Historians have an increasingly strong incentive to tell dramatic stories

> This is such an academic take!

Of course it is, we're talking about academics.

> There's tons of fraud, bias and false claims in research, therefore the fix is to get rid of whatever trivial checks and balances we already have and let researchers do or say whatever they want.

You've completely mischaracterized the argument. The problem is that the metrics don't work and that the metrics themselves incentive fraud more than they incentive quality research. Call it the cobra effect or goodhart's law, but it's the same thing.

The reason I present this solution is because I'm saying to get rid of the carrot that leads us towards fraud. You may also be able to infer that I'm suggesting to take the money out of academics too, which is a different carrot pointed in the same direction. The point is to make the system only enticing for those that want to research and learn, not to those that want to get rich.

Unfortunately they'll always be able to publish false claims. That's because claims are almost never immediately verifiable. It can take decades many times. We also do not want to cause people to fear being wrong so much that they never attempt to make any meaningful progress. The process of researching is being wrong over and over until you find something that doesn't look wrong. Then you communicate it to others. And like I said, one thing that is discouraged by the system that I want to encourage is that of replication. This is your metric that you seek. Unfortunately it still doesn't completely been verified results but it does build high confidence. Replication allows us to make the mistakes and learn from them. It teases out the charlitians as their work will always fail to be reproduced. Just because you do not think there are guides to my system does not mean there aren't.

> They need to be micromanaged far more aggressively, by people who aren't impressed by them.

This is such a business person take! To overly simplify metrics, to believe everything can be codified by numbers, and any complexity is non-existent to the system. I say this as a mathematician. Metrics are guides and always limited. You assume naively that everyone is chasing money when most academics aren't. Pursuing my PhD is pretty similar to turning down hundreds of thousands of dollars, and profit maximization is at the masters level.

The problem you have is that you're looking at the goals wrong. You're making this system profit oriented. But the goal instead is to educate and to expand human knowledge. You may think that these are aligned. I'm sure you've created a good explanation for why these are aligned. But the common mistake far too many make is that you have to challenge your metrics and ask yourself how else they can be achieved besides your intended route. You must constantly ask yourself this and try to find easier paths than the intended one, especially as time goes on. But you're just one person and there's a reason all code eventually gets hacked. There's a certain irony that on a forum called hacker news that we pretend that mathematics/metrics can't be hacked but all code can, despite these being just different representations.

All metrics can be hacked because they aren't perfect rulers with all the nuances built in. It is why all rules are meant to be broken, because the environment changes and a simple set of words does not completely codify the intent and spirit of those words. You are always required to be human, not automata, because the incompleteness is fundamental to the system, to the world.

While the author is of course correct in his assessment, I do wonder if he is doing the same thing. Certainly his point could be better or at least simpler made without such a "dramatic" attack on storytelling. Stories make us less intelligent?!! Easy does it lad!

I think what you're describing is a sort of ensemble. I agree sometimes "conspiratorial" thinking is right, as in the invasion of Ukraine, or Hitler. Then the difficult part becomes how to weight the different world views in this particular time in history. But I agree at the very least you need to be able to simulate the thinking of ideologues in order to understand their thinking and actions. The invasion of Ukraine is a very interesting example where people underestimated how nefarious Putin really was, but if you read his writings on Ukraine, and do a bayesian update after Georgia and Crimea/Donbas the invasion shouldn't have been all that surprising. At least we should have assigned something like 10-15% chance.

See my comments on PAGES2K on another thread, reachable from my profile link. But as you've worked with the data presumably nothing in those posts is news to you, and you're well aware of the sorry state of paleoclimatology.

[flagged]