I can see that. Especially if you want to keep things practical in the current state of affairs with Homey.
Allthough I’m not really comfortable with the notion that we should have to use first-party hubs because Homey is lacking. That, to me, is just not good enough for a product that pretends to be able to connect everything (not worlds, like you say, but devices).
Personally I don’t trust Xiaomi/Aqara devices directly connected to the internet. I try to have as few internet-connected/internet dependent devices in my home as possible. It is not totally avoidable (I have a Homey after all). In my mind: more manufacturers (Hue, Plugwise, Homey, Ikea, Innr, Aqara, Netatmo,Nest in my case) means more possible angles of attack. I’d rather have one device exposed to the internet instead of many.
I would like to have this rewrite done since IAS clusters are unsupported currently. For movement and doorsensors that is a must. It includes tamper alerts etc. Does not work like on/off switch.
For all other purposes i really still use a system of systems (Hue Bridge, Plugwise Tom, still wishing to connect Qivicon). I dont want to get my house cold because someone unplugged Homey, and my wive kills me if a lamp switch doesnt work as expected. I don’t think a burglar alert should be powered by an unprotected device like Homey. It is sitting in my living room and not in a closed shelf.
I don’t think Homey is meant to be a profesional grade security alarm. It can do that too, yes. But if you seriuously rely on it, I’d buy a dedicated alarm system.
But I agree. Homey relies too much on correct/more advanced Zigbee implementations of other vendors.
Hi @rampage, Agree with you fully. For similar reason I would rather connect the burglar alert to Homey for reporting purposes only. Not even to manage it.
With respect to systems of systems, it works fine but will enforce multiple Zigbee and Zwave networks to be used. Currently I am running 3 Zigbee networks (Hue and Plugwise and Qivicon) without any problems. Zwave runs on Homey and Toon without interference as well.
They might focus more on the Zigbee rewrite they promised years ago instead of other useless options that came in the period after that. Like the whole energy thing.
One of the primary ways a smart home could save you some money is by reducing energy waste. It is an economic selling point.
Now we can debate on how useful Homey is for that, granted. But in terms of marketing, and the ever rising costs for gas and electricity, it is certainly not useless to all.
I guess if you’re an environment type of person this may be a selling point as well. Although with all the smart devices leeching standby power all the time this may not be very convincing.
Not sure if this is much of a selling point in the US. I believe Homey wants to enter the US market in the near future. Homey needs more distinction from competitors I guess.
The Zigbee rewrite can be considered a poweruser-feature. It helps people who already own Homey. And a small portion of them, like us, who have more than 20 Zigbee devices.There is no direct money to be made of it.
Unless they make the Zigbee functionality so that it can recognize any device in learn mode. So you would be able to add any cheap device to Homey, witout having to install an app. I have not been succesful at that, maybe others have. Same goes for RF and Z-wave btw.
Then the claim that Homey connects to almost any device would become true. And that would be a great advantage.
But I do agree with you and wish they would prioritize it. But that is purely for selfish reasons.
The Zigbee rewrite will not be able to fix the 20-devices-directly-connected limit (well, okay, Athom may be able to squeeze out support for perhaps a few more), that’s a hardware limitation.
The main reason people (/developers) are waiting for the rewrite is so that Homey will be able to support certain types of devices, like remotes or particular types of sensors, that it doesn’t support now because it lacks support for particular Zigbee profiles (which is a software limitation). That’s something that will benefit both existing and new users.
The necessity to have an app to support specific devices is fundamental to how Homey works, and moving away from that to a more generic implementation (similar to deCONZ and zigbee2mqtt) doesn’t really fit the current model. And it’s not specific to Zigbee.
But I bet it is a huge disadvantage when using more generic/unknown devices. Just out of curiosity: Why the App model instead of a learning mode? What are the benefits of the app model?
There is something to be said for having an app-based model where third party developers can add device support without Athom having to do anything (other than deal with the app management logistics).
Home Assistant, for instance, uses a different model: their project repository is open to everyone, and integrations (which are basically the HA-equivalent of “apps”) are part of the main project. It’s more open (everyone can validate/check new integrations) but as user you need to upgrade the entire app if you want to pull in a new integration.
But I do agree that Homey has a big disadvantage because it cannot really deal with generic/unknown devices. You need an app to get the devices working, and developers are running into issues with Homey’s Zigbee implementation being incomplete.
Within Home Assistant and the use of the Conbee 2 Zigbee adapter you are dependent on which kind of brand/devices they support. You can’t add every device but having that said it is way more stable and faster then within Homey. There is also no need to be within 10cm reach of Home Assitant for adding a new device.
So it depends when the rewrite is finished if it can compare with it.
That’s true, but because it supports more Zigbee profiles, device support is typically added quite quickly. And even when a device isn’t specifically supported, you may still be able to get it working by using bindings (there was/is an issue with support for the Xiaomi illumination sensor where you need to explicitly create a binding between it and the controller, but when you do, it works just fine).
Athom learned the hard way that a generic approach doesn’t work (at least: for zwave). In the early days the whole zwave-stack was turned off and had to be rewritten…
I think if they won’t come fast with the necessary update, they are going to loose the fight.
With the economic recession underway it’s gonna be difficult to survive with a product that isn’t reliable.
Homeassistant is more and more growing to be the succesor of Homey.
Its a shame really, and this is the kicker and main annoying point from my point of view and frustrations.
I brought Homey rather than go open source, because it had zwave, zigbee and wifi and RF (never used RF but its there) all in one, and a web interface, so i could manage over local network.
Down in New Zealand we have limited ranges of zwave things, as we are not on the EU or USA frequencies, so Zigbee had the ability to fill the gap. The buttons work most of the time (better now i did the antenna mod), temperature sensors update anywhere from 30 mins to 4 days, so i cant rely on them to trigger stuff. zigbee led controllers, some cant dim and some cant turn on and off, in basic mode even, so they are still in their box unused.
So essentially I’ve been waiting almost 2 years to be able to use my Homey like it said you could on the box.
Athom has finally mentioned the Zigbee rewrite in their Facebook post about the tablet apps (see the replies).
It ain’t much, but more than nothing…
Hopefully the fact that the meantion it, means we don’t have to wait very long anymore.
Well, since I am already waiting for over 2,5 years I consider this as good news. Our suffering seems to come to an end soon … Whatever Athom’s definition of soon may be
It might be good for Athom to place this information on this specific topic, or on the forum’s update page. I’m sure this information would make people more willingly to stay with Homey.
Although i’m not consideringto move away, i see a lot of people doign a workaround for Zigbee.
but having that said it is way more stable and faster then within Homey. There is also no need to be within 10cm reach of Home Assitant for adding a new device.
