As someone who successfully raised one puppy (disgustingly shameless clickbait: [1]) I can recommend a critical principle that applies to software developers as well:
Only positive reinforcement: for the dog, you don't even express disapproval when they mess in the house. You just carry them outside, plus express wild approval when they do it outside.
I don't recommend picking up your developers and carrying them around, but you get the idea.
"But they have to learn what's bad, too!" you're saying. Yes, but they'll notice that you're not doing that "wild approval" thing for certain behaviors, and instead showing them the correct one. They're a lot smarter than laboratory rats. Most developers are as well.
But the overriding reason is: you want the dog to be happy. Because then he'll make you happy.
I think the idea is to correct incorrect actions, not to emotionally penalise them - this is basically the same as no-blame culture where you change the processes that enabled the mistake to be made rather than punishing the person who happened to make it.
Note I don't say "intentionally hide it." I suggest leaving it to the dog (person) to notice the absence of approval, and also that he showed a better action.
Only positive reinforcement: for the dog, you don't even express disapproval when they mess in the house. You just carry them outside, plus express wild approval when they do it outside.
I don't recommend picking up your developers and carrying them around, but you get the idea.
"But they have to learn what's bad, too!" you're saying. Yes, but they'll notice that you're not doing that "wild approval" thing for certain behaviors, and instead showing them the correct one. They're a lot smarter than laboratory rats. Most developers are as well.
But the overriding reason is: you want the dog to be happy. Because then he'll make you happy.
[1] https://drive.google.com/file/d/1GhuA-XPj932SRtnVLReOBYXt0wb...