I tried to report scientific misconduct. How did it go? • Hacker News Board

Re: I tried to report scientific misconduct. How did it go?

#251

Post by hinkley » Wed, Jan 27, 2021, 3:05 PM UTC

andrewon wrote: post #226

dnautics wrote: post #31

In graduate school, in my lab there was a grad student who was kind of an unlikely "professor's pet". He was tall and had surfer's long hair with a bit of a hippie aesthetic. Anyways, he was also really completely clueless about how to do science correctly, but also, I guess, really good about playing politics (there was a time when he asked me to put some bacterial plasmid DNA on my mammalian cells. I told him "it d…

Having worked in both academia and industry in biotech field, I have to say that the bar of reproducibility is a lot higher in industry. In academia, the goal is to publish. The peer-review process won't care to repeat your experiments. And the chance that other lab repeating your experiments was slim -- why spending time repeating other people's success? In contrast, in industry, an experiment has to be bullet-proof…

I could sense this sort of problem even in CS and could not wait to get into an applied position as soon as possible. If you cannot build the thing you’re an expert in, you’re no kind of expert I understand.

Re: I tried to report scientific misconduct. How did it go?

#252

Post by fer » Wed, Jan 27, 2021, 3:06 PM UTC

missingrib wrote: post #220

DarkWiiPlayer wrote: post #195

Earlier quoted context omitted.

> Wikipedia is an encyclopaedia An encyclopaedia with rather low standards that many people sadly treat as an absolute source of truth. You're right that this isn't really a wikipedia problem though. It's a matter of education because an overwhelming majority of the population isn't competent enough to fact-check memes on facebook, let alone wikipedia, and if wikipedia doesn't do it either, then that responsibility i…

Is Wikipedia really as bad as you are claiming? Can you give some examples? It seems to me like a lot of the articles are accurate, and some of the check marked or featured articles are downright great.

Wikipedia is overall great, though highly politicized in some topics, for science/engineering topics there's actual review by other Wikipedians and sourcing is solid.

Moreover, the talk page always has anything that might be controversial about the article that you might be interested in.

Sure, it will rarely display incorrect data, but it happens less and less as antivandalism bots become smarter.

Re: I tried to report scientific misconduct. How did it go?

#253

Post by speeder » Wed, Jan 27, 2021, 3:08 PM UTC

mam2 wrote: post #184

cies wrote: post #178

Many stories like this and people wonder why the general public is less "believing" in science lately. I think sadly the general public is right not to believe as strongly in science as it maybe once did, the results though are dramatic in "baby bathwater" kind of proportions. I think science should fix itself. Just publishing paper should not be the metric to reward. A retraction should seriously reward the flaw fin…

> Many stories like this and people wonder why the general public is less "believing" in science lately. I. Eh im not sure bad studies is the cause. Scientists, especially doctors, wanting to use their authority i some debates while 2 of them can be saying completely opposed things maybe, however, contribute...

Problem is more serious than that.

What is happening is that the bad studies are being used for policymaking.

Examples: the "nutrition pyramid" that encouraged carbohydrates and blamed health issues on animal-based food, was later found out to be based on research that was blatantly corrupt, with researchers getting bribes from food industry to manipulate or hide results (a case of hiding results: one researcher that found out that vegetable oil causes decrease of blood cholesterol, also found WHY it happened, but omitted that part from his paper... the reason is that cholesterol is needed for cell maintenance, and consuming only vegetable oils cause a deficit from it, the body pulls cholesterol from the blood to repair itself, and even that might not be enough, with some people suffering damage).

Or a lot of pharma circlejerking that turns into law or regulations.

Or the paper mentioned in the article, that was about video-games and aggression, with many countries passing laws regulating video game consumption based and such papers.

Or the original reason Cannabis was banned (long story short: part of the reason is that they wanted to ban hemp fibers, that was being an obstacle to some newly invented synthetic fibers, some of the government people involved, had stocks of Dupont and other fiber companies, and "accidentally" banned hemp fibers while "trying" to ban the drug, based on manipulated and fraudulent science).

Or more seriously: the papers that recommended "Austerity" and basically destroyed the livehoods of millions of people, later were found out to have math errors that changed the conclusion completely.

And the list goes and goes on.

Re: I tried to report scientific misconduct. How did it go?

#254

Post by psim1 » Wed, Jan 27, 2021, 3:09 PM UTC

LockAndLol wrote: post #187

How are we, the laymen, supposed to trust published, peer-reviewed papers? They seem to be just a bunch of words now with little meaning. Before, I used to consider scientific papers... well, scientific and would try to base an informed opinion on the abstracts and conclusions. After reading blog entries like these and actually reading some papers (especially soft science papers which are easier to understand), even…

For one, consider the institutions and nations that the research is from. It's too bad that journals are not more discriminating, but we as consumers of articles can be.

Re: I tried to report scientific misconduct. How did it go?

#255

Post by TeaDrunk » Wed, Jan 27, 2021, 3:11 PM UTC

hobofan wrote: post #158

m12k wrote: post #77

Earlier quoted context omitted.

You're absolutely right - the big question of the day is not just "how do we counter disinformation?" but "how do we counter it at scale?" The bad-faith actors of the world have realized how incredibly cost effective disinformation is in an online world that can massively amplify messages, and which algorithmically selects for divisiveness and "engagement" rather than factually and utility. We need a CAPTCHA for trut…

Isn't the answer well know (improving education and encouraging critical thinking), but unsatisfying because it is hard to implement and changes only crystalize slowly? If everyone were to e.g. request sources for claims instead of taking them at face value that would act similar to a CAPTCHA and prevent the spread of misinformation.

I don't know if improving education and encouraging critical thinking is actually the silver bullet one may hope it is. HN considers itself a cohort with significantly more critical thinking ability, but it will make wide and unsubstantiated, highly upvoted comments whenever a political subject comes up involving censorship, hallucinogens, systemic oppression, or unions.

Of course, HN is also an invaluable resource when it comes to tech and sometimes other STEM subjects. It's just significantly less valuable for areas completely outside of it. I wouldn't trust HN as a neutral or critically thinking source for, say, the usefulness (or lack of usefulness) of gender studies.

Re: I tried to report scientific misconduct. How did it go?

#256

Post by m463 » Wed, Jan 27, 2021, 3:11 PM UTC

Johnny555 wrote: post #16

The article mentions Brandolini’s Law: “The amount of energy needed to refute bullshit is an order of magnitude larger than to produce it.” It's the first time I've heard it, but it's a very appropriate observation in today's world where misinformation travels faster and wider than correct information. If you're just making stuff up, it's much faster than looking up sources.

This goes double for privacy.

Re: I tried to report scientific misconduct. How did it go?

#257

Post by singhrac » Wed, Jan 27, 2021, 3:11 PM UTC

andrewon wrote: post #226

dnautics wrote: post #31

In graduate school, in my lab there was a grad student who was kind of an unlikely "professor's pet". He was tall and had surfer's long hair with a bit of a hippie aesthetic. Anyways, he was also really completely clueless about how to do science correctly, but also, I guess, really good about playing politics (there was a time when he asked me to put some bacterial plasmid DNA on my mammalian cells. I told him "it d…

Having worked in both academia and industry in biotech field, I have to say that the bar of reproducibility is a lot higher in industry. In academia, the goal is to publish. The peer-review process won't care to repeat your experiments. And the chance that other lab repeating your experiments was slim -- why spending time repeating other people's success? In contrast, in industry, an experiment has to be bullet-proof…

I can second this. Working in industry, the bar is quite high for rigor. The general attitude of industrial researchers is to be very very skeptical of academia, since a lot of things just don't reproduce (cherry-picked data, p-hacking, only work in a narrow domain, etc., etc.). These researchers are almost all people with PhDs in various science fields, so not exactly skeptics.

Re: I tried to report scientific misconduct. How did it go?

#258

Post by jjoonathan » Wed, Jan 27, 2021, 3:11 PM UTC

bnralt wrote: post #246

jjoonathan wrote: post #233

Earlier quoted context omitted.

I don't like the science-bro movement, but I also think they might fill an important niche. The anti-science movement has too many people and too much time and too high of a (answer time / question time) gish-gallop ratio for scientists to possibly engage with. If scientists try to fight the anti-science crowd, they will lose. Science bros, for all their faults, can trade blows on more even footing, and that's someth…

Science bros often misuse "anti-science" to try to shutdown opinions they disagree with. Hence people worried about the unlikely event of being killed by a nuclear power plant are anti-science, but people worried about the even more unlikely event of being killed by a super intelligent AI aren't. Misusing the word "science" (particularly by people who don't seem to have a good grasp on it) and turning it into a rheto…

Do you have an alternative for addressing the gish gallop issue?

We agree that science bros have problems, but unless you have an alternative I see them as a net positive, and not by a small margin.

Consensus formation is always messy, but that's not solved by losing.

Re: I tried to report scientific misconduct. How did it go?

#259

Post by tinyhouse » Wed, Jan 27, 2021, 3:14 PM UTC

hinkley wrote: post #251

andrewon wrote: post #226

Earlier quoted context omitted.

Having worked in both academia and industry in biotech field, I have to say that the bar of reproducibility is a lot higher in industry. In academia, the goal is to publish. The peer-review process won't care to repeat your experiments. And the chance that other lab repeating your experiments was slim -- why spending time repeating other people's success? In contrast, in industry, an experiment has to be bullet-proof…

I could sense this sort of problem even in CS and could not wait to get into an applied position as soon as possible. If you cannot build the thing you’re an expert in, you’re no kind of expert I understand.

It's true CS has this problem. But that doesn't mean you cannot do reproducible research in academia. It's up to you.

Re: I tried to report scientific misconduct. How did it go?

#260

Post by mooseburger » Wed, Jan 27, 2021, 3:14 PM UTC

dnautics wrote: post #31

In graduate school, in my lab there was a grad student who was kind of an unlikely "professor's pet". He was tall and had surfer's long hair with a bit of a hippie aesthetic. Anyways, he was also really completely clueless about how to do science correctly, but also, I guess, really good about playing politics (there was a time when he asked me to put some bacterial plasmid DNA on my mammalian cells. I told him "it d…

Shouldn't you name and shame this guy in this comment? It doesn't seem like he deserves anonimity.