Allegations of Scientific Misconduct Mount as Physicist Makes His Biggest Claim | Hacker News Reader

Allegations of Scientific Misconduct Mount as Physicist Makes His Biggest Claim

physics.aps.org

155 pointssohkamyung62 comments

Some people just get too big for their britches. If you're going to commit scientific fraud, you need to do it in the edges where nobody really cares, nobody's going to try to replicate it, nobody's going to try to build on it, and nobody's going to try to get you to explain more details because they want to build on it. You want to be working in the dead-end cul-de-sacs.

You do not want to be going "I've got a room temperature superconductor" or "I've detected a dark matter particle" or "I've got a simple procedure for flipping matter to anti-matter" or "I've cured all cancers". People are going to want follow-up details and to reproduce it and to begin building on it and turning it into engineering.

You gotta keep your aspirations realistic. You need to be working off in the darkness somewhere, not trying to conquer the territory where the brightest lights are shining. Don't try to fraud your way to Nobel-prize level work.

but what if you actually do have a room-temperature superconductor, but the entire community is dead set against considering even the possibility?

we already have plenty of evidence that even science people (flawed humans, like us all) cannot "do proper science" because they are swayed by irrational arguments such as distaste, shame, appeals to tradition, conformism etc. etc.

but what if you actually do have a room-temperature superconductor, but the entire community is dead set against considering even the possibility?

Well then you'll have no trouble demonstrating it and you will receive the accolades, the fame and the fortune that creating the first room-temperature superconductor will bring you.

Exactly. Even just in medicine: you’ll sell an MRI that is far cheaper than anyone else on the market, with no need for a quenching system, and with bigger margins. Etc.

> but what if you actually do have a room-temperature superconductor, but the entire community is dead set against considering even the possibility?

That's not the case, though. If you say "I've got a room temperature semiconductor" (or a reactionless drive, or an antigravity machine, or whatever) and tell people how you did it, people will _absolutely_ try to reproduce that. Not just academic institutions, either; the US Navy has a bit of a history of trying to reproduce obvious nonsense, in particular.

Who is this community that’s dead-set against this?

There’s lots and lots of academic research. And a fair bit of Department of Energy money flowing into those labs. There’s lots of interest. There’s lots of papers.

And the great thing about those who “do proper science” in materials research: you can do actual science by repeating and verifying the experiments.

This sounds like a reflexive take without any thought about what science actually is.

I would guess that’s where Dias started. His thesis probably wasn’t his first piece of academic fraud. After getting away with fraud for so long, it’s possible the fraud got bigger in the classic “I’m too smart to get caught” way.

Do something bad enough times and you will think it’s normal and no big deal.

Medicine is a great area for scientific fraud. Unless you research a super common disease you need to have the relevant patients (which means only a handful of people can replicate your study), the sample groups tend to be small, and there is a million other factors to explain away why your results may differ from another study. To the point that there is a large grey area between outright fraud and unconvincing results (small impact that may very well be noise).

Medicine is an interesting area because all of the incentives lean towards incremental, long-term subscriptions for small-but-measurable gains. So if your fraud is an improvement by 2.5% instead of 2.45%, it's both unlikely to get a lot of attention and quite likely to get incorporated into the new drug development pipeline. All of the players except for the general public seem to have reached their goals here. Even if the fraud is detected during drug production it's probably small enough not to warrant a scandal, and who wants to shut down their next great drug production pipeline for what might or might not be a fraud and even if it was doesn't amount to very much?

Clearly we need adversarial science teams. Some people not connected and perhaps by law never connected directly or indirectly to each other that try to disprove the other team’s results. No husbands and wives and relatives and kids working for the other team.

You keep what you kill. If you proof something wrong you gain its funds for research.

Or "I can run dozens of tests on just one drop of blood"?

> If you're going to commit scientific fraud, you need to do it in the edges where nobody really cares, nobody's going to try to replicate it, nobody's going to try to build on it, and nobody's going to try to get you to explain more details because they want to build on it. You want to be working in the dead-end cul-de-sacs.

But then what would even be the point to do that?

Tenured professorship

Likewise, see the phrase "grist to the mill".

https://www.collinsdictionary.com/dictionary/english/grist-t...

It reminds me of Jan Hendrik Schön, who looked like physics' next emerging all-star wunderkind until his "discoveries" of crystalline semiconductors won him so many awards and so much attention that it was discovered his entire career had been built on falsified data. A huge embarassment for Bell Labs and a black mark on the final years of more than one great physicist's careers there, since so many people thought they were helping a great new mind succeed when they were in reality perpetuating his fraud.

There's a great 3-part documentary on youtube about it, which argues Schön was probably a contender for the Nobel Prize in Physics before everything collapsed: https://www.youtube.com/watch?v=nfDoml-Db64

I was instantly reminded of the same. I can recommend the documentary, it's the first time I saw an actual example of scientific fraud at this scale.

It's a stunningly well produced, informative and entertaining video with excellent visuals. It puts a lot of expensive tv, never-mind youtube to shame.

my assumption is that scientific fraud is rarely premeditated. perpetrators are not setting out to lie or cheat. they start honestly, pick their field like anyone else would, then more like with N-rays[1], they start to believe their own false - or even valid but difficult to reproduce - results. then I suspect that, in competitive fields like Alzheimer’s or superconductors, the pressure to get results is such that the temptation for some to cheat just a little to “get the paper over the line” is too high

it’s understandable really. if you’re working with extremely unreplicable results, as they are here, and you genuinely believe in your results, then it’s only a small step. it’s still inexcusable of course, but it’s easy to see the thought process

this is not the same as what you’re describing, which is setting out to cheat from the beginning, which is almost certainly much rarer and undoubtedly harder to get away with, even in obscure cul-de-sacs

[1] - https://en.m.wikipedia.org/wiki/N-ray

I think there is also a lot of pressure from funders to be the first to provide proof of a paradigm-breaking discovery, and tout it as a breakthrough before the claims can be verified.

By chance, about a decade ago I met the lead researcher of a NASA group that in 2010 made a shocking claim of arsenic being incorporated into bacteria's DNA in place of phosphorus at Mono Lake, CA. They later had to retract the statement after others were unable to replicate. But the research I did into the follow up indicated that NASA, who employed the scientist, did not have its astrobiologists follow basic standard microbiology protocols against sample contamination, and pushed the arsenic hypothesis as a game-changing breakthrough before it had been subjected to peer review.

Nice advice for future pro cheaters :)

> While Hamlin was digging into the CSH data, he came across familiar-looking sentences in the paper that presented the raw CSH data—lines that he recollected writing in his 2007 PhD thesis. On a hunch, he pulled up Dias’ 2013 thesis and fed both his thesis and Dias’ into a plagiarism checker. His computer screen lit up; the two documents contained numerous identical passages.

Oof. When the guy investigating your paper is the same guy you plagiarized for your thesis.

My memory is a little fuzzy, but back in 2012 or so, this programming whiz applied for a job at Square on the security team (I think?). He was impressive, and had written an SSL implementation (EDIT: it was actually a certificate authority) entirely in Ruby. Breezed through the coding challenges. Just our luck that one of the interviewers was the actual author of that implementation!

Seriously, it was really just luck. It makes me wonder how many people successfully bluff their way through their careers.

Reminds me of when the MIT admissions dean was found to have faked her degrees: https://www.bostonglobe.com/magazine/2019/05/15/mit-dean-for...

Marilee Jones (that article is subscriberwalled). "...and co-author of the popular guide to the college admission process 'Less Stress, More Success: A New Approach to Guiding Your Teen Through College Admissions and Beyond' (American Academy of Pediatrics, 2006)"(!)

https://en.wikipedia.org/wiki/Marilee_Jones

Interestingly it's really easy to fake degrees because the universities are prohibited by law from sharing information on students.

Universities have a process for providing proof of degrees as long as the student authorizes it. Thing is when you're in mid-late career, no employer cares.

Did that applicant claim to be the original author of that implementation? I don't see why it would be an issue otherwise...

Yes, that's exactly what they did and why it was a problem. They represented themselves as the sole author. I want to say that they'd never even sent in a single PR, but this was over 10 years ago and I just heard the story over lunch (the real author told us all about it).

Seriously? You think that if people insinuate plagiarized stuff was theirs but don't say so, that's not a problem per se. I've seen this loads of times. You don't think insinuating is a very bad character signal?

You could have just asked: "what's wrong with SSL?" and anybody who ever wrote actual SSL / X509 / CA related code would not stop talking :)

My go-to Android interview question for years was simply, "Tell me your pain". A developer with any experience would be overflowing with angst about the platform and it was a really good way to gauge their ability.

Why would someone do something like this? He’s essentially going to become persona non grata in the scientific community. His career in science is basically over.

On top of that, according to a Reddit thread, he raised $20 million in VC funding for his “room-temperature superconductor”. Due to the deception, he’s likely going to not only lose any money, but likely face federal securities fraud (and other fraud) charges.

He’s literally looking at prison time right now.

The baffling stupidity that could cause someone engage in such idiotic (and self-destructive) conduct is perplexing, to say the least.

edit

RPI is pretty legit. Or at least, it was back when I was in college

RPI? RPI = Rensselaer Polytechnic Institute which is not mentioned in the article. The researcher is from the University of Rochester.

ah, I was responding to that mention in my parent comment and assumed it was correct, but now that comment says "edit" so presumably it was a faulty assumption

> The baffling stupidity that could cause someone engage in such idiotic (and self-destructive) conduct is perplexing, to say the least.

...and yet, here I am, refactoring unsanitized parameters.

I'm not in a research field but I feel like someone looking for and then raising $20M in funding has already taken a very large step onto that slippery slope.

If I had $20M looking over my shoulder at my lab results I would feel a certain obligation to "perform" — or certainly to appease.

At first I can imagine data is legit but maybe a bit "curated" for the client. Later some data is dropped altogether. Maybe conditions are tweaked to get more promising looking data. Results that have an error range are penciled in to one end of the range or the other to, again, show more favorable results.

Naively I assumed there was a "purity" to research that would not even start on the path to looking for cash. Grants, on the other hand, as I understand them, don't have the same assumptions of showing a profit.

> he raised $20 million in VC funding for his “room-temperature superconductor”

Did they tell the VCs the "room temperature superconductors" needs to operate under more pressure than you'd find in the center of the Earth?

Pressure that is well within humanity's capability to produce and maintain at small-ish scales (not intercontinental transmission lines, but within a factory). 100x less pressure than the previous record. If these results were real, it would absolutely be an enormous scientific leap forward and lead to new technologies, and I don't see a reason to minimize that. Of course, it's all fabricated bullshit, so I guess it's a moot point.

Yeah that's an obvious question for people who _don't_ do this kind of stuff. Similar behavior by other people, like George Santos or SBF, come to mind - I think a big part of it is Ego and probably another mental disposition like severe narcissism or something. The frustrating thing about these questions is we'll never really know the answer, even if these individuals told us why we'd be hard pressed to believe them.

The Laws of Human Stupidity (Carlo Cipolla):

1: "Always and inevitably everyone underestimates the number of stupid individuals in circulation." Do not be fooled by traits assumed to be exclusive of stupidity such as education or career.

2: "The probability that a certain person be stupid is independent of any other characteristic of that person."

3: "A stupid person is a person who causes losses to another person or to a group of persons while himself deriving no gain and even possibly incurring losses." The corollary is that the exact nature of stupidity is difficult to predict and therefore guard against. (Refer to the QZ article covering these laws for an interesting graph comparing stupid people to all other types of people.)

4: "Non-stupid people always underestimate the damaging power of stupid individuals. In particular non-stupid people constantly forget that at all times and places and under any circumstances, to deal and/or associate with stupid people always turns out to be a costly mistake."

5: A stupid person is the most dangerous type of person - more dangerous than an intelligent but selfishly-motivated actor.

I would strongly recommend reading the critique of the data originally published by these authors:

https://www.worldscientific.com/doi/abs/10.1142/S02179792237...

The fact that "measured" data fits a cubic spline so well is extremely strong evidence of misconduct. This just doesn't happen with real data. The misuse of significant figures is also implicative.

The first paper was retracted by the journal due to a lack of detail around a key measurement, which doesn't automatically mean misconduct but certainly leaves room for it. And the same journal knew about this when they accepted the new paper, so hopefully they made sure the data was looking better for this one?

Now that _would_ be the journal providing a valuable service indeed, we'll just have to wait and see if they actually did.

Discussed here yesterday:

https://news.ycombinator.com/item?id=35070643

The subtext in the article claims the room temperature superconductor exists at 'near-ambient conditions', though reading further in the article this material is a superconductor at about a gigapascal (about 9.8k atm).

I have heard of diamond anvil room temperature superconductors, it appears this is in that class.

The ones you've heard of are likely this guy's 2020-era work, which is currently in doubt.

Regarding the near-ambient thing, that's probably relative to their prior work, which was at ~2.7 million atm.

This is much lower pressure than those others. It's not that it's practical to make a useful device, but it is much more practical to do in a lab where researchers can analyze the structure and hopefully design new ones. So it might be a very useful stepping stone.

> The writer of this story—Dan Garisto—had no communication about this story with his father, Robert Garisto, the Managing Editor of [Physical Review Letters].

!!!

I'm sure that's genuinely fine, but it's a slightly awkward way to wrap up an article about scientific impropriety.

People are deadset on the fraud allegrations arguing that one must be incredibly stupid to do this. It stands to reason then, that they haven't.

I feel they are simply lazy-on-the-paperwork tinkerers. I can understand since academic publishing seems to be such an ordeal.

There is a reason why people do not use their 'nonstandard' way of subtracting noise.

Probably, I'm not familiar with the domain. I'm suggesting their data and publishing practice could be dogshit while their process might still wield a room-temp superconductor, in which case it would be a little unfair to conclude they are fraudsters.

stupidity or fraud, their results should not be trusted until successfully and independently reproduced.

If you take two independent source datasets, A and B, and combine them (add, subtract, whatever) to produce C, then C is correlated to A and B, but A and B are uncorrelated.

So in the paper, there are two datasets D and E, and a third F. The noise in D and E are uncorrelated and the noise in F is correlated [edit: to D and E]. Clearly F is the result of combining D and E. The problem is that D is labeled "raw", E is labeled "final, clean result", and F is labeled "background". Thats... pretty damning. They took raw measurements, added them to a desired result, and claimed the result was "background".

Edit: to my mind, the biggest problem here isn't that an exciting new physics discovery is reversed. It's that so many people show little regard for truth. I'm not speaking to the parent of this post so much as to entire community. In the APS article it says that Rochester

    > "'determined that there was no evidence that supported the concerns.' But the university has not made the remit of the investigations public and has not provided any rationale for the investigations or details on how they reached their conclusions."

In regards to clearly plagarized sections in his PhD thesis Rochester also said

    > “Dr. Dias has taken responsibility for these errors and is working with his thesis advisor…to amend the thesis.”

Even one of the scientists who uncovered problems with the first 'discovery':

    > “'Still, I don’t want to believe [the allegations] because it’s too serious,' Eremets says. He would prefer the field simply forget about the irreproducible CSH result and move on."

It stands to reason that it happens rarely, but it clearly does happen occasionally. Scientific fraud exists. Probably a lot of it never even gets caught. Unfortunately you can't trust the logic "if it's that stupid, no one would do it". The reason we invented the word "stupid" is that people are stupid.

People who commit fraud at this level usually have a psychosis akin to narcissism. Their brain is dysfunctional and they are literally compelled to lie bigger and bigger -- they are addicted to lying. Do not look into your own brain and think "I'd never do that, therefore they wouldn't either"; you can't extrapolate from a healthy brain to a clinically compulsive liar / narcissist. Once these things start to unravel, you find a seemingly endless stream of lies these people have woven into their lives from their earliest years. We're already beginning to see this with the 2013 thesis being plagiarized. This person's entire life is a lie. It's a psychosis.

We think that your claims are bunk AND that you are a plagiarist.

When your case offers 2 totally different accusations like that... it's weird.

Scientific misconduct -> copying and pasting data, falsifying data

Biggest Claim -> announcing (again) that he's found a high temperature low pressure super conducting material

This is basically a non story turned in to clickbait. Someone no one believes announces a world altering breakthrough.

Reply on news.ycombinator.com