(Some others have noted that it’s taken significantly less time than 15 minutes to complete; we’re also aware of a ranking question that is reversed from others. We don’t make changes to surveys after we’ve started to receive responses, but we’ve noted these for future surveys.)
“Ideal” includes a huge number of hardware configurations since all the (no-video) portions of HA can run on almost anything that has a very small amount of power! One can spend thousands more to shave nanoseconds off of something that was already fast. One additional minor point, the quality and responsiveness (latency) of both the network and the devices to which HA connects probably is a bigger concern in many configurations - correcting same may end up with a much bigger “bang for the buck”.
Hardware setup here (talking about the HA portion only) - on the right a PironMan5 case with an RPI5 8G and an ssd installed running HAOS on a VM under QEMU. The one on the left is an older RPI4 8g running weewx listening to an AmbientWeather WS-2902C, transmitting that weather data to weather web sites worldwide and also using REST api calls via several python daemons to update status of those transmissions to a custom HA dashboard (as well as CPU temp, % and memory % for both the RPI4 and RPI5 into HA dashboards on the RPI5 HA instance). Evidently the bare metal CPU temp and % is not available to a VM on the same machine, so this also handles that workaround.
It would have been more bang for the buck to put it all on one NUC but I already had the RPI4 and running everything (with HA Supervised) for a long time while I sat on the RPI5 before they went up inn price scratching my head on my preferred migration. Getting the PironMan5 and setting it up this way was more fun as an “executive desk toy” rather than some old used ugly NUC lol
(P.S. For fun with Music Assistant that I just added to HA, I then put all my music on the RPI4 (you see underneath it’s running off a USB SSD), then installed a JellyFin server on the RPI4. The folder with the music is shared with HA for Music assistant. Our phones have both Music Assistant from the RPI5 HA and JellyFin standalone app (connecting to the RPI4) to play music on the house stereo via bluetooth from the phones.) Why? Why not!?
(P.P.S. I do have video via RING but no other cameras - next step down the road may be video from other cameras I would add but it’s a not an urgent need and I probably would make major changes to the setup to add that (I would probably want more horsepower as I prefer to never run anything maxed out)).
From the perspective of 99% of the HA owners out there I would say a used NUC with HAOS on a VM, anything else would propbably be hardware (and wallet) overkill.
Just submitted my responses, but hey — a few quick questions.
Who’s conducting this survey about hardware, the nonprofit Open Home Foundation, which IIRC doesn’t produce hardware on its own, or Nabu Casa, the for-profit company with quite a bit of hardware on the market today?
I know Missy said (edited to add: for clarity, she said this on Reddit) she’s looking into what the plan is about sharing results at the end of this — but assuming this is being done under the auspices of the OHF (which the footer at the bottom of the form says is the organization that created the survey), will staff from Nabu Casa get to see the results before the public (or get to see results the public never gets to see)?
That’s not to suggest anything shady is going on, please let me be clear. But the OHF and Nabu Casa nonprofit/for-profit arrangement is still pretty new, and there are still some unanswered questions about where the division falls between them. Particularly if the most monetizable parts of these survey responses are going to be treated as privileged information for a for-profit company (even a company like Nabu Casa), I think that should be clearly labeled for the community.
I will take the survey, but from my years of IT experience, I’d like to share an observation which I think should be foundational for all developers with regard to hardware:
Developers love to use the latest, fastest, sleekest, trendiest cutting-edge hardware.
Users like to keep their old, slow, bulky, unfashionable hardware for as long as possible.
As a developer, my desk always had the oldest models of hardware anyone in our organization still used. That’s what I developed to, even if it wasn’t my “daily driver.”
Look around. Every time Micro$oft announces a new version which won’t run on old hardware, users scream. There are lots of people still running Windows 7 because they refuse to give up their old hardware.
Yeah. I know. We can laugh. But it’s a lesson we need to keep in mind. Just because you have the coolest new stuff, doesn’t mean everyone does.
I also know that the OHF isn’t like Micro$oft. But we still want HA to be the world’s best smart home system. We want it to be used and loved by as many people as possible. Not just those who like to tinker and buy new hardware all the time.
I don’t want to make it seem like I am ignoring your question here but I genuinely don’t have an answer at this time, either. I’ll see how I can clarify the way the results are shared, however people who can help answer are now on time off and that will delay my response here. So, pause on that.
You are correct, though, the division is still new - even for us! While Nabu Casa is a for-profit company, it’s also the only commercial partner of the Open Home Foundation. However, your point about making it clearly labeled in the survey for the community is very valid. The survey is being conducted by the Open Home Foundation, indeed.
Trying to find out how we can/will publish results will also knock out sharing the feedback about clarity in how they are shared in future surveys themselves. Appreciate you raising your questions!
Totally understand the right people not being around to answer this quickly (that’s literally been the running theme of the week at my work, ha).
I really appreciate the thought that’s gone into defining the distinction between Nabu Casa and the OHF overall — clearly y’all are putting more brain cycles into preventing enshittification than most big projects.
And I presume you’ve kept the lines of communication pretty open between the two — in terms of Google workspaces and Slack orgs and all the things you navigate to do your jobs — so I kind of imagine things seem pretty porous between the two in daily life.
But I want to feel like the OHF could survive Nabu Casa getting sold to the worst, most craven VC firm out there, so I think it’s important to be considering those boundaries. And that’s just where I’m coming from (as we head into Labor Day weekend here in America, no less).
Thanks for the response and I look forward to seeing more about the result-sharing on this survey once folks are back from their well-deserved time off.
Oh, wow. That’s a scary thought. And unfortunately, not at all unlikely.
It would make absolutely no sense (in today’s world) for the owners or shareholders in Nabu Casa to sit on an asset of any value, and not sell it to a VC or PE firm. Customers, users and employees are irrelevant. It’s all about enriching shareholders these days.
I’ve contributed to at least two projects in the past which were sold out to faceless corporations which bled them dry and abandoned their customers and contributors. It’s not a good feeling to know that all the time and effort you donated was leveraged to make some corporation even more profit.
Sorry, none of this has anything to do with hardware. Just a real eye-opener for me this morning.
I’d like to just point out here that there are no shareholders for Nabu Casa, it’s a bootstrapped company. And Paulus is still CEO. I don’t expect him to sell it off in any fashion.
Could that change? Possibly. But it doesn’t feel like something that would happen, even if I am over at the foundation now.
As for this survey, our Researcher Annika is back and reviewing the submissions. I’ll be keeping in mind how we share these results out, but probably won’t have more information until October for that.
Thanks Missy! I appreciate you keeping us all up-to-date.
And I’m glad there are no plans to sell out anytime soon. Still, I have seen it happen. More than once. At some point the owner/founder moves on to other things, or wants to retire, or gets an offer they can’t refuse.
It’s really sad to see all the work, donated freely by volunteers who believed in the project, go to enrich a faceless corporation with no motivation to continue in the original spirit of the project, instead just mining it for profit until it dies from neglect.
Again, I’m not saying it will happen. But it can happen. It most certainly has happened, numerous times.
