And the option is what, buy a Zigbee dongle and a raspberry pi run some code written by unpaid enthusiasts? 3D print a case for it and mount it on the wall, running updates and fixing it ever few months when some package update breaks it?
I like the concept of lights that run from an app. I don’t have any of the physical Hue switches for my system and it’s fine. But I do not want an app that abuses me, and I do not want to maintain some fragile project made from slapped together code. I want robust open hardware with open source software.
I’m convinced that we can achieve this, but it won’t be with the current model of business and engineering we have today.
That’s the other (mostly) good thing about… they’re Wi-Fi native and don’t use any sort of hub.
Their iOS app is reasonably decent, and there are honesceeen widgets.
Easy to build fairly advanced automations… for instance I have a rule that turns my bedside fan off and turns the bedroom lights on that runs 5 minutes before my alarm goes off, weekdays only.
Most of what I use are their bulbs, which are both relatively cheap ($30 for a 4 pack), with excellent CRI, variable color temp (including a nightshirt style automation that goes whiter during the day, and both warmer and dimmer at night), and full RGB color.
I probably wouldn’t recommend them for a whole home setup (you’d need pretty serious Wi-Fi routers) but for a case like mine where I’m only using it in a few rooms, it’s great.
You don't need the app if you're using Home Assistant, py-Kasa or something like that since the devices create their own WiFi network on first boot and you can program them from there.
I run 30+ Wi-Fi bulbs on an eero mesh. These devices are noisy but they are responsive (slower than zigbee). Only disadvantage is that they live or die by DHCP. If they can't renew their lease, it can take a while for them to come back online.