I went into programming PLCs for industrial applications. Totally stagnant, backwards industry, were the platform suppliers basically make a living, handing in

I went into programming PLCs for industrial applications. Totally stagnant, backwards industry, were the platform suppliers basically make a living, handing in advances from 20 years ago to the industries suppliers in the walled garden. Its stressful, every machine developed is a unicum, software reuse is widely not used and often not possible, due to the "lowest common denominator" setting up what can work for a team. All good programmers leave the field as fast as they can. The others, who cant and are worth the salt, are in eternal fear of the A-day or G-day. The day one of the large valley software companies turns around and says "boo". They could easily replace the whole infrastructure of used software for all machines involved and install a steam like software market place for industrial controller components. No more "laptop-guy" sitting next to the machine for half a year and "thats normal". The only thing keeping this industry "alive" is the phobia of the great ones from anything "physical". I tried to leave, but this job is like a taint on your resume. People will not touch you with a stick, cause normal software companies have usually made bad experiences with the "programmers" leaving industrial automation. There are actually companies who try to lock there programmers in, by teaching them a "unique" bad software development style. I had people explaining git to me as a hot, new thing in 2018, at a company were we suggested it 8 years earlier. Yes, that resistant to change. In every company you come, you are viewed as a sort of union-busting freak of nature, because that is the only way one could be that "good" at making software. Obviously i must be possible to memorize a million globals, instead of packing them into objects and systematically develop my stuff. And you have that moment over and over again. In dubio pro rubicon.