Very good analysis. I agree that Home Assistant is basically a huge toolbox, especially when you consider other tools which can be used with HA such as MQTT, NodeRed, et al. There is no shortage of opportunity to tinker and fiddle.
I never got to OpenHab in my search. Since I wasn’t trying to find the “best” system, but rather was just looking for one that would do what I wanted, I stopped looking when I got the HA setup running. It’s quite possible that OpenHab is better, or something else. I stopped at HA when it worked for me.
It’s correct that HA has a lot of releases, each of which brings in some kind of new feature, so there’s lots of opportunity for neverending hobby activity.
But in my own experience, after I got all of my 100+ devices moved over, and a few simple NodeRed “scripts” in place, I could simply forget about the system for months and it would work just the same. There were likely a handful of new HA releases in that time, but I didn’t feel any need to install them.
That is, IMHO, the big difference I noticed after bringing up the HA system. I don’t have a long list of ongoing problems, quirks, and annoyances that I hope might be fixed in the next release. Or might not. Or might break something else. So I don’t feel the need to perform the endless updates. It all seems to “just work” today as well as it did when I installed my last release.
I do have a small list of features that will be worth an update. E.G., I have a handful of scene controllers that aren’t supported currently in HA. But I learned that is because the current HA release is using OpenZwave (OZW) 1.4, which doesn’t support those specific devices. However, OZW is now at release 1.6, which does support them. Some future release of HA will upgrade to use OZW 1.6, and then my controllers should work. When that happens, upgrading my HA might be worthwhile.
Similarly, my ZWave setup is not 100% reliable. Sometimes (few times a month), some light or switch might not go on or off as it should. I’m pretty sure that’s because of the ZWave polling that I still have to use because of my old ZWave devices. I know how to fix that, it just takes a little time and money, but in the meanwhile I can live with an occasional glitch.
Knowing what’s going on, and making my own decisions about what to do, removes a lot of the anxiety and angst.
I agree that there’s a future product release lurking in the HA world. In fact, there’s already a company “Nabu Casa”, which sells HA-based add-on services and has been hiring some of the developers…
I suspect (and hope) that this is heading toward a structure similar to what now exists for Ubuntu Linux - the basic Ubuntu system is open-source and free, and people with the interest and skill can download the system and set up their own computing environments. People (e.g., corporations that use computers) who don’t want the hassle can hire the company behind Ubuntu (Canonical) to use the same Ubuntu toolbox to build a system tailored to that customer’s needs and provide ongoing support. I.e., you can download, tinker, and fiddle to your heart’s content, or you can pay someone else to do it. Similarly, there are a lot of Ubuntu releases. But there are also occasional “LTS” (Long Term Support) releases, which are intended for install-and-forget environments. You can be on the bleeding edge, or keep your current LTS installation for 5 years. Your choice.
Something like that for “home automation” would be a win, IMHO.
Perhaps Ezlo will do it. Meanwhile, my HA setup does what I need.