Finally, I would argue that the large problem stems not from within science, but from how "science administratio/funding" has changed. Essentially, the idea of making science more (a)countable/measurable has lead to a state where scientists spend large fractions (>30%) of their time applying for funding. Also the funding is very much tied to some measurable outcomes, so there is just no time left to dedicade to reproducing results, because you can't get funding for it.

I believe we as society need to come back to a view of seeing science as a general good, without the need to produce some immediate measurable outcomes (in fact this is true for many aspects of society) . Unfortunately the MBAs are running the show almost everywhere so I don't see this happening anytime soon.

Yes, this is part of what I was referring to when I said that humans do all kinds of crazy things for all kinds of crazy reasons.

> In the end the best science always won

I'm not sure I agree. I would agree if you limited your claim to hard science--areas where we can nail down the best theory by lots of data and controlled experiments--but many areas that are called "science" are not amenable to that.

> by the very nature of us speaking about it, it can't be an existential crisis

I'm not sure I agree with that either.

> I believe we as society need to come back to a view of seeing science as a general good, without the need to produce some immediate measurable outcomes

I agree with this, but I would also add that science should be viewed as a tool for finding good predictive theories, no matter what those theories end up predicting. Unfortunately much funding of science now is given with the expectation that a particular kind of result will be what "science" predicts.

I think it's rather that we have had a bunch of great curious people who worked on interesting stuff and found out a lot about nature. But we can't equate that with institutional credentialed authoritative "Science".

I guess there is some underlying scientific ethos, that can be learned between the lines, but I don't think it's a clear method (as in the intro textbook definition).

No, we don’t just say that. Recognizing that “knowledge” about the universe is always tentative and contingent and that an understanding of the universe is to be replaced when one with strictly better at predicting observations is identified is the scientific process. And it's a pretty big change from the approaches taken prior to it's adoption.

> In this way science can never be really wrong, because it is always open to being proved wrong

“Science” is almost entirely a combination of openness to being proven wrong and a particular understanding of what it means for something to be proven wrong. Since it is an epistemology, it cannot be “really wrong” if it is consistent, except in terms of a different epistemology.

> I think it's rather that we have had a bunch of great curious people who worked on interesting stuff and found out a lot about nature.

We had a lot of that without the scientific method, too. What the scientific method did is to provide a framework in which the work of great (and not so great) curious people could be effectively aggregated so that they built-in each other, whereas pre-scientific approaches did that notably less well in terms of building an ever-better understanding of the physict universe, because work was judged by standards other than predictive power.

> But we can't equate that with institutional credentialed authoritative "Science".

Science is exactly about authority having little to do with institutional credentials.

Much of the 3 was created using approx 1, before 2 really. 3 is the original source of respect for the masses, the landmark results, the household names, Newton, Galileo, Copernicus, Darwin, Einstein etc. But then come along 2 with their little games and citations and positions and visa requirements and recommendations, CVs, conferences, journals, publish or perish, the push for sexy results for PR announcements and the next grant, p hacking, tyrant supervisors who only care about having a paper at any cost, massaging the data, not speaking up because if funding dries up for the institute, the newly birthed PhD student may lose her job and be deported, the big name prof must be worshipped etc. They co-opt the reputation accrued by the giants, and then if anyone has any criticism, they are scolded for being against the ideal 1, the noble endeavor and ethos and upright honest methodology. In many cases people good at playing game 2 need quite different skillsets than people who'd be good at 1.

But who will produce the next big item for category 3?

Is 2 a good mechanism to encourage practicing 1 to get to 3? I don't think so.

The ethos is simple: all scientific models are judged by their predictive power, and nothing else. If your model does not make accurate predictions, it is wrong, period. Nothing else matters.

I agree that there is no single "method" that guarantees that models will satisfy the above criterion.

Then String Theory is rubbish because it predicts nothing?

Every single theory in science started out with some 'predicate knowledge' and based on 'intuition', scientists reached out a few leaps with some ideas.

Some proved right, many proved wrong.

Some took a while to come together.

"String Theory" is different - it was not based on 'intuition' or any predicate foundation.

'Strings' are a fantasy - just completely made up.

Some people took a fantasy 'new origin' and put tons of math around it.

Since then, they've never been able to map it to reality, or even conceive of an experiment let alone do one.

What it looks like is religion - not only in it's foundation, but in the behaviour of it's acolyltes.

How does a Priest react if you tell him 'God does not exist'?

How does a String Theorist react if you tell him 'There is no String Theory it's just your imagination'.

Perhaps in the greatest of ironies, I actually believe God is more likely to exist than String Theory (!) or at least, there might be some kind of actual spiritual reality to our existence.

Strings? It's just an idea that snowballed into something.

How you come up with ideas and directions is a mystery. It's luck but not just. Kinda like investing.

Feynman also just described that step as "Guess it".

Its not clear at all that our institutional organized science process (grants, papers, academic positions and evaluations) correlate to better success at finding out stuff.

It depends on what the disdain is based on. If the disdain is just based on "your field isn't physics (or chemistry, etc.)", I agree that's not justified. But if the disdain is based on "your models don't make good predictions", I think it's justified. Scientific models are supposed to be judged based on whether or not they make good predictions; if they don't, they're wrong. Models that make good predictions are much, much rarer in the "soft" sciences--many such fields basically have no models with any predictive power at all, yet their proponents claim they are "sciences" and want their models to be used to, for example, make public policy.

Nowadays we have a meritocratic expectations and want to give equal opportunity to all, our cultural ideal of governance is democracy etc. It this framework a big thing to avoid and a big source of angst is that people will take these tax paid jobs and do jack shit and just sit around having fun. We want to avoid that, and replace any lazy non working people with people who will do it better.

For this we need some accountability. To see that papers get written and not only that, but they are also having an impact on other research (citations). What else can you do? You can have time sheets of working hours or computer software that tracks usage to see if someone just watches YouTube or reads the news at work... I mean, sure you could base it on common sense and talking with the researchers, but that's almost like asking them to write a report, but then why not publish the underlying work as well if it was good work?

Earlier however, science was more like a hobby and the background system was various forms of feudalism. Many scientists were rich noblemen who had the wealth to pursue their interests. If nothing came of it, they were still well fed and had nobody to report to. Or it was enough to have a patron who liked work enough that he fed the scientist. Or the it happened within the church with some theological justification/disguise, but that wasn't democratic either.

Since we now conceptualize power bottom up, from the masses to the intellectuals and not top down from the monarch/God to noblemen/intellectuals to the masses, we need a system to justify funding decisions to the people. And the easiest way is to point to papers and awards and experiments and machines etc.

In a way the old aristocratic system perhaps gave more freedom to a small number of scientists, today we have tons of scientists who all need to follow quite a rigid path.

I guess there was a time in-between, till the seventies or eighties maybe, when things were pseudo aristocratic with relatively few scientists with lots of freedom to do whatever the hell they please, including just reading newspapers or books all day long if they so decided.

I think our collective conscience is catching up and demands objective accountability to things are working in a just manner.

But papers and citations are the wrong metric for accountability. What we want from government funded science is models that make good predictions. That's what we need if the knowledge gained from government funded science is going to be useful to the government and the taxpayers. So that's the metric we should use to judge it. Number of papers written and number of citations are, at best, irrelevant to that metric, and at worst, actively opposed to it; it's much easier to generate papers and citations if they don't have to be based on models that actually make good predictions.

This is very frequently repeated, but I have never encountered any reason to believe it is true. Certainly, better science has been seen to win, in cases, but the amount of shit hurled first has always been enough to drive any sensible person to do something else less likely to attract hurled shit.

We can get a sense of how often correct theories are suppressed by looking at cases where a now-accepted theory was anticipated decades or even centuries earlier, but failed to gain traction at the time, not for lack of evidence, but simply because the evidence was unwelcome.

Theory of causation was understood a full century ago, but suppressed by the statistics mandarinate, and is only in the past 20 years starting to resurface. Evidence conclusively demonstrating human presence in the Americas before 15kya is still routinely dismissed. Requirement for surgeons to wash their hands took decades to catch on; to this day, more hospital ICUs have the traditional 4% incidence of fatal infection than the very gradually spreading 0%. The entire lack of any negative health impact from eating saturated fat was well demonstrated decades ago, but has only been accepted in this decade. These are just cases I know about; most have been better interred.

I am forced to conclude that there remain many more successfully-suppressed correct theories than accepted correct theories, and that things are getting worse, not better. I.e., correct theories still win sometimes, but the fraction of correct theories that win is decreasing in the face of tenured grant committees.

People still believe the Earth is flat.

Similarly in Alzheimer's research, still enforcing the failed amyloid "consensus".

Humanity can not afford to spend all their energy and resources for scientific research. We have many things we need for survival, and some for entertainment, etc. Obviously we do want to invest some amount of resources and human energy into research - but the question is how much is enough and which fields should we invest more or less ?

I claim that these questions are fundamentally unknowable / unsolvable. So we guess blindly (and very occasionally with a tiny bit of very myopic foresight), and mostly end up unwittingly relying on various historical accidents.

This is a very, very difficult question for humanity overall, and we're never going to get "right" amount of funding or "right" structure as there's no right answer. "MBAs" do and can hinder progress unnecessarily, but not always - they do exist for a reason and they represent society's collective attempt to avoid waste or runaway spending.

Governments spending bazillions just want some measure of accountability and that's it.

Aside from the more existential issues you've noted ... there probably a host of things that could be improved.

One thing may be the recognition that a lot of science just isn't worth repeating at all. To simply 'dial back the faith' we have any in any single experiment, and assume that every first pass is fuzzy.

I don't think anyone is accusing anyone of fraud here, this is not a problem of 'a few bad apples' - it's more of a broad systematic issue of 'looking for positive results'.

Perhaps some opportunities lie in getting more behind null hypothesis stuff as being 'respectable'.

I agree it is a cliche, but it is a very observable one over the last decades. Regarding resource constrains. I argue it's just the other way around. If we were resource constrained we would not build huge administrative bodies to check that the scientists are actually doing what they are supposed to. Take an example people here are likely more familiar with: in startups there are generally much less administrative processes, those come only in once companies get to a certain scale. Startups can simply not afford to pay lots of people to administer their "progress".

> Humanity can not afford to spend all their energy and resources for scientific research. We have many things we need for survival, and some for entertainment, etc. Obviously we do want to invest some amount of resources and human energy into research - but the question is how much is enough and which fields should we invest more or less ?

Considering the amount of energy we spend on things that are not needed and are actively hurting the planet and our survival, I would say we can't afford not to spend more on science.

> I claim that these questions are fundamentally unknowable / unsolvable. So we guess blindly (and very occasionally with a tiny bit of very myopic foresight), and mostly end up unwittingly relying on various historical accidents.

> This is a very, very difficult question for humanity overall, and we're never going to get "right" amount of funding or "right" structure as there's no right answer. "MBAs" do and can hinder progress unnecessarily, but not always - they do exist for a reason and they represent society's collective attempt to avoid waste or runaway spending.

MBAs are quite ill equipped for this tasks though. That's the my main gripe they try to apply a process that has problems even on the scale of a business (the problem of focusing on short term gains instead of long term sustainability) and try to apply it to society at large, even though there is no evidence that it would work.