I switched from OpenHab2 to HA (I avoid using HASS because it’s “Hate” in German) couple of weeks ago. While I am very into Java, know nothing about Python and absolutely hate YAML (expressing context by indentation, how crazy is this?) I was able to “re-create” my existing setup in an easier and faster way than I was with OH2. In addition (or as a consequence) automation works more reliable than it did with OpenHAB2.
I think, at the core OH2 and HA are close to each other. Discovery is working okay at both, the PaperUI of OH2 is pleasant and most things work without excessive configuration. The rule engine of OH2 with the Xtext-based, Java look-alike syntax is possibly more powerful than automation in HA (haven’t played with AppDaemon yet), but it’s difficult to learn even with using the Eclipse based “Designer” application you never get it right at the first time.
I’m a convert from OpenHAB to HA as well. I am mostly z-wave and wifi devices. I have a few zigbee sensors that I wasn’t able to use when moving to HA as there isn’t any non-hub related way to talk to them (not that I have found yet anyway). At first I preferred the rule creation in OH (code based) over the YAML in HA but have since started using appdaemon which lets me use code based again for the more complicated stuff.
I also found that with HA and its base on python it was much easier for me to get involved without all the java development overhead.
Only thing I really miss is the zwave management that was kinda available with OH via the habmin addon. Similar is available with HA using openzwave but can’t reconfigure zwave easily on the fly with HA running. HA has been taking a few small steps in adding some zwave management into the HA config though.
I am a Java developer, have experience in C, C# and C++ and some other languages but never had used python before. I’d say definitely check out python, it is pretty easy to pick up I just started a few weeks ago. It is pretty powerful, and tons of resources online of course.
Right now I just use my own scripts and call them through shell commands, but I did install appdaemon, just haven’t written anything for it just yet…
I have been an openhab user for some time. It’s a great system although the learning curve is steep. Way steeper than HA.
Rule engine in OH is quite lacking and sometimes the solution is quite untrivial.
Also the development pace is quite… slow. Sometimes things get broken and there is nobody to fix it, for months. And it’s not only the case wtih bindings.
But hopefully it will be on tracks and developing. Always good to have an alternative.
Thank you all… we also have very helpful and responsive community here
Yes… I was reading the documentation and the examples on the appDaemon and have to agree with you. Seems very powerful and I would start there for sure for my automations. Allows you to write real apps with complex logic.
What still keeps to jump right in is the lack of a nice GUI/interface to have some control through my tablet. Quickly scanned through the HADashboard and the setup looks complicated. Is this true or just my first impression was wrong?
I also have to figure out what’s involved in creating a custom “Thing” which support events and other parameters. With Smarthings I have used a tablet (my custom Windows app controlled with http through SmartThings) to display “SmartTiles” through automated IE for house control, display keypad to disable alarm in AWAY mode, also as media player to play sounds and speak. I would like to reuse this setup and just create a thing to control it through http/json with HA (sorry, did not know HAAS means hate in German)
Any pointer on what’s involved in creating a Thing for HA? Is it a scripts based or i need to recompile to binaries? or I can use appDeamon for this? (in SmartThings was a plug in script or a “DeviceHandler” I can update at any point while system running)
For what it’s worth, I am the author of AppDaemon and HADashboard - I came from SmartThings and liked the idea of SmartApps so I reinvented them in AppDaemon, HADAshboard is born out of a community project In Smart things but v2 is a complete rewrite. I find the community here awesome, and the raw power and extensibility of HASS has let me extend its capabilities in ways that seem natural to a bunch of folks. Put simply, I see HASS + AppDamon + HADashboard as the potential that SmartThings never lived up to, coupled with an awesome open and helpful community, and rapid feature addition - I’m here to stay.
aimic,
thanks a lot for your awesome contributions and for taking time to reply to me. AppDeaemon and HADashboard already made me a convert I cannot wait to start digging in and understand more of the architecture
HASS: It’s python and other languages, open source, and I’d call it “modern”. It’s very IoT-ish and hacker-ish. Very little support for professional home automation products. I ended up with this product because I didn’t want Java and was intrigued with learning more python even though HASS has poor support for my hardware.
OpenHAB: Java and other languages. Open source, I’d call it more “traditional”. Less IoT-ish. More support for professional home automation products. I found this one easier to get running because the scripting is less abstract.
I am thinking about switching to OpenHab from Home Assistant. (becouse of better and free Alexa support in OH). Did someone switched to OH (or maybe tried and went back), Why?
Its probably worth noting with HASS Cloud this will likely bring HASS inline with OH2 which already has a similar concept and has done for a few months now.
I’ll take a stab at this. I tried Homekit with OpenHAB briefly. I believe it’s built directly in whereas HASS I think has a separate program to install. I’m not really sure functionally how they compare. I wasn’t happy with it when I tried it in OpenHAB so I never went back to it when I switched to HASS.
Are you sure you want to take that leap. I love OH and used it for 2 years. But the automation engine… The speed of development (major bugs were not fixed like for several months). That’s what brought me here. And now i don’t think i will ever go back…
Introduction of a new paid option (Cloud) without publishing the source code of the respective funktions behind is a signal for me to thing about alternatives. Usually, it is a start of deeper commercialisation… unfortunately. Thereare lots of examples…
