the way when you cannot help.
The mouse ring doesn't know when I am sweating. It also does not try to compensate or average the input or ask me to calibrate. It gives me a button. I decide when the ring is on and when it is not. That decision costs me five seconds in either direction.
I have tested rings that cost less and promised more. I have tested rings that monitored my sleep and were wrong, rings that tracked my steps and drifted, rings that required a login to access data about my own body. Every one of them was described as smart.
The mouse ring has two buttons and a glide surface. Nothing else.
It is the smartest ring I have bought.
Field notes — Day 6.
Today it did not power on.
I pressed and held for five seconds. Nothing. I pressed again. Nothing. The ring that had a reliable answer for every input — even sweat, even range — had no answer.
I tried to charge it. I cannot tell if it is charging. The LED blinks. I do not know what the blink means. There is no app to query. There is no screen to read. There is no manual that maps the blink pattern to a state. I am looking at a light, and the light is telling me something, and I have no way to know what.
This is the part where the observation-first method runs into its limit: I cannot observe what I cannot read. The ring is a sealed system with one output channel — an LED — and no dictionary for that channel.
I have a hypothesis. I wore the ring while washing my hands. I wore it while washing dishes. Water contact, repeated, across several days. The ring may have been rated for it. I did not check before I started wearing it that way — not because I was careless, but because I assumed. Most wearables handle incidental water. I treated it the same.
If water is the cause: the lesson costs $70. The same lesson on a ring that costs $500 would cost $500.
But here is what the failure clarified, regardless of cause: a device you cannot read is a device you cannot manage. The mouse ring has two buttons and that was enough — when it worked, no state changed without my input — I pressed on, it was on; I pressed off, it was off; the ring could not be somewhere I had not sent it. Now the device is in a state I did not choose, and I have no instrument to find out what that state is. Battery level, charge progress, fault condition: the ring knows. I do not.
I have been saying for a while that battery indicators are not a luxury feature. A device that monitors your body but withholds its own status is not observing you — it is hiding from you. The mouse ring showed me the inverse of that argument: physical buttons gave me control that capacitive surfaces never did. The missing feedback gave me helplessness that any small indicator light — a simple, readable one — would have prevented.
Two states, legible to the user. That is all it would take.
The ring may recover. I will update here when I know. The observation stands either way.
Field notes — Final.
It did not recover.
The ring does not charge. It does not function. The physical buttons are there — the mechanism is intact — but there is nothing behind them. I will keep it for that. The clicking still works. The idea worked. The execution did not survive water.
This is the most disappointing device I have tested. Not because it failed — devices fail. Because it was so close to right. The button philosophy was correct. The interaction model was correct. The only thing missing was a seal.
Physical buttons without waterproofing is a design gap that does not need to exist. Handwashing is not an edge case. It is the first ten minutes of any day. A device worn on the finger that cannot survive soap and water has not finished thinking through its own use.
Of the rings I have tested: the one worth recommending is the ring with a screen. The one with physical buttons was worth recommending until it wasn't. The failure is precise, correctable, and — if the manufacturer is paying attention — the next version.
The observation stands either way. So does the $70 lesson.
I test wearables. I write about what I find.
