Become a fan of Slashdot on Facebook

 


Forgot your password?
Close
typodupeerror
×
Science

Another Retraction Imminent for Controversial Physicist (nature.com) 61

Posted by msmash from the closer-look dept.
A prominent journal has decided to retract a paper by Ranga Dias, a physicist at the University of Rochester in New York who has made controversial claims about discovering room-temperature superconductors -- materials that would not require any cooling to conduct electricity with zero resistance. From a report: The forthcoming retraction, of a paper published by Physical Review Letters (PRL) in 20211, is significant because the Nature news team has learnt that it is the result of an investigation that found apparent data fabrication. PRL's decision follows allegations that Dias plagiarized substantial portions of his PhD thesis and a separate retraction of one of Dias's papers on room-temperature superconductivity by Nature last September.

After receiving an e-mail last year expressing concern about possible data fabrication in Dias's PRL paper -- a study, not about room-temperature superconductivity, but about the electrical properties of manganese disulfide (MnS2) -- the journal commissioned an investigation by four independent referees. Nature's news team has obtained documents about the investigation, including e-mails and three reports of its outcome, from sources who have asked to remain anonymous. "The findings back up the allegations of data fabrication/falsification convincingly," PRL's editors wrote in an e-mail obtained by Nature.
This discussion has been archived. No new comments can be posted.

Another Retraction Imminent for Controversial Physicist More

Another Retraction Imminent for Controversial Physicist

Comments Filter:

  • Published in 20211? (Score:4, Funny)

    by andyring ( 100627 ) on Wednesday July 26, 2023 @09:33AM (#63715864) Homepage

    So it was published almost 20,000 years in the future? I mean, by then we probably will have room temperature superconductors!

    • Nah, they are just using the Alantean Calendar.

    • Re:Published in 20211? (Score:5, Funny)

      by RitchCraft ( 6454710 ) on Wednesday July 26, 2023 @09:53AM (#63715910)

      He's simply ahead of his time.

    • Re:Published in 20211? (Score:4, Funny)

      by ArchieBunker ( 132337 ) on Wednesday July 26, 2023 @09:56AM (#63715918)

      Using the Kelvin calendar.

      • Common misconception, that's the Rankine calendar. The Kelvin calendar would have had it in the year 11228.

    • Re: (Score:3)

      by pesho ( 843750 )
      No wander he needs to retract, he messed up in which millennia he needs to publish. He is yet to run the experiments.

    • Re: (Score:2)

      by ls671 ( 1122017 )

      So it was published almost 20,000 years in the future? I mean, by then we probably will have room temperature superconductors!

      That's why he needs to retract it while he still has time to contrarily to many scientists who have said things lately, in the recent past that they will need to retract in the near future !

    • Re: (Score:2)

      by Zak3056 ( 69287 )

      Obviously, it's not a year, it's a zip code, and the summary means it was "published in Washington, D.C.". This is still a minor error, as the APS (publisher of PRL) office is located in the 20045 zip code a couple of miles away, and I put this down to slipping journalistic standards where a lazy intern just googled "DC zip code" and got 20211... the state of the world today, I tell you.

      (</humor> for the impaired).

  • Not the only super (Score:5, Informative)

    by Neuroelectronic ( 643221 ) on Wednesday July 26, 2023 @09:45AM (#63715882)

    This is not the same person as the recently published paper on room temp superconductors covered on hacker news.

    https://arxiv.org/abs/2307.120... [arxiv.org]

    • More a hoax than a scam if it's fabricated. Sure wasting people's time is still not ethical, but there seems no plan to get away with it for any significant amount of time ... certainly not enough to get into Nature.

  • Ambient pressure RTSC claim announced... (Score:5, Insightful)

    by ebrandsberg ( 75344 ) on Wednesday July 26, 2023 @09:46AM (#63715884)

    https://arxiv.org/abs/2307.12008

    Given the claims of ambient pressure and even Tc above water boiling point, it is MUCH easier to validate this claim, and they have a video of the miesner effect online as well. They provided instructions on how to create the new material, and based on this, I expect claims of reproduction (or failure to do so) within the week. The time may have come.

    • Wait, and itâ(TM)s not even him? I was thinking that this retraction would be about his last auperconductivity âoebreakthroughâ⦠though I thought that was replicated a couple of months back.

    • It's rather obvious it's MUCH easier to go with sensationalist clickbait headlines from a prominent journal first, in order to derive profit long before content is proven true or false.

      Apparently all it takes for you to be known as a "controversial" physicist, is to screw up more than once, since this retraction represents his whopping second in total. The bar for tar and feathering, is quite fucking low among what we ironically call "intelligent" people.

      Ben Franklin would have been committed to an insane

  • Goodbye, Dias, we hardly knew ye anyway.

  • Fake studies making it to publications (Score:5, Insightful)

    by SmaryJerry ( 2759091 ) on Wednesday July 26, 2023 @09:47AM (#63715894)
    This isnâ(TM)t the first time a paper was found to have issues. There was another hoax where they completely made up entire studies, not just plagiarized, and they were accepted by supposedly âoepeer reviewedâ journals. The bar is too low now and apparently no one has the time to validate anything so it just gets trusted and accepted. The fault is in the publications.

    • Re: Fake studies making it to publications (Score:5, Interesting)

      by FudRucker ( 866063 ) on Wednesday July 26, 2023 @09:51AM (#63715902)
      with AI/ChatGPT gaining in popularity and usage i expect to see more fake discoveries & research papers

    • Does anyone actually read these publications? Seems like they are mostly used as citations when writing your own paper.

      • Does anyone actually read these publications? Seems like they are mostly used as citations when writing your own paper.

        It's scary considering how your suggestion here should make us look across the entire multi-billion dollar industry of "publications" driving funding.

        I highly doubt this will be the only field we find potential manipulation in. We might even find people read publications like lawmakers "read" 2,000-page bills before they vote on them.

        • Re:Fake studies making it to publications (Score:5, Informative)

          by excelsior_gr ( 969383 ) on Wednesday July 26, 2023 @11:12AM (#63716074)
          When you do your research, you usually read the title, if that's interesting, then the abstract. It's a matter of taste (among other things) what you do after that if the abstract is intersting as well. A professor of mine at the university once told me he skipped to the conclusions section. More often than not, I find these completely worthless. I skip to the tables and the diagrams, because that's always where the intersting bits are. If a diagram is interesting, then I read the text around it for more details and clarifications. If the paper is still interesting, then you read it from the beginning. All articles I cited in my thesis were read this way, but I didn't cite any articles that I didn't read.

          • Re: (Score:3)

            by ceoyoyo ( 59147 )

            I suspect you can tell a lot about a field by the order of the sections in a paper. I was taught to write them intro, methods, results, discussion. My PhD supervisor told me the discussion was mostly a chance to speculate but sometimes had interesting stuff. Conclusions, if there were any, were pretty uniformly bullshit, and several of the journals explicitly banned any section by that name.

            Many medical journals put the methods last, in smaller print. Some of them even banish them to supplemental material.

      • Does anyone actually read these publications? Seems like they are mostly used as citations when writing your own paper.

        Think about that for a second - your premise that no one reads the papers means that there will never be one retraction ever, jut everyone accepting whatever is published.

        Science is not like pop culture or religion - experiments are made, and other scientists try to replicate them, and other scientists look through the papers.

        https://retractionwatch.com/ [retractionwatch.com] is alive and well. And it's doing good work too.

        And there lies the difference between the science minded and other people. These retractions and t

    • It's almost as if large numbers of eyes and the ability to peer review the work are a silly facade that in no way guarantees anything.

      Who'd have thought?

      • IT's a lot like how open source works. People just assume lots of people are reviewing a widely used security library's code, they decades later somebody actually does and finds a very obvious buffer overrun through the use of the C API. Looks like the only one reviewing open source code are the hackers.

      • Re: (Score:2)

        by jythie ( 914043 )
        Meh, I think the problem is that people expect flawlessness in things they don't like, and anything less is proof of uselessness. Problems happen, problems get found, problems get fixed. The whole concept of 'guarantee' is more of a thing detractors make up than people who use a system claim.

    • There's a big difference between profit driven no name publishing paper mills and well established journals of the likes of Nature, The Lancet, etc.

      Managing to get published in a top journal is highly prized & scores big career points. And you're hearing this news because the journal did its due diligence to investigate suspicions of malpractice.

      • Re: (Score:2)

        by ceoyoyo ( 59147 )

        Yeah, no. Journals don't really do any "due diligence." The editor filters things by whether they're interesting or not, then they send them off to peer reviewers. The peer reviewers might be someone who knows what they're doing, they might be that guy's student, or they might be an author who recommended themselves by another e-mail address. Good journals make some effort to get good reviewers. If you want lots and lots of work, be a scientist who's friends with a journal editor.

        Nature has a *higher* rate

      • Umm, no. NAture dropped the ball on this guy. Twice.

    • Maybe only in this particular case, given Ranga Dias history -- they should have asked him for a sample .. but Ranga Dias would probably claim he lost the only sample (he's done that before.) But in the general sense, the publication of these fabrications is NOT the fault of the publications. Do you know how much it costs to replicate some experiments? There's no way Nature can fund hundreds of thousands of dollars to replicate every study. Reviewers can only ask you some due diligence questions and accept

    • Verifying the data used is not (and has never been) the responsibility of the publication or peer reviewers. Peer review judges a paper based on the paper itself (at least it's supposed to): is the conclusion supported by the data in the papers, is the analysis properly done, is any needed data missing, are the results notable and worth publishing, etc. It cannot, does not, and is not *meant* to catch fake results (except for the most obviously fake results). Obviously the reviewers can and will reject pape

  • Cold Fusion redux (Score:3)

    by CoderFool ( 1366191 ) on Wednesday July 26, 2023 @09:58AM (#63715922)
    All of this has happened before. All of this will happen again.

  • It is common to publish parts of your phd thesis. They are now calling this self-plagiarism?

    • It is common to publish parts of your phd thesis. They are now calling this self-plagiarism?

      I came here to say this. Also, I think it is typical that one's PhD thesis is copyright by the author. (Mine is.) So, one can re-use it any way one wants (subject to academic honesty of course.)

  • Many times. One reference of many: https://slashdot.org/comments.... [slashdot.org]

  • The President of Stanford also had to step down under similar circumstances.

    When people wonder why normies don't reflexively "trust the science" anymore, this is why a lot of the more educated ones don't do that.

    People are indoctrinated that science is "self-correcting." There is a reproducibility crisis in science now. You don't have to take my word for it; just Google it and you'll find a lot of mainstream sources admitting and lamenting it. The fact is that science is doing a very crappy job of "self-cor

    • No, you can't make it a felony. That would impede science even more. For one thing, it would make people wary of publishing or science careers because it is easy to get a false result for a variety of reasons unintentionally such as instrument error. Second, it would reduce the number of collaborations and/or make them much higher overhead/micromanaged because people would be afraid of losing their freedom -- rather than an academic scandal they could walk away from.

    • Re:"Trust the science" (Score:5, Interesting)

      by alvinrod ( 889928 ) on Wednesday July 26, 2023 @11:51AM (#63716130)
      The scientific process itself is sound and trustworthy, but the same cannot always be said of the people professing to follow it. Frankly, no one result should be trusted with absolute certainty. That's only part of the process and until something can be replicated and potential alternative hypotheses rejected, anything should be treated as a work in progress, and even if widely accepted still subject to being overturned as we gain additional understanding of our world.

      If people don't understand this, it indicates that our education system has done a poor job of teaching even a surface level understanding of what science really is. I wouldn't be surprised if many students come away with the idea that science is little more than a series of facts to be remembered and regurgitated on an exam, much akin to how they might be asked to learn and recite some passages of scripture. Most facts are entirely useless, but teaching people how to discover them as necessary for themselves is beyond invaluable.

      • Re: (Score:2)

        by sfcat ( 872532 )
        You are speaking only of theory. Now let's talk about how things actually work. The way we judge professors and researchers comes from one of a couple complicated formulas that measure number of papers, citations, etc. What that formula doesn't reward is replicating other's experiments. What it doesn't punish is getting papers retracted. You aren't wrong, but you are also glossing over human nature and how the incentives of academia work these days. The reproducibility crisis is real and it didn't hap

        • Re: (Score:2)

          by tragedy ( 27079 )

          The way we judge professors and researchers comes from one of a couple complicated formulas that measure number of papers, citations, etc. What that formula doesn't reward is replicating other's experiments

          Those two sentences are contradictory. When you reproduce someone else's experiment, you write a paper about it and that paper gets cited along with the original when people write about the original work. So, in that system, the reward for reproducing results is the generation of papers that are heavily cited and, therefore, greater professional esteem.

    • Re: (Score:2)

      by ceoyoyo ( 59147 )

      It is self-correcting. You're just complaining about the time frame.

      It would be great if reproduction were more timely, but the people complaining most loudly about the "crisis" are people who've spent a bunch of time working on something based on a particular underproduced result (i.e. reproducing it), found it doesn't hold, and are pissed that someone else didn't do the reproduction for them.

      The classic example being pharma companies that spend a bunch of time trying to reproduce some drug effect, can't,

      • Re: (Score:2)

        by sfcat ( 872532 )
        A recent example is a pharma company trying to reproduce the last 10 most important papers about a specific condition and finding 80% of them to be frauds. And most of those papers were more than a decade old. Saying this is about time frame is pretty self-serving, a bit like those physicists who claim their pet hypothesis will be proved with the next bigger particle collider. I mean, if the time frame on average is a century, its a pretty moot point isn't it? If it is more than decades, its pointless a

        • Re: (Score:2)

          by ceoyoyo ( 59147 )

          They're not "frauds." They were non-reproducible. That could mean they're just wrong, they're among the 5% false positive results you expect statistically in that field, the drug company's reproduction was among the 10-20% false negatives you typically should expect, or they didn't replicate the procedure correctly.

          If someone writes a paper about some result and nobody checks it in a decade, AND nobody does anything that depends on it, it's not really a high priority, is it?

          The scientific literature is a re

  • Peer reviews (Score:3, Interesting)

    by chen2011 ( 1971334 ) on Wednesday July 26, 2023 @11:21AM (#63716086)
    I have done many such reviews. Unlike in math, a reviewer can check the logic and derivation fully. In sciences, one can't check the result of an experiment unless one conducts it, which reviewers typically don't do. What they check is whether there is enough information for someone else to repeat the experiment.

    • Reviewers (the good ones) also check to see whether the conclusions of the authors follow from the data gathered, and whether the instruments or methodology for gathering the data have flaws or weaknesses. Both can be done without reproducing the experiment.

Slashdot Top Deals

Don't tell me how hard you work. Tell me how much you get done. -- James J. Ling

Close