I can’t read your 2nd pic.
I’m very skeptical of this. Maybe I’m wrong, but I don’t see any major brands taking this up. Simply because of liability reasons.
HA is very far from being ready for the masses, so to speak. Things move around all the time, things break all the time, nothing is really easy or stable unless you tinker around a lot. If a big brand slaps a ‘works with HA’ logo on their mass market product and things break with some random breaking change down the road (and it inevitably will), then customers will reach out to their support rather than the HA forums. A manufacturer will have to pretty much guarantee perfect functionality of their product with an ever changing, highly technical and volatile open source project, where devs do major changes on a whim. The support load would be huge. I just don’t think this would be financially viable for a big consumer brand.
3 Likes
Christ my son can’t even get his xiaomi “works with alexa” lights to work properly.
1 Like
Supporting Amazons smart speakers is pretty much a requirement in todays market in order to stay relevant. Manufacturers will have to put up with support around those, that’s part of the business. Supporting HA is not, it’s purely optional for them.
I was merely trying to refer to the irony of guaranteeing things will work, when a major manufacturer doesn’t seem to be able to get turning a light on to work.
Yeah I know. The user experience with many mass market smart home products downright sucks and you’re left wondering how they can’t even get something that basic working properly.
But that’s exactly the point. If a big brand company can’t even get their stuff to work properly with something like Alexa, which has a well established UX, very basic user interaction without much configurability, a rather stable and well documented SDK, then how are they going to manage properly integrating with a platform like HA, that has no stable API, very bad or non existent documentation, ever changing UX and where users typically expect a lot more configurability. And these manufacturers would have to provide support for this. On top of all the Echos, Google, Apple and whatever other ecosystems they have to support already.
Financially speaking, compared to something like Alexa, HA is a tiny niche thing.
2 Likes
Yes I agree.
Tuya was such an effing disaster.
2 Likes
In theory all zigbee devices should work with ha out of the box (as in as soon as they come to market). I say in theory because zigbee is not Z-Wave and manufacturers have more leeway to do what they want such as tuya not adhering to the standards at all. When a device does not adhere to the standards a quirk is needed for it to work properly. Aqara devices require quirks and Ikea for example.
1 Like
Personally I doubt they would announce something like this without having a lawyer that developed the contract for the program. And they certainly wouldn’t go to that expense without a company lined up. Maybe it won’t be ubiquitous, but I really don’t see it having 0 adoption
1 Like
Naturally this hasn’t been communicated to us consumers, and I think long term it really should so consumers can have confidence in the badge, but I am sure this is way more specific in the contract.
All that has been said here is for the “cloud” and other badges, the manufacturer must maintain the integration themselves.
Home Assistant is open source, all of it: Yes, of course, is the answer.
Yup, that is the risk of the cloud, so you are absolutely correct. I would not recommend buying cloud devices. Yet, they can join the “Works with Home Assistant” program. The program is about things like (not limited to): interoperability.
For example, a weather forecast service could carry “Works with Home Assistant”, and works really well, is supported, open source, platinum quality scale, just works™. But a local weather forecast service is kinda non-existing. Just like an integration scoring platinum on the integration quality scale, can be using cloud-based communications.
Home Assistant is agnostic, cloud, local, zigbee, z-wave, MQTT, whatever… Together we all try to make it all work regardless of what it is or where it is coming from.
Buying local operating devices is regardless of that, still recommended.
Could be, maybe, maybe not. You are comparing two different things there.
There are many certified Zigbee devices that don’t work well with Home Assitant (or any other Zigbee coordinator). Zigbee tells something about the protocol, not inoperability.
5 Likes
So this program isn’t limited to “real” devices but also allows like “virtual” ones like payed services/subscriptions?
No, the blog post even starts talking about APIs and about “integrating their products”. It doesn’t say devices in the announcement anywhere. The menu is called devices and services.
Wouldn’t it be nice to consider this in the appearance of the badges?
It does, and also written and explained in the blog post explicitly.
Like on the main page written:
People always pull this out of context when they see fit. When questioned in this context, it means one doesn’t understand it. Home Assistant automations and data are local, you are in control over your data. It values the privacy of data. What you do with that data and automations and who you are sharing it with, is up to you.
Take a moment to compare data flows, devices & data access for automations using, for example, just Google Assistant.
That is where the difference lies. Again Home Assistant is agnostic and tries to be as inoperable as possible. We don’t deny cloud usage. Yet, we don’t recommend it.
4 Likes
Come to think of it: How would that work? both Zigbee and z-wave have multiple integrations that support different devices each. deCONZ, ZHA and zigbee2mqtt have different lists of supported devices, so do zwavejs and zwavejs2mqtt.
2 Likes
Zwavejs and zwavejs2mqtt use the same driver under the hood. As far as I know there should not be a different between those 2 for device compatibility.
Zha and zigbee2mqtt are entirely different, the code is not the same and handlers (quirks) are not the same.
1 Like
That’s a service level agreement, that wasn’t described but is certainly a good question. Hard to see how nabua casa can force something like that given that our (home assistant users) market share is a nothing compared to the big dogs.
1 Like
What happens if a core update breaks something in the internal HA API that the service or device uses to interface with HA and the integration breaks as a result ? Will it be stripped off the badge until it is fixed ? Who is responsible for fixing this ? Is the third party manufacturer contractually obliged to keep up with all core changes as long as they want the badge ? If yes, are there time limits ? What keeps us from ending up with a bunch of ‘certified’ integrations that don’t work because of core changes the manufacturer didn’t anticipate or keep up with ? Who would be in charge to QA this after each core update and possibly to strip / award the badge back ?
Seeing how that Tuya thing turned out…
2 Likes
You answered that already above. And you know the answer considering the discussion above.
I also answered that above as well. So I’m not sure why you are asking now again?
2 Likes
Tell me about it… I’m personally hit by that as in: Wasted effort (at least, that is how it feels for me, the reality is that I’ve been able to get lots more devices working at least).
That said, yeah, we (and sadly me too) have learned from that. They were not partnered btw, nor will they have been able to meet any of the program requirements. Honestly, I hope they will meet the requirements for the program one day, as that would mean: Stuff would actually work nicely.
3 Likes
So… You quote this:
Which conveniently (not sure if you did that on purpose or not), quotes part of that sentence.
Next you quote:
minimum time frame for maintenance and keep cloud servers running for example?
The latter quote makes no sense, as that is not written in the blog post. I feel like you are trying to bend things into things you’d like to hear. However, as already mentioned above, the focus is on inoperability and a good experience. If it isn’t operable as per the agreement, the “Works with Home Assistant” partnership can obviously not continue for that specific partner.
…/Frenck
1 Like
no zigbee is a real crap shoot for devices ACTUALLY working, or FULLY working without special work. For example an outlet might turn on and off out of the box and join the network but depending how they implemented power monitoring may not report power use out of the box.
Unique devices like AirQuality sensors may have extra metrics that need to be specifically polled.
Combo devices might appear as multiple devices out of the box, but would be better served if the HUB combined them as one device.
Configurable devices might have flags that can be turned on and off that need to be known and setup in a profile.
1 Like