[grandempire]

I think you wait a few decades and you have thousands of people using their full brain power and attention to continue to secure that funding. And not just money - but a continuing affirmation that their work is valuable.

> you just don't know in advance which ones.

at the limit, this sounds like “you can’t question what’s good science because it’s unknowable, but it’s also a moral imperative to fund” which leads back to my original question. If that’s true, which projects and fields get to forgoe scrutiny in hopes of paying off?

[jltsiren]

It's fundamentally no different from what VCs are doing, except the timescales are longer. You expect that most startups are going to fail but some are going to be really successful. Because you can't predict the outcomes in advance, you fund a large number of startups, hoping to catch the successful ones. There are some heuristics that should help you pick the winners. Or at least you hope so. But you also know there is a real chance that your intuition is wrong and the heuristics just make your choices worse.

[grandempire]

VCs vet investments. A good opportunity is one that might make money. A bad investment is one that won’t,

Shouldn’t we be asking a similar question? Which research projects are going to deliver insight and value for the public?

That doesn’t mean every project needs to be perfect from the start, but it does mean a lot of what’s called science now is not included.

[acdha]

> Shouldn’t we be asking a similar question? Which research projects are going to deliver insight and value for the public?

Yes, we’ve always done that - and quite extensively. I would recommend learning more about this process: it’s run by people who care deeply about scientific progress - nobody gets into it for the low pay - and if there seems to be a simple improvement, the odds are high that someone made it in the previous century.

I have a number of friends and family members who are academics and they spend a lot of time on each grant explaining how their research will advance our scientific understanding and linking it to other benefits (e.g. low-level neuroscience isn’t going to lead to new medical treatments directly but it provides the foundational knowledge which those treatments are based on).

Echoing the peer comment, if you were to pick any flaw it’s that we probably spend too much money on betting relative to the savings. There’s a lot of good research which doesn’t get funded, so it’s not hard to fill your budget every year with qualified proposals.

[grandempire]

All of that can be true, and yet there are simultaneously so many papers and projects that are not worthwhile being funded. How does our university system exist in the current set? Where is the money coming from?

> they spend a lot of time on each grant explaining how their research will advance our scientific understanding and linking it to other benefits

Do those come true? Or is this just an exercise in diligence.

> I would recommend learning more about this process

You’re right. I should learn more. I really want to understand.

[jltsiren]

And there is plenty of vetting.

Writing a grant application typically takes weeks of full time work, split between the prospective PI, their collaborators and trainees, and administrators. When the funding agency receives the application, there is administrative vetting to ensure that the application meets all formal requirements. Then there is vetting by internal and external experts, who evaluate the application for both scientific merit and whatever other values politicians happen to prioritize at the moment. If it looks like that the grant will be awarded (typical success rates are 20-25%), there is further administrative vetting to ensure compliance with various regulations. And this entire process typically takes anything from a year to a year and a half.

If anything, the process is inefficient, because there is too much vetting. Especially considering how small the individual grants are.

[grandempire]

You’re telling me we know it’s quality because there are hoops to jump through to get the funding. But shouldn’t the results tell us that? Do we have a list of the most recent successes from these grants?

[jmye]

In your view, the government should’ve not funded the research that resulted in, say, the internet, because who could’ve known it would be valuable before it was done? Is that right?

Then there’s this weirdly pervasive (on the political right) attack on researchers that supposes they’re all just out for funding, damn the actual science. Is that a reflection of your values, projected on people you don’t know? Is it in actual, widespread evidence? Are you just begging the question for fun?

These are basically nonsensical objections that, I’m guessing, have no basis in reality as no evidence is given. There are a whole lot of listed studies available - it seems like it would be easy to find examples of things that shouldn’t be funded.

[grandempire]

> In your view, the government should’ve not funded the research that resulted in, say, the internet

No. I’m saying just because things had unforeseen value in the past does not mean we should not scrutinize which projects we fund.

In other words, having bad prospects for utility or success is not a virtue.

The limit of that argument is that anything and everything deserves funding because it might be useful even if its prospects look terrible.

> These are basically nonsensical objections

What’s nonsensical is to say funding science means good things happen, not funding science means bad things happen. What’s science? I don’t know. Everything from particle physics, to elementary school surveys.

[jmye]

> What’s nonsensical is to say funding science means good things happen, not funding science means bad things happen. What’s science? I don’t know. Everything from particle physics, to elementary school surveys.

I think, in aggregate, it’s trivially true that funding science results in good things happening. I think if there are specific studies that appear valueless, they could be assessed for future potential outcomes, wherein “we don’t know” is a viable and not unvaluable answer.

I don’t think anyone disagrees that there’s value in oversight, they , and I, just disagree that not knowing, right now, the future outcomes isn’t really indicative of anything and is not a comment on the presence or absence of that future value.

[grandempire]

> it’s trivially true that funding science results in good things happening

It’s trivially true in that you’ve defined science to mean “good and valuable research”.

Let’s make that more concrete. What qualifies as science and what’s not with funding. Is doing a survey about whether having the color red in a classroom helps student performance, science?