'Science Has a Nasty Photoshopping Problem' (nytimes.com) 190
Dr. Bik is a microbiologist who has worked at Stanford University and for the Dutch National Institute for Health who is "blessed" with "what I'm told is a better-than-average ability to spot repeating patterns," according to their new Op-Ed in the New York Times.
In 2014 they'd spotted the same photo "being used in two different papers to represent results from three entirely different experiments...." Although this was eight years ago, I distinctly recall how angry it made me. This was cheating, pure and simple. By editing an image to produce a desired result, a scientist can manufacture proof for a favored hypothesis, or create a signal out of noise. Scientists must rely on and build on one another's work. Cheating is a transgression against everything that science should be. If scientific papers contain errors or — much worse — fraudulent data and fabricated imagery, other researchers are likely to waste time and grant money chasing theories based on made-up results.....
But were those duplicated images just an isolated case? With little clue about how big this would get, I began searching for suspicious figures in biomedical journals.... By day I went to my job in a lab at Stanford University, but I was soon spending every evening and most weekends looking for suspicious images. In 2016, I published an analysis of 20,621 peer-reviewed papers, discovering problematic images in no fewer than one in 25. Half of these appeared to have been manipulated deliberately — rotated, flipped, stretched or otherwise photoshopped. With a sense of unease about how much bad science might be in journals, I quit my full-time job in 2019 so that I could devote myself to finding and reporting more cases of scientific fraud.
Using my pattern-matching eyes and lots of caffeine, I have analyzed more than 100,000 papers since 2014 and found apparent image duplication in 4,800 and similar evidence of error, cheating or other ethical problems in an additional 1,700. I've reported 2,500 of these to their journals' editors and — after learning the hard way that journals often do not respond to these cases — posted many of those papers along with 3,500 more to PubPeer, a website where scientific literature is discussed in public....
Unfortunately, many scientific journals and academic institutions are slow to respond to evidence of image manipulation — if they take action at all. So far, my work has resulted in 956 corrections and 923 retractions, but a majority of the papers I have reported to the journals remain unaddressed.
Manipulated images "raise questions about an entire line of research, which means potentially millions of dollars of wasted grant money and years of false hope for patients." Part of the problem is that despite "peer review" at scientific journals, "peer review is unpaid and undervalued, and the system is based on a trusting, non-adversarial relationship. Peer review is not set up to detect fraud."
But there's other problems. Most of my fellow detectives remain anonymous, operating under pseudonyms such as Smut Clyde or Cheshire. Criticizing other scientists' work is often not well received, and concerns about negative career consequences can prevent scientists from speaking out. Image problems I have reported under my full name have resulted in hateful messages, angry videos on social media sites and two lawsuit threats....
Things could be about to get even worse. Artificial intelligence might help detect duplicated data in research, but it can also be used to generate fake data. It is easy nowadays to produce fabricated photos or videos of events that never happened, and A.I.-generated images might have already started to poison the scientific literature. As A.I. technology develops, it will become significantly harder to distinguish fake from real.
Science needs to get serious about research fraud.
Among their proposed solutions? "Journals should pay the data detectives who find fatal errors or misconduct in published papers, similar to how tech companies pay bounties to computer security experts who find bugs in software."
In 2014 they'd spotted the same photo "being used in two different papers to represent results from three entirely different experiments...." Although this was eight years ago, I distinctly recall how angry it made me. This was cheating, pure and simple. By editing an image to produce a desired result, a scientist can manufacture proof for a favored hypothesis, or create a signal out of noise. Scientists must rely on and build on one another's work. Cheating is a transgression against everything that science should be. If scientific papers contain errors or — much worse — fraudulent data and fabricated imagery, other researchers are likely to waste time and grant money chasing theories based on made-up results.....
But were those duplicated images just an isolated case? With little clue about how big this would get, I began searching for suspicious figures in biomedical journals.... By day I went to my job in a lab at Stanford University, but I was soon spending every evening and most weekends looking for suspicious images. In 2016, I published an analysis of 20,621 peer-reviewed papers, discovering problematic images in no fewer than one in 25. Half of these appeared to have been manipulated deliberately — rotated, flipped, stretched or otherwise photoshopped. With a sense of unease about how much bad science might be in journals, I quit my full-time job in 2019 so that I could devote myself to finding and reporting more cases of scientific fraud.
Using my pattern-matching eyes and lots of caffeine, I have analyzed more than 100,000 papers since 2014 and found apparent image duplication in 4,800 and similar evidence of error, cheating or other ethical problems in an additional 1,700. I've reported 2,500 of these to their journals' editors and — after learning the hard way that journals often do not respond to these cases — posted many of those papers along with 3,500 more to PubPeer, a website where scientific literature is discussed in public....
Unfortunately, many scientific journals and academic institutions are slow to respond to evidence of image manipulation — if they take action at all. So far, my work has resulted in 956 corrections and 923 retractions, but a majority of the papers I have reported to the journals remain unaddressed.
Manipulated images "raise questions about an entire line of research, which means potentially millions of dollars of wasted grant money and years of false hope for patients." Part of the problem is that despite "peer review" at scientific journals, "peer review is unpaid and undervalued, and the system is based on a trusting, non-adversarial relationship. Peer review is not set up to detect fraud."
But there's other problems. Most of my fellow detectives remain anonymous, operating under pseudonyms such as Smut Clyde or Cheshire. Criticizing other scientists' work is often not well received, and concerns about negative career consequences can prevent scientists from speaking out. Image problems I have reported under my full name have resulted in hateful messages, angry videos on social media sites and two lawsuit threats....
Things could be about to get even worse. Artificial intelligence might help detect duplicated data in research, but it can also be used to generate fake data. It is easy nowadays to produce fabricated photos or videos of events that never happened, and A.I.-generated images might have already started to poison the scientific literature. As A.I. technology develops, it will become significantly harder to distinguish fake from real.
Science needs to get serious about research fraud.
Among their proposed solutions? "Journals should pay the data detectives who find fatal errors or misconduct in published papers, similar to how tech companies pay bounties to computer security experts who find bugs in software."
Reverse image search on all newly submitted images (Score:5, Interesting)
It's sad that the pressure to publish leads to this sort of behaviour. Unfortunately at the moment the incentives are to cut corners to get on the career ladder and to keep climbing. We need to see people very publicly shamed when things like this come out; if the first name on the paper always got a public rap on the knuckles, perhaps this would deter senior researchers from grabbing the credit for their juniors' work. And if publishers just ran the images in a new paper against internet's images, we might get a few more caught.
The problem of course is the same as for any whistleblower in terms of reporting abuse in your lab, except your status as a junior researcher makes your position even more fragile. But overall this is part of a general collapse in moral standards on all sides; we need to admit there is a problem in society that makes this sort of behaviour somewhat acceptable.
Re:Reverse image search on all newly submitted ima (Score:5, Interesting)
It's sad that the pressure to publish leads to this sort of behaviour.
Yeah, I came here prepared to post a comment titled Abolish the 'Publish or Perish' tradition. But I'm at a loss to suggest how we might get governments and corporations to fund scientists to simply investigate and follow clues with no expectation of "results" - whatever that term means at any given time and in any given context.
I guess an early step in that direction might be re-calibrating our standards of usefulness to include null and negative results, which are valuable in eliminating unpromising avenues of investigation. As it stands, these are considered useless and are covered up, lest funding be withdrawn and careers ended.
Re:Reverse image search on all newly submitted ima (Score:5, Interesting)
But I'm at a loss to suggest how we might get governments and corporations to fund scientists
It is called a tort in legalese. Basically, if a scientist gets caught, the institution they work for (or whoever funded them for more complex research funding) pays a fine. The longer the time between publishing and getting caught, the bigger the fine. Also the bigger the false claim, the bigger the fine. Some part of the value of the fine goes to whoever caught the problem 1st. The rest goes to pay for peer review or possibly the original funder of the research. Its like in 3rd grade when the teacher punished everyone else instead of the student who misbehaved. That kid never did it again, did they.
Re: (Score:3)
I don't think this will work, corporations will simply force scientists to sign some kind of waiver to say they are not responsible. E.g. they will be contractors not employees. In the end I think you will just get richer lawyers, and higher costs: Look how risky research is we need to pass that onto our customers.
I think this type of behavior is trained into us from at least school age. I remember doing science experiments in school, and when the results didn't match I would adjust them to be more "correct
Re: (Score:2)
I am more concerned about censorship (Score:4, Interesting)
Senior Virologists have managed to stifle all debate about the Origin of Covid. Nothing supporting a Lab Leak ever gets published, while very weak articles supporting a natural origin do.
Whatever the ultimate truth, there is considerable scholarly analysis on both sides of the debate, and both should be heard.
Senior virologists do not want the Lab Leak discussed because it could lead to restrictions on their Gain of Function experiments.
The censorship has been quite chilling, and for me, surprising.
Re:I am more concerned about censorship (Score:5, Informative)
Senior Virologists have managed to stifle all debate about the Origin of Covid.
Yes, they did. They not only suppressed scientific debate about the possible lab leak but even managed to stifle public debate by labeling it as a "fringe theory" that was censored on social media.
This was a shocking overreach of censorship, a sharp rebuke to those who said it would never happen, and a great opportunity for those of us who warned of a slippery slope to say, "We told you so."
To be fair, the stifling has stopped, Facebook has apologized for suppressing the discussion, and scientific journals are now accepting evidence of man-made Covid.
Facebook lifts ban on claims of man-made Covid [theguardian.com]
Re: (Score:2)
Of which there is none. Studies have shown there is no correlation between the samples [cnn.com] of covid taken in China [deseret.com]. There were multiple strains rather than one, and they came from different locations.
All FB has done is let the conspiracy theories rise once again. Which they need to do to keep the eyeballs rolling in for advertisers.
Re:I am more concerned about censorship (Score:5, Insightful)
and in that year, there's no evidence that covid was man made.
False. There is plenty of evidence that it was manmade. There is no proof that it was manmade. We don't even know what they were doing in Wuhan. There were doing suspicious things, both in America and China.
We don't know where SARS-CoV-2 virus came from. There is some evidence it came from animals, some evidence it came from a lab, some evidence it came from animals and a lab. We don't have conclusive evidence either way. So it is fine to say, "we don't know pending more evidence."
And now I appeal to the logic in your sig: "Absolute statements are never true"
Re: (Score:3, Insightful)
False. There is plenty of evidence that it was manmade.
Nope. There's plenty of evidence that there wasn't. The whole "Scientists are not allowed to discuss it" claim is conspiracy bullshit peddled by those who ignore the papers which actually looked into it and found the virus to show no signs of either a) being man made, or b) exhibiting any desired traits of a virus a human would attempt to make.
Everything points to a completely natural mutation. Sticking your fingers in your ears, closing your eyes and shouting "la la la" doesn't change that.
Re: (Score:2)
There is a great deal of evidence supporting a lab leak, it just isn't definitive. Nor will it be because China prevented any real investigation and is obviously trying to save face regardless of what the truth is. They are even spreading propaganda that the virus actually originated in the US.
The claim that the virus naturally jumped from bats to humans right next door to a massive lab performing gain of function research on the precursor virus is the extraordinary claim. Especially in China, a hostile com
Re: (Score:3)
I don't know if scientists are no longer allowed to discuss it, but I have definitely seen it happen among doctors.
https://www.rnz.co.nz/news/nat... [rnz.co.nz] which states:
Council takes these matters very seriously and our concern is demonstrated by the publication of our recent guidance emphasising council's view that there is no place for anti-vaccination messages in professional practice, nor any promotion of anti-vaccination claims, including on social media and advertising by health practitioners,"
Look I am not anti-vax I got one as soon as I could but I cannot deny that there is significant pressure on people to conform. Imagine having spent 7 years training, and a student loan to match and then losing your ability to earn money from that. If that isn't a massive disincentive to have an opinion that opposes the norm, I don't know what is.
Th
Re: (Score:2)
False. There is plenty of evidence that it was manmade.
Is there though? There were editorials from 2015 IIRC (impossible to find now, since there is now so much coronavirus literature) about coronavirus being the next pandemic. This was after two near misses with SARS and MERS. This has been considered somewhat inevitable for a while by epidemiologists.
Oh and from 2015:
https://www.nature.com/article... [nature.com]
It was already known that something very similar to the OG human SARS virus was endemic in bats and showin
Re: (Score:3)
extraordinary claims require extraordinary evidence.
Ugh! If I hear this one more time I'll throw up. Who's to say that your claim isn't extraordinary? Why is the theory that COVID-19 jumped from animals to humans less extraordinary than it being developed in a lab and being released into the world by mistake? Maybe you put too much trust in what you're told. I, for one, have come to the point that I don't believe anything I hear any longer and thus everything said is plausible. It's better than blindly choosing one over the other. If it's something th
Re: (Score:3, Informative)
Re:I am more concerned about censorship (Score:4, Interesting)
Lab leaks also happen all the time. Some well documented, like SARS-2 escaping a lab in Taiwan.
People that assert that it cannot be a lab leak never actually discuss the actual evidence for and against. They only discuss authority.
In the unlikely event that you care, the following site summarizes the evidence, with lots of verifiable links.
http://www.originofcovid.org/ [originofcovid.org]
Also, note how my original post has been censored as Troll. You may disagree, but it was certainly not troll.
Re: (Score:2, Interesting)
I had a look at the site, but the very first bullet point is misleading. It says that no intermediate has been found, which is true, but not unsurprising. It took 5 years of surveillance to find the intermediate for RaTG13, for example.
Source: https://www.ncbi.nlm.nih.gov/p... [nih.gov]
So it doesn't seem like a very fair summary of the facts, and I didn't bother checking the rest of it.
The bigger issue is political. When COVID first started, we needed data. We needed to understand it. The WHO is the organization task
Re: I am more concerned about censorship (Score:2)
Re: (Score:2)
I think his meaning is the only plausible meaning of intermediate in this context. Viruses mutate all the time. Don't take the "genetic lineages" that they talk about too seriously. They exist as statistical probabilities, but mutations can happen de novo that replicate existing mutations. I understand that COVID mutates a bit less frequently than some others, but it still mutates frequently. The population of a virus is HUGE, and any mutation which is more successful is likely to reappear independentl
Re: (Score:2)
Having looked at that site, it seems to build a circumstantial case that since there was a biological research laboratory nearby, it must be the source.
On the flip side, looking up bat species that are known to carry coronaviruses, it seems that Wuhan is in or near their range.
Now, what's more likely - a virus crossing the species barrier outside of the lab, and thus causing a major outbreak? Or a virus crossing the species barrier inside of the lab, escaping, and causing a major outbreak?
I don't re
Re: (Score:2)
Your first there paragraphs are accurate. The implied conclusion doesn't follow. I don't think there *is* a valid conclusion. For one thing most COVID cases are asymptomatic, so when you're looking at human cases, you're missing most of them.
I have a pet theory that COVID was circulating in a small isolated village for quite awhile before someone with the disease took some stuff to market and passed along the disease. But there's no way to estimate how likely that is.
Re: (Score:2)
and a highly active nuclear weapons development program.
Chinese nuclear program is proof of covid escaping lab. Got it.
Re: (Score:2)
The available evidence does not allow that to be decided. So it's not a scientific question, it's a political question.
P.S.: Why do you even care? The question is what should be done, not who can I blame.
Re: (Score:2)
Pushing conspiracy theories in the middle of a pandemic is absolutely justifably cracked down on.
I don't think its justifiable, because it doesn't make that conspiracy theory go away, I think in general it just reinforces it, it has the Streisand effect. The more you try to hide it the more people do not believe you. Its like when someone keeps insisting they are telling the truth, all it does is make you think they are lying even more.
Re: (Score:2)
Re:Reverse image search on all newly submitted ima (Score:5, Insightful)
Re: (Score:3)
Re: improving the quality & efficiency of research, yes, can change that with things like pre-r
Re: (Score:2)
>Rather than focusing on punishing those who succumb to the pressures placed upon them by the current system, why not change the pressures?
Why not do both? Certainly they are at fault. You call it succumbing to the pressures. I call it lying, cheating, and stealing money. It *should* be illegal to take money from a source and produce a fraudulent result. You wouldn't settle for that behaviour in a carpenter or auto mechanic who is fixing your breaks, would you?
Carpenters and mechanics *also* face th
Re: (Score:2)
"The average salary for an Assistant Professor Non Tenure Track is $108,979 per year in US." [glassdoor.com]
"College professors work anywhere from nine to 12 hours per week teaching classes, an additional 20-30 hours preparing for classes and around 10 hours a week grading, reviewing and evaluating course assignments."
"Other studies report similar findings, with an average 53 hours per week spent on all activities"
I don't know where you've pulled that job from but you've forgotten to include how many hours per week they spend on research. Since this discussion is about research, could you please include this information?
BTW, the post-doc researchers I know don't make anywhere near that much money. Some can barely make ends meet. They're the ones with the most precarious working conditions & under the most pressure to publish or perish.
Re: (Score:2)
We should make replication (or falsification) sexy, because there's a shortage of it in some fields. It's good if scientists are publishing, but we should reward them for doing the work that's going undone, since that's how it gets done. Expecting people to come up with new cool shit all the time is maybe not the most realistic or desirable situation.
It happens, but not easy (Score:3)
We need to see people very publicly shamed when things like this come out
It does happen - recently the Nature paper on room temperature superconductivity at high pressures was retracted [nature.com]. But that took a huge amount of work from a few dedicated individuals determined to not let the result stand since there was strong evidence of data manipulation. But some conferences and even the institute of the main authors were quite happy to ignore the whole controversy.
Re: Reverse image search on all newly submitted im (Score:5, Interesting)
I have very little sympathy for the pressure to publish.
The pressure to profit exists as well, yet I could not imagine anyone defending a bank executive for cooking the books. Unethical behavior is unethical, whatever the reasons.
Re: (Score:2)
Re: (Score:2)
Re: (Score:3)
It's sad that the pressure to publish leads to this sort of behaviour.
Why does pressure to publish lead to cheating to ensure the hypothesis is correct? When did science become about being right instead of publishing a result. Even if something didn't work, it would often be a publishable result, if for no other reason than to allow other people not to repeat the same test expecting a different result. .
Re: (Score:2)
In the real world 'interesting' results get you publicity which turbo charges your career. Discover a new super conductor - tenure assured. Show that yet another mixture of metals doesn't have super conducting properties, back to the treadmill. Of course this should not be, but sadly it's true.
Proud supporter of her Patreon. (Score:3)
It put paid to the lie that people in science get there by merit.
Re: (Score:3)
I was about to do the same, so I googled "Dr. Bik Patreon" and I found this Reddit post, which claims that Bik is a sensationalist [reddit.com]. I don't know one way or another. The only red flag I saw in this Slashdot summary is the fact that Dr. Bik doesn't like images that are "rotated, flipped, stretched or otherwise photoshopped" which makes no sense. But I assumed that was just the summary is missing context so I thought nothing of it. The reddit post cites an example where Bik's theory was disproven, but she
Re: (Score:2)
From what I've seen of her process, she doesn't just cry fraud at everything that seems "rotated, flipped, stretched or otherwise photoshopped". That counter-criticism sounds more like a strawman. That just makes her look further into it, like actually reading what the paper says some figure is supposed to represent. Sometimes there's a good reason for
Re: (Score:2)
somebody who thinks they are better than everybody else
Someone who is "better than average" is not "better than everyone else"; they're better than half of everyone else. That leaves a lot of room for others to be better than them. Your "red flag" doesn't seem to be either red or a flag.
Re: (Score:2)
Bullshit, better than half would be the median, not the average. She is obviously so much better than everybody else that she skews the average to be the only one above it.
The median is valid as the center value only for highly skewed (relative to Gaussian) distributions. In this case, the sample size is so large that one individual cannot skew the distribution significantly just by being "better than average".
Re: (Score:3)
It put paid to the lie that people in science get there by merit.
Finding bad apples and cheaters does not make the statement a lie. If it did we can call every career a lie, which is just stupid on the face of it. Call out fraudsters when you do, but then don't make underlying assumptions about an entire field as a result.
Unless you have a peer reviewed article analysing the data showing that scientists are overwhelmingly fraudsters. Do you? If not, stop peddling anti-science rhetoric.
Re: (Score:2)
Unless you have a peer reviewed article analysing the data showing that scientists are overwhelmingly fraudsters.
Well, how would trust that study that said that? :)
I wouldn't call that a photoshopping problem (Score:5, Insightful)
It's a fake research problem. A gaming the system problem. Duplicate photos will just be the easiest component to identify.
Re:I wouldn't call that a photoshopping problem (Score:5, Insightful)
It's a fake research problem. A gaming the system problem. Duplicate photos will just be the easiest component to identify.
100% this.
Complaining that this is a Photoshop problem is like bitching about the shade of black used in the swastikas plastered all over the burning church. The actual issue, is a bit more significant as you've identified.
Re: (Score:2)
Science is a 2wonderful and beautiful thing (Score:3, Interesting)
Unfortunately scientists are people and in general not particularly better than the average person.
What's worse is that because scientists have spent so much of their lives pursuing very narrow professional goals they can be much easier to manipulate than the average person that views their morality, family and community as the pillars of their life.
This is why you can even see "revered" publications like Scientific American whoring themselves out for relevance
https://www.scientificamerican... [scientificamerican.com]
Note that doesn't say 2 genders but 2 sexes.
Re: (Score:2)
The language has changes: "Sex" is used by most publications today to describe physical biology, "Gender" is used by most to reflect social identity. And yes, there are many who deliberately confuse them, and many who spout absolute falsehoods about each of them, sheltering their nonsense behind political claims.
We want Sex. (Score:2)
Who cares Gender? We want Sex :)
Re: (Score:2)
Can you describe a distinction that is not "made-up"?
Any distinction which is provable is not made up. Gender distinctions are difficult to prove, but sex distinctions are relatively easy, so long as your definitions are flexible. Saying an organism is strictly male or female is complicated, but those words have accepted scientific definitions so in some cases you can, and in others you can't, and you can quantify why or why not. Definitions of gender are more multitudinous, and often not based on science at all, or only barely or peripherally. Consequently,
Re: (Score:2)
Human language is too fuzzy for things to be that simple, drinkypoo. Even when in the domain of technical jargon, terms can have ambiguities, multiple meanings, overlapping meanings, and so on. But the main weakness of language is oversimplification, and that weakness cannot be repaired.
For example, it is precisely because of oversimplification that I can say "there is a heap of salt on the table" without having to know precisely how many grams of salt are in that heap before I can speak. The fuzziness o
Re: (Score:2)
Even when discussing the scientific/biological distinction of "sex" into male and female, we find that these two terms encompass more than one fact about the person in question. For example, female has multiple separate characteristics, such as "vagina, breasts, high voice, absence of beard, low muscle mass."
Haha, what? Show me where that's written in a science-based book post-1800s
Re: (Score:2)
So, it appears that you are deflecting. Rather than respond to the points made, you have zeroed-in on a specific detail of how one point was explained, and fixated on that, thus derailing the conversation rather than engaging in it.
Perhaps I could have said something like: according to wikipedia, "female" has a scientific definition of "the sex of an organism that produces the large non-motile ova (egg cells)." And then we could have the same discussion about cases that occur such as hermaphrodites. The
Re: (Score:2)
So, it appears that you are deflecting. Rather than respond to the points made
If you don't want me to respond to your most ridiculous point, don't be ridiculous.
Re: (Score:2)
Unfortunately scientists are people and in general not particularly better than the average person...
This is why you can even see "revered" publications like Scientific American whoring themselves out for relevance https://www.scientificamerican... [scientificamerican.com] Note that doesn't say 2 genders but 2 sexes.
I agree with what you said, but I don't necessarily agree that your example is relevant. If sex according to chromosomal makeup differs from sex according genitalia, which do you use to determine sex?
I dislike most wokeness as much as the next sensible person, but I think you might be reaching a bit here with your implication that the SA article is an example of it.
Re: (Score:2)
Unfortunately scientists are people and in general not particularly better than the average person.
Here I disagree, or, at least, have deep regrets. The whole idea of the scientist is to look at things dispassionately, with no aim to personal gain.
Re: (Score:2)
Unfortunately scientists are people and in general not particularly better than the average person.
Here I disagree, or, at least, have deep regrets. The whole idea of the scientist is to look at things dispassionately, with no aim to personal gain.
Different experiences. Of the scientists I have known maybe at best 3 out of 10 had a real calling. The rest were people looking for respectable work.
Re:Science is a 2wonderful and beautiful thing (Score:5, Insightful)
What sex would you call someone who has mosaic chimerism with a set of XX and a set of XY cells all mixed up throughout their body?
You're not a scientist, you're not a biologist, you're an idiot with an axe to grind.
Re: (Score:2)
The fact those people exist, and for example, there are plenty of real life examples of "hermaphrodites" (a term they themselves hate..), doesn't fit the anti-woke/religious narrative, so gets happily ignored.
Re: (Score:2)
It doesn't fit the general cultural narrative in most of the modern world either, which is why sex assignment based solely on evaluation of external genitalia at birth remains a thing almost everywhere, and also why intersex surgery on infants is a commonly accepted practice [healthlaw.org]. Rather than learn to treat one another like humans regardless of whether or not we're unusual, the bulk of cultures make everyone as similar as possible, by violence if necessary.
Re: (Score:2)
And most people have no clue how many of those there actually are. It's sad.
Re: Science is a 2wonderful and beautiful thing (Score:2)
Intersex.
Since they comprise two different genetic entities in one body, their default pronouns are "they/them."
With a mind for sensitivity toward minorities, and not appropriating for one's self those things that minorities come by honestly from their genetics and heritage, anyone who isn't a chimera who uses "they/them" as their pronouns is guilty of the most heinous form of appropriation and marginalization imaginable.
One wonders how people appropriating "they/them" and everyone who supports them could b
Are the photos data or just illustrations? (Score:5, Interesting)
When I was in junior high, our science textbook had a photo of the Milky Way galaxy. I asked my science teacher how it was taken. He said "with a telescope". I asked how a telescope could see the whole galaxy when we are inside it. He got a puzzled look on his face and said he'd look into it. Turns out, it was a common stock photo of the Andromeda galaxy, a spiral galaxy that (at the time) was assumed to look very similar to the Milky Way. I realize it was just a way to illustrate something that there was no actual photo of, but it always struck me as dishonest, even in something meant for children.
Re:Are the photos data or just illustrations? (Score:4, Interesting)
it's great that your teacher followed up on the question though! It sounds like you both learned something:-)
We see the same problem with modern journalism. For example, any article about plastic pollution will usually have a stock image of a beach with rubbish on it, or that turtle with the straw stuck in its nose. Anything about power generation will have a stock image of some smokestacks etc. There's usually an image credit, but I think it could be made more clear that the pictures aren't pictures of the article's actual subject.
Re: Are the photos data or just illustrations? (Score:4, Interesting)
Mismatched imagery is an old, lazy fallback. What strikes me in contemporary reporting is the constant need to include the reader (or viewer or listener) in the report.
Shit like, "Susan was already delayed on her lunch break. How would you feel waking up after walking in front of a bus?"
Or, "Brian had heard of crypto scams. How would you respond to this email?"
I see it everyday now. It's like nothing can be interesting if it couldn't apply to yourself.
Pathologically narcissistic.
Re: (Score:2)
Yes! I don't know started this meme in journalism school (journalists still GO to school don't they?) but it seems to have crept into the "serious" reporting as well as the fluff pieces. I have no objection to fluff or lifestyle articles, but their tone doesn't have to bleed over to science reporting, or stories about war.
Re: (Score:3)
The photos are data. "Simulation results" mostly in graphical form and put in quotes for obvious reasons given what we're discussing. There's been a couple of articles on Slashdot about this before.
Re: (Score:2)
Build a Dyson wall!
No defense (Score:5, Interesting)
When it works it is wonderful. I recall finding a small article from 30 years past that let the researchers I work for build a new structure. I was in a lab where a grad student built a new molecule, and as others did it her data point became science. On the their hand, another gas student made an error in coding that lead to decade of building a theory that went no where
The pint is that peer review means the theory, math, technique is relatively solid. And the findings are of some interest. It dies not mean the finding has validity or are in any sense reflective of a general reality. This is not the pope imposing his will on creation. These are humble being trying to eek out clues of creation
It has been wildly successful. What we know is as we move from physical science to life science to the pseudo science of medicine and human society, the rigor becomes less prominent. Partially because isolating variables become harder, but also because the temptation for fraud is greater.
Re: (Score:2)
No mod points today, but thumbs up anyway.
Re: (Score:2)
The pint is that peer review means the theory, math, technique is relatively solid.
I'd argue it's a bit weaker than that. It's generally at the point where it means it's worth a scientist in the field taking a look to see if there's something worth looking at more deeply. Probably not obviously junk.
Cheating gets rewarded (Score:5, Interesting)
And people that do honest scientific work get punished. My own PhD took about a year longer because some cheater published data and an approach that was deeply flawed. I could only continue my work after all of the authors (minus the cheater) basically published a retraction a year later. The really sad thing is that I spotted all the flaws within about 20 minutes of getting the paper but then was unable to convince my prof of what was happening. He only saw the "reputable" conference and the other names on the paper and then his mind was shut. Of course, I never got an apology or anything like it. Now, I was lucky that enough people spotted the deception and complained to those authors. Without that retraction I would likely have not gotten that PhD. But I know of people that were not so lucky.
With incentives like that, is it any surprise people are cheating? And those 5% out of 100'000 papers are just the ones cheating by edition images. There will be tons of falsified data, manufactured date, and sometimes even manufactured references in there. It is, yet again, a case of 10-20% assholes messing it up for everybody.
Re: (Score:2)
That doesn't sound like a science problem as much as you worked for a toxic professor. Sorry to hear about your tribulations but happy that it worked out in the end. Nothing is quite as demoralising as being held back for no reason.
Re: (Score:2)
Thanks. The prof was reasonably non-toxic. But there was a cretin "senior researcher" in the picture with a level of actual understanding that was exceptionally low and a huge ego.
There is a science problem here though: That paper got accepted. That should never have happened. And the retraction should have cost the first author his PhD because that was his core contribution. That has not happened. Yes, I get that his funding was running out and he had to have something, but publishing cleverly disguised li
is it evidence or just explanatory images? (Score:5, Interesting)
Re: is it evidence or just explanatory images? (Score:2)
I agree. I want to know what fraction of the papers based their scientific conclusions upon the images in question. Many, many authors reproduce images from other papers, sometimes with correct attribution, sometimes not. These are typically introductory images that don't impact scientific conclusions - conceptual flow diagrams or experimental schematics. I consider that on a different level than falsified or manipulated data. But I don't want to dismiss this effort! What she's doing is really important,
Re: (Score:2)
Even parody papers can be cited accidentally. Have any of the "Sokal Hoax" papers been cited?
https://en.wikipedia.org/wiki/... [wikipedia.org]
Re: is it evidence or just explanatory images? (Score:2)
I wonder what you would get if you asked DALL-E to "make a post-trauma photograph of novel carbonfiber-weave demonstrating improved shear tolerance".
Re: (Score:2)
Changing it to "post-damage photograph of novel carbonfiber-weave demonstrating improved shear tolerance" (swap trauma with damage) give this:
https://labs.openai.com/s/2pOH... [openai.com]
https://labs.openai.com/s/79xE... [openai.com]
Pay the Piper (Score:2)
It would be interesting to see the areas of research affected and what position or claims were made in these frauds.
Another reason (Score:2)
At least be creative (Score:2)
journalists can fix this (Score:2)
Especially dodgy psychology stories
Journalists make iffy scientists look reputable (Score:2)
The problem for journalism is the same as for scientists - publish or perish. With journalists however the requirement is to publish far more often, and not to be left having failed to report the latest 'news'. Therefore to ask a journalist to ignore an interesting science story because it hasn't been confirmed is a VERY hard call because they know if they don't report it, some other paper / site / station will, and they will be in trouble with their boss.
Thanks to Publish or Perish (Score:2)
What Publish or Perish means, is that by challenging published papers, you are challenging the continued employment of some people.
Paraphrase a saying, it is very difficult to get someone to understand a thing when his continued employment depended on him not understanding it. It is of no surprise that few took action when the problem was exposed.
To make people act, you need to make it so that their continued employment depends on dealing with this problem, preferably properly rather than sweeping it under
Failures of peer review... (Score:5, Interesting)
A lot of this is a failure of peer review. Photoshopped images ought to be caught by the reviewers. So should fudged data. Too often reviewers just assume the authors are telling the truth, and pick on the presentation and the references. Reviewers need to be just that bit skeptical, and keep an eye open for cheating.
All raw data, source code, and everything else needed for verification and full replication must be made public, along with the paper.
This is also where independent replication needs to be funded and encouraged.
What world do you want? (Score:2)
When the imperative is to publish to sustain funding or maintain employment and quantity trumps quality this result should come as no surprise. (I have long time friends in academia and have just a hint of how vicious this can get). But wait, there is more. Not only is there a photoshop problem but more seriously there is a fake data problem. More than a few research papers cannot be reproduced by others carefully following the described methodologies in the paper. So photoshop is just one part of the fake
Anti-cheating software applied to papers? (Score:2)
I got my degree before plagiarism via the internet was a thing but I believe it is common or routine for student assignments in college to be processed with plagiarism detection software when it is turned in. How is this not a thing for research papers? Or even authors of regular books? Maybe it is a narrower problem to solve to detect when a student has copied wikipedia but this image problem does seem like it might have been detected via a reverse image search.
Get ready for diffusion models (Score:3)
You think image reuse is a problem? Wait until you get a load of image generation.
In the mean time, kids, make sure to flip or rotate the image you're borrowing :)
Science is a religion now (Score:2)
Science as is known to history and physicists has largely broken down. Most disciplines are now faith-based. Papers get published because they confirm our ideas of what's right. None of the normal checks and balances are used anymore. It's a priesthood of debt slaves all trying to maintain the illusion.
Big data killed science (Score:3)
As soon as scientists started publishing the results of "big data" analysis, science's death knell rung out. Rather than form hypotheses and develop rational tests, we now just feed more data into a computer until it starts to see patterns where none exist. It's the computer age equivalent of numerology. If you keep throwing in new numbers, eventually you will start to find relationships.
Ben Goodacre's 'Bad Science' (Score:2)
Has a great discussion of this problem. The answer in the medical research community has been the requirement to register tests and what they are looking for to prevent such data mining. It's certainly a partial solution. But beyond medicine, where the profits are obvious and so have attracted attention, where the benefit is in terms of tenure / status it's hard to prevent. Add in the honest mistake, where the data did actually point to the conclusion in the test conducted, and it gets very hard to adequate
In related news... (Score:2)
A small percentage of people are cheating. No matter what group you are talking about, that contingent will always be present. The fact that it's in medical research is only of minor interest. Before I read the article, if I was asked to bet on it, I would have put nearly all of my money on, "some people are cheating."
This guy really took a paragraph to say: (Score:2)
>Using my pattern-matching eyes and lots of caffeine, I have analyzed more than 100,000 papers since 2014 and found apparent image duplication in 4,800 and similar evidence of error, cheating or other ethical problems in an additional 1,700. I've reported 2,500 of these to their journals' editors and — after learning the hard way that journals often do not respond to these cases — posted many of those papers along with 3,500 more to PubPeer, a website where scientific literature is discussed
Publish or perish (Score:2)
Back in the seventies, I read, about 70% of professors were tenured. Now it's under 30%, and indentured servants, er, grad student T/As are *finally* beginning to unionize because they're underpaid... and have zero guarantee of a job next year.
But you have to publish or perish. Never mind if you're a wonderful teacher, and your students rate you that way regularly (and not because you're an 'easy grader').
Meanwhile, in the US at least, the coaches get millions. And tuition keeps going up. And the products o
Re: (Score:3)
Re: (Score:2)
No, this is bullshit capitalist rewriting of reality.
People need food and shelter, which in the modern world is about holding down a job and getting paid. If you're not independently wealthy you can't afford to put too much time into hobbies otherwise you start to fall short on the first two. Even if your hobby is spotting scientific fraud.
It's not greed it's survival. If it were greed then the person in question would already be doing something more remunerative.
Re: (Score:2)
Funny you mention capitalism
You did.
it's the single most successful economic system when measured by the number of people it lifts out of poverty.
Holy non sequitur, Batman!
I realize this isn't a popular stance. Doesn't make it any less true.
It also doesn't make it more true.
Re: (Score:3)
Yep, we can't trust science. I'm off to trust Alex Jones instead. I mean sure science has had just astonishing success over the last few hundred years but the presence of even one flawed human nearby is enough to put me off. Besides I like how angry Mr. Jones is about gay frogs.
Re: (Score:2)
Did you read the part of the summary where it says that she has been finding the same image being used to describe different phenomena. That is not a problem of copying between OS's or programs. That's fraud.