I regard myself as a fairly technical hobbyist. I’m an electronics engineer but not daunted by programming. I’m not trained but can passingly learn most languages to get things done although I’ll never win awards for coding.
Over the past 5 years I’ve been into-outof-into and currently outof OH. I admire its goals, architecture and adoption but it hasn’t been friendly for me. I feel I’m regarded as one of the encumbrances drawn in and then disparingly supported, or dismissed by the inner circle. I’ve had some helpful but mostly dismissive support and I don’t think I’ll ever be up to the contributory level as I’ve seen some better, arduous and well intentioned peoples journey result in frustration and abandonment. All in all perhaps it’s a little conceptually beyond me and so I struggle to feel comfortable.
Why am I posting this ? Because I like the technical quality and adherence of OH to standards but I’m not on board now and quite honestly I’m enjoying the alternatives, contributing and getting supportive feedback. I feel OH is at a pivotal moment, especially with the 2.5 changes and questionable forward stability and compatibility.
However for me OH increasingly comes over as elitist and I’m not in that club, or wish to be, I want it to be evolving, feature leading, liked, friendly and adopted and to feel comfortable with it as one of my possible choices, but I don’t and I just wanted to let you know…
@J-N-K, @sihui, based on a close reading of @xAPPO’s post, I believe he is referring to contributing to the code and not support he has received here on the forum.
I’m not personally involved in contributing code to OH, but I suspect, based on what I know, that what you are interpreting as elitist is probably more of a combination of cultural differences (the bulk of the core developers are German I believe and some of the terseness in their postings and writings can be wrongly interpreted as rude, angry, or dismissive by native English speakers) and the fact that the core developers are way overloaded with work. Some of this might be self inflicted (the build system move might have been better handled incrementally or in a separate branch) but that doesn’t change the fact that they are all stressed and have little time to help newcomers right now.
There are two aspects that need to be handled separately.
Contributing to core:
I fully agree with the OP. Core contributions are hard, no visionary discussion possible, strictly sticking to the current paradigms. The code is ageing, some parts have never been refactored and very much reflect OH1 concepts with OH2 stuff wrapped around it.
Also: No documentation, not even simple readme files for the different bundles. No automatically build javadoc
Binding development: I think this target is moving forwards. Many people recently helped to make development possible in different IDEs. If you do not use dependencies, the code compiles fast, tests run, deployment can happen directly to a running OH instance.
The current initial setup is not finished though. Kai, as the initiator of OH, proposes to use the Eclipse Setup. But that takes 45minutes to complete on a powerful machine with high speed internet. That need to be improved dramatically. Compare that to 5 mins to setup an add-on development setup for home assistant.
Second weakness is dependency management and how to distribute build artifacts, eg share with others.
