Hacker News Reader

Comment by chatterhead | Hacker News Reader

We have to come to terms with this.

Academia was bought and paid for long ago and the money was used to build an incredibly broken and overly political bureaucratic engine of scholarly and scientific work that doesn't get anywhere near as much peer-review scrutiny as it should and commands far more respect in politics and legal proceedings than we should allow.

Universities and experts are the best we can do sometimes so we have to rely on it, but it doesn't mean it's truth or absolute and people like to use it as if it is to sell ideas like global warming instead of educating people on climate change.

View on HN

The 2 academics I know aren’t bought and paid for. Either could make far more money in the private sector than they make now. They work in their fields out of love of research.

This was my impression of most of my professors years ago too, especially as we saw so many quit for high paying engineering jobs at companies year to year — the ones who stayed aren’t in it for the money

I suspect it’s the same at the top: most senior administrators I bet would make more in senior Fortune 500 jobs

Agree. It's stagnant, fearful and bureaucratic. "Trust the science" never changed anyones mind.

However, I think your climate change example is a bit strange. If anything, it's big oil who's been trying to sell the idea that we don't have to stop using fossile fuels. Hiding evidence, and spreading confusion by paying lobbyists and scientists. Global warming was proved beyond reasonable doubt decades ago.

Climate change is one of the most politicized fields at the moment.

Any questioning it, even slightly means being banned from grants and academia.

Its also interesting that most climate models are NOT open source. Most recorded data from satellites is also NOT open source. So everyone works with a pre cleaned data set.

Its also worth pointing out that data sets like HadCRUT have never been audited by any respected scientist/group of scientists. This data was collected in stations not meant for long term measurements and they have a lot of errors. Just download it yourself and see. (Climate scientists are not really data experts, since they go from clean datasets in school to "clean" data sets in real life)

Calculating global temperature is also one of those things that is done in quite an obscure way, extrapolating too much IMO.

You seem to be much more knowledgeable about this than myself. But I was under the impression that while large climate models are used to infer the extent and impact of a changing climate, they are not actually an important component in determining climate change caused by carbon emissions to be a real phenomenon.

No other hypotheses about geophysics are required to show realistic and fault free simulations of the climate of the planet over several centuries to be generally accepted. Why should we apply such extreme prejudice to the hypothesis of climate change caused by the greenhouse effect? The basic mechanisms is quite simple and well understood, there is a variety of kinds of measurements supporting the claim that the temperature of the planet is increasing (meteorological temperature measurements, glaciers disappearing etc.). Full understanding of all the geophysical processes and feedback loops involved is not necessary, and very likely impossible.

I also have a hard time understanding the motives for such an enormous scam. Who would stand to gain from this except a relatively small number of researchers and the renewable energy industry? On the other hand, it's well documented that the fossile fuel industry has tried to sabotage climate science for the purpose of limiting political action on the issue.

> Most recorded data from satellites is also NOT open source.

This is false. NASA and ESA science data is free.

There is often an embargo period for very novel sensors, and always a delay of hours-to-weeks to allow processing to catch up, but it's free.

If it's the source code of the analysis pipeline you mean -- even though you said data - that's a harder lift, because the processing is complex. But even that is changing (https://science.nasa.gov/open-science-overview).

Even in the absence of the open science initiative above, today you can always get the raw data ("Level 1 radiances") or sometimes even uncalibrated straight-off-the-sensor data ("Level 0"), if you want to process it. (https://www.earthdata.nasa.gov/engage/open-data-services-and... -- "All EOS instruments must have Level 1 Standard Data Products (SDPs)")

And if you want to look in to how the processing works, there are detailed documents ("ATBD's") that explain how the pipeline works, for each data product. Also free.

> Climate scientists are not really data experts, ....

Dreadfully wrong. Do you work in this area at all?

Do you have any evidence of this?

HN isn't a journal and so academic sources of evidence really are overkill. However, I'll throw you a bone. Go to your favorite major journal and start looking at the "conflicts of interest" and "grants" section.

Once you take money from someone (except for NIST, in my experience) you're basically beholden to try your hardest to get the results they're looking for. Some scientists are moral enough to still return bad results. A lot of scientists aren't. There's a lot of garbage out there, and the worse the journal, the more garbage it gets. Famously, Phillip Morris studies "passed" the scrutiny of several major journals. It's amazing what greasing a few palms will get you.

"HN isn't a journal and so academic sources of evidence really are overkill"

I disagee, you should backup your accusations or statements (unless widly accepted) with some reputable source. This person claimed that science was bought and paid for. Did they mean all of it? Or most? That's an insane accusation that requires evidence.

"Go to your favorite major journal and start looking at the "conflicts of interest" and "grants" section. Once you take money from someone (except for NIST, in my experience) you're basically beholden to try your hardest to get the results they're looking for"

This doesn't mean people falsely data. You're simply providing motive.

Everyone wants money, you are using greed to then claim mass fraud in science

Is it really that insane of an accusation? This is how almost everything in the world works.

This isn’t only about greed either. People want their research published for reasons other than greed. For example, they want to move up in their career or achieve recognition.

After looking at a lot of medical studies related to COVID during the last couple years, I have seen first hand how biased and inaccurate many of them are. Some of these studies are even mentioned in major news outlet despite their obvious flaws when you actually begin to scrutinize them. Think big pharma providing research grants for studies that conclude their products are effective.

The OP never said that people falsify data as a result of receiving grants from interested parties. They often don’t have to. They can simply design the experiment in a way that doesn’t account for specific variables or behaviors then use the resulting data to reach a specific conclusion.

I remember seeing an article related to AI research on HN a little while ago that somewhat explained this problem. The grant money all goes to people researching deep neural networks which creates a reinforcing feedback loop. Since all the money goes to one branch of research, it creates very few opportunities to research competing ideas. I believe it was this one:

https://nautil.us/deep-learning-is-hitting-a-wall-14467/

I just looked at the first 10 articles in Nature.

Most declare no conflicts of interest. One author of one paper seems to have started a company based on similar technology: potentially a bias, but also potentially putting one's money where their mouth is. One other author lists some consulting work for a few companies.

As for grants, I doubt people are bending their results to appease the NSF or NIH. There's certainly groupthink in what gets funded. We're still throwing money down the ABeta-for-Alzhemier's hole, for example. That eventually shapes what topics get published, but maybe not the specific results. The recent Abeta articles are pretty negative, for example.

Reply on news.ycombinator.com