Ah, there seems to be a mis-understanding. If I see it properly, the core of your critique is that experts’ data cannot make a normative statement. In the first section, you said “But they don’t decide if you should go home or not.” The second section, similarly, “ tell you what should or should not be done.” In the third, that “ data doesn’t care if you live or die.”
I agree, data cannot create values or goals. I accept that statement as fact, because I see no logical way for data to create a goal. Data, itself, doesn’t care.
I hope that I was clear about that point, when I said “The public makes the decision about relative values — ‘Is readability or stylization more important, and roughly how much more important?’” and that “Surveys would assess what I value…” The public is surveyed to determine what is valued. That’s where the normative statement, the should, comes from. The sailors all voted that they wanted to go home, but only the navigator is equipped to say which direction will get them there.
You also mentioned that “A competent CEO hires a qualified team of experts and tells them the overall direction they want the company to head towards.” That is true. Yet, is the CEO’s role worth their salary and benefits? As I mentioned, the company Gore and numerous others are quite successful at decision-making, without any managers. So, a company’s success does not depend upon the existence of a CEO — the CEO is redundant, not worth their salary. Most management is redundant, a drain on productivity.