There are several things that are weird enough not to be taken as absolute logical premises. I mean they certainly can and have worked for many people in many c

There are several things that are weird enough not to be taken as absolute logical premises. I mean they certainly can and have worked for many people in many companies, but let's just think a little more about them, especially in an industry of R&D knowledge workers. Going from "IC" to a non-practicing manager (like the article describes and even recommends) is in many ways a lateral move: you are not going to use your preciously acquired past skills and expertise a lot, and you are actually going to do quite a different job. The only thing that will help you in said new job is your general knowledge about the industry, maybe the subfield, and maybe some of your memories about your past job. This should not be a very common "promotion" path (and maybe in many contexts this should not even be considered a promotion, more changing profession), this actually already is necessarily pretty "rare"; unless you are "managing" just one or two people (and even then), or people that just come and go quickly, it is doubtful it is actually so common, because even with a very large growth of the industry, this would require just not a sustainable overall exponential growth rate for it to be the most common career change. As for "creating an environment where others can solve the problem", support staff would often be more valuable. Last, we can also consider big successful community projects as more up-to-date examples of more modern organizations. Is Linus a "hands-off manager"? Certainly not, even if he mainly does not do exactly the same thing as when he started Linux; also when he intervenes and gives strong opinions on very technical subjects (even often down to quite detailed ones), is it some form of counter-productive "micromanagement"? In my opinion: also not. If your main subject is mostly R&D, technical leadership should be more important than "old-school" management (and to be clear, the same caveats apply about the possibility for tons of people to really become technical leaders, it is not possible everywhere just because of mere growth reasons). I doubt the described traditional hierarchical organizations are the most effective in this context.