[mmooss]

If the solutions are so simple, what are they? Yes, incentives are distorted, but what system - that does not rely on everyone being on a higher plane - would have better incentives?

By your definition, every human endeavor is dismal and always has been - all are corrupted and flawed to some degree. Is there evidence that current science is more dismal than others or than before? You can look at any day in history and see people saying the same things about how it's so dismal and not like the good old days.

> No wonder the general public distrusts "the intelectual elite", we deserved it.

The general public has no idea about scientific publishing, publish or perish, or the distorted incentives it creates. Science has delivered at an incredible level for centuries, arguably more than any other human enterprise. Covid-19 vaccines were available in record time - it wasn't the science that caused it to go somewhat off the rails.

[zemvpferreira]

I didn't claim there's a simple solution, rather that the simplified system of publish-or-perish doesn't work. Personally I believe you can't replace a management layer that's aligned with the institution with a career progression based on metrics. Embrace complexity, let managers decide who should be promoted, sack or pay those managers based on their performance. Propagate onwards through the ownership chain.

The fact that Covid vaccines worked so well is precisely an endorsement of a hands-on, results-driven approach.

[mmooss]

> Embrace complexity, let managers decide who should be promoted

You know better than I do about academia, but speaking generally about management that sounds like an issue not of complexity but of bias and corruption, which are prevelant enough in academia as I understand it: Faculty advisors have so much power they can crush careers, leading to abuse (including sexual harassment and abuse) and corruption.

For a manager to do it effectively, they need objective, quantified metrics as a major input. 'I can't manage what I can't measure.' And to manage managers based on performance, we need to measure their performance. Which would seem to bring us back to publishing, impact factor, etc.

> management layer that's aligned with the institution

Why do we care about alignment with the institution (you mean the university?)? Very broadly, institutional interests are often quite corrupt and cruel to people and society. It would seem to pervert research, encourage corruption, and harm scientists personally.

> The fact that Covid vaccines worked so well is precisely an endorsement of a hands-on, results-driven approach

What do you mean here? I don't know how the Covid vaccine research was managed, other than non-tenure-track faculty earned a Nobel!

> the simplified system of publish-or-perish doesn't work

The point is, maybe there's no better solution. That's often the case in practice. The way I think of it is that theory prioritizes truth over having a solution - no 'truth', no theory; that's science. Practice prioritizes having a solution - no 'truth', do the best you can; that's engineering, management, clinical healthcare, etc.