[HN Gopher] Superconductivity scandal: the story of deception in...
___________________________________________________________________
Superconductivity scandal: the story of deception in a physics lab
Author : rntn
Score : 93 points
Date : 2024-03-08 15:46 UTC (7 hours ago)
(HTM) web link (www.nature.com)
(TXT) w3m dump (www.nature.com)
| wnevets wrote:
| IIRC people assumed it was a mistake when the claimed turned out
| to be false I didn't realize he was a career conman.
| dotnet00 wrote:
| In case others are coming into this thinking about LK-99, the
| article isn't about that.
|
| I had figured the LK-99 team might actually have something
| interesting considering that they were promising a proper
| presentation of their work at the American Physical Society. I
| had heard it had been withdrawn and wrote it off as another case
| of academic fraud, but looking it up right now, apparently it's
| weirder than that.
|
| Someone pretended to be the corresponding author and submitted a
| withdrawal, but this was fixed and the conference the
| presentation is at should be happening around these days.
| DarmokJalad1701 wrote:
| It already happened a few days ago (March 4th). They showed
| more results, but nothing ground-breaking. There is another
| group that is also replicating the results and publishing
| results (SCT Lab or something). Their paper is currently
| pending.
| adastra22 wrote:
| LK-99 is interesting because it really is like another example
| of cold fusion.
|
| And I don't mean what you probably infer from that. If you look
| into the way that hydrogen adsorbs into a platinum lattice, and
| how phonon resonance modes can induce tremendous crushing
| forces on anything between two platinum atoms, cold fusion
| actually looks quite plausible. It's not like perpetual motion
| --there's a clear possible mechanism for it which is
| surprisingly fully compatible with the laws of physics as we
| understand them. That said, Pons and Fleischmann rushed
| publication on shoddy data, making a mockery of themselves and
| causing cold fusion to become an untouchable subject no one
| would go near a generation later due to perceived guilt by
| association.
|
| The theory behind LK-99 is pretty solid. The linear structure
| of these apatite compounds create chains of conductive electron
| bands that in theory might have sufficient band gaps to enable
| superconductivity. If you could get atomic precision in the
| copper/lead alternating pattern that creates those stresses.
| However I fear any research into room-temperature linear
| superconductors will now be as taboo as cold fusion.
| worik wrote:
| > That said, Pons and Fleischmann rushed publication on
| shoddy data, making a mockery of themselves and causing cold
| fusion to become an untouchable subject no one would go near
| a generation later
|
| Yes, they were shoddy
|
| But no, the field was swamped within weeks with people trying
| to replicate it. It has lost its luster, and become a parody,
| but that is because it was wrong
|
| It took a decade for the excitement to die down
| 1024core wrote:
| "Publish or perish" takes yet another victim.
| Simulacra wrote:
| It could be worse than that. How did he get into the program in
| the first place should be examined.
| fifnir wrote:
| How amazing is it to read all this on nature itself, the journal
| whose peer-review method failed again and again.
| semi-extrinsic wrote:
| Fun fact: Nature used to publish papers without any peer review
| whatsoever right up to the mid-60s. The selection was based on
| what the editors thought was interesting or provocative.
|
| Legend has it that the submissions tracking system consisted of
| a particularly wide windowsill, in which the submitted
| manuscripts were stacked up in piles for each month.
| pfisherman wrote:
| Nature and Science are kind like the tabloids of scientific
| publishing. Legitimate news mixed with unsubstantiated gossip;
| all with a high "sexiness" factor that draws attention and
| moves copy.
| dsign wrote:
| Exactly. It's bad motivations all the way:
|
| - Nature published a paper twice after the reviewers voted
| "no". Of course, it is not a voting process, the editor makes
| their decision. But evidently the editor didn't use the
| referee's input for that decision. That input, must I say, is
| free work that academics do for the journal. Click-baitiness
| simply weighted more to the editor and her supervisors. People
| should just stop paying attention to what Nature publishes.
|
| - The students... I think this is the worst part of the story.
| Sometimes, students are on a student visa that depends on them
| holding for dear life to their university position. Been there,
| done that. I remember when I did my PhD: four to five
| international students--very well prepared I must say--to one
| from the host country. The ones from the host country were the
| ones with the best soft skills but they rarely put the long
| hours. I'll never forget that Iranian student who was a
| political refugee passing as a PhD student, because she
| couldn't bear the indignity of an asylum process. She couldn't
| return to her home country either.
|
| - The replication...For the first retraction, I could find
| nowhere in the article that replication attempts were made. If
| those replication attempts were impossible to produce with the
| data published in the article itself, then its not a scientific
| paper, but a promotional brochure. In fact, some promotional
| materials are more extensive than Nature's articles; they
| really want to make the thing very short and academics are
| forced to jump through hoops to condense their publications,
| until they are practically impossible to understand, not to
| mention replicate. Again, the only solution to that problem is
| to ignore Nature.
| araes wrote:
| The people writing the Nature article did not seem to notice
| the issue. The third part especially was what was occurring
| to me while reading.
|
| The LK-99 paper, lots of attempts at replicating the results.
| Crazy quick. Like front-running the published results.
|
| These results, like people never even had the thought to try,
| and it devolved into arguing sampling issues rather than just
| trying the material.
|
| There is some certain sub-segment of humanity that often
| feels that way. Like 8B humans will all bend over backward to
| let them get away with almost anything. Totally abandon all
| normal procedures, safety checks, laws, replication,
| validation, ...
| londons_explore wrote:
| Nature should simply require the main article be short and
| snappy, but then put all the details needed for a replication
| in an online-only appendix.
| autoexec wrote:
| For all the problems that Nature (and science in general) has,
| and while I'm sure this incident will be used by some as
| "evidence" that we shouldn't trust in science, ultimately this
| incident was another win for science. It should have happened
| much much sooner, but in the end, the truth was found and the
| record corrected. That's exactly the result we want.
|
| I appreciate that Nature's news team is willing to publish
| information about Nature's massively embarrassing failure to do
| their job reviewing the paper, and I hope they mean it when
| they say "We are looking at this case carefully to see what
| lessons can be learnt for the future."
| dudeinjapan wrote:
| It's a "win" for science in the sense that the truth was
| finally revealed.
|
| It's "loss" in the sense that our truth determining apparatus
| (e.g. peer review) appears to be highly unreliable, and a
| massive, unquantified amount of bullshit research has already
| passed it and is actively being used as a foundation for
| subsequent research.
| dekhn wrote:
| Realistically, we can't reduce false positives to
| absolutely zero without giving up on some True Positives
| (sensitivity/specificity tradeoff).
|
| Nature intentionally plays loose with potentially world-
| changing discoveries. It probably makes sense to have some
| journals that publish with fairly high rates of (post-hoc
| determined) false positives, simply to move the fields
| forward more quickly.
| worik wrote:
| > and while I'm sure this incident will be used by some as
| "evidence" that we shouldn't trust in science
|
| How else can it be interpreted?
|
| Most fraud uncovered has been careless. So the fastidious
| fraudsters are flying high?
|
| That would be the logical conclusion
|
| Science needs to win its credibility back, not claim this is
| "...[a] win for science."
|
| Science needs to be realistic about its decaying social
| license
| perihelions wrote:
| - _" 203 K (70 degC)"_
|
| 203 K is -70 degC, not +70 degC
| evilduck wrote:
| Maybe the author was an accountant?
| wyldfire wrote:
| Wouldn't they put ((70 degC)) in that case?
| dekhn wrote:
| no, they write it in red
| levocardia wrote:
| >Students say that Dias gave them an ultimatum: remove their
| names, or let him send the draft. Despite their worries, the
| students say they had no choice but to acquiesce. "I just
| remember being very intimidated," one student says. The student
| says they regret not speaking up more to Dias. "But it's scary at
| the time. What if I do and he makes the rest of my life
| miserable?"
|
| Indeed it _is_ scary, because as a student your PI _can_ in fact
| ruin your life and ruin your career, without much recourse in
| most cases. This behind-the-scenes is a good insight into the bad
| incentives and lack of accountability for PIs in academic
| research. I suspect this kind of thing happens all the time in
| smaller fields, and never gets discovered because it 's not a
| headline-grabbing topic like room-temp superconductivity.
|
| I had a bioengineering PhD friend who said she spotted obviously
| fraudulent papers all the time in her field--publications
| claiming they grew such-and-such bacteria in so-and-so medium,
| when she knew damn well that said bacteria _cannot_ grow in said
| medium.
| jcranmer wrote:
| Yep, when I was reading this article, as I kept hearing about
| what Dias was doing to his students, I was getting ready to
| throttle him.
|
| The students here didn't really do anything wrong, but Dias
| used his position of authority to essentially blackmail the
| students into becoming complicit in fraud, and attempted to
| justify it essentially with an "everybody does it" excuse.
| Kudos to them for taking the initiative to retract their paper
| over the head of their PI! That takes a lot of courage, given
| the immense powers an advisor holds over their students.
| strangattractor wrote:
| > Soon after the CSH paper was published, Jorge Hirsch, a
| condensed-matter theorist at the University of California, San
| Diego, began pressing Dias to release the raw magnetic-
| susceptibility data, which were not included in the paper. More
| than a year later, Dias and Salamat finally made the raw data
| public.
|
| Explain to me why Nature would publish a paper without having
| access to/requiring the data. Could it be Nature wanted to be
| the first to publish such a ground breaking discovery?
| Dishonest people are simply a fact of life. The entire point -
| supposedly - of peer review is to insure that scientist do not
| make unsubstantiated claims. Job well done Nature. Job well
| done.
| tornato7 wrote:
| I would like it noted that I had Jorge Hirsch as a professor
| many years ago, and his grading was _extremely_ harsh - you
| needed to be flawless at showing your work, not just getting
| the right answer. I remember getting a 14% on one of his
| tests, the worst score I ever got on any test in my life. It
| turned out to be a B+ after the curve was applied.
|
| Hence, it doesn't surprise me at all to see that he was the
| one to call out Dias. Some things never change!
| 0cf8612b2e1e wrote:
| Peer review is not designed to detect fraud. It is meant to
| improve the quality of work, by giving a publication a fresh
| set of eyes who might identify gaps in the original research.
| strangattractor wrote:
| Not having data is quite a significant gap.
|
| BTW: I am not criticizing the actual reviewer of the paper.
| Data availability should be a policy of the journal and
| required for publication. How can you improve quality
| without access to the data? In this instance requiring data
| would have likely prevented the paper from even reaching
| peer review and wasting a busy individuals time.
| 0cf8612b2e1e wrote:
| That was one particular dataset being requested. There
| are near infinite ways to characterize a material, and
| even the best lab is only going to have the time,
| resources, and expertise to collect and present some of
| them.
|
| I am not in the field, so I cannot comment on the
| specific relevance of the request, but in my area, there
| are absolutely some mandatory data inclusions which
| should block a paper from being published. Presumably
| this was not on that list.
| strangattractor wrote:
| There are numerous ways to embargo data and make it
| available to reviewers before publication and to the
| public after publication. [1] This isn't a new idea. They
| simply did not do it.
|
| Nature is a "premier journal" known for publishing
| cutting edge research. Any data - in this case a bunch of
| numbers - used to support the conclusion should be made
| available. I am sure there are disciplines where the
| evidence would be difficult to provide directly -
| fossils, physical items etc. Maybe this PI did the
| experiment himself - thought he won a Noble - only to
| find he could not in fact reproduce the results after
| publication - oops. IMO Nature did not do its job. Just
| pointing that out.
|
| [1] https://plos.org/resource/how-to-store-and-manage-
| your-data/
| BeetleB wrote:
| > Not having data is quite a significant gap.
|
| Requiring (raw) data is relatively new, and is only in a
| few fields. For most of the history of peer review data
| was not required. In fact, none of the PIs I worked in
| during grad school would ever agree to releasing the raw
| data. I'm pretty sure they're still not in most of
| physics.
|
| IMO, it's reasonable to require it, but understand that
| this is a relatively new requirement and not part of peer
| review culture.
| worik wrote:
| > Requiring (raw) data is relatively new, and is only in
| a few fields.
|
| That on its own is scandalous
|
| I was reading of the "reproducibility crisis", in
| computing, in the nineties
|
| Science still wants to be the warm friendly club, whilst
| playing competitively for high personal stakes.
|
| Cannot be both. Fraud, deliberately cynical fraud and
| hopeful naive fraud, has been rampant for decades
|
| The journals benefit directly and seem to be utterly
| complicit. The system is dreadfully broken
|
| Nobody should wonder why people do not trust science
| dekhn wrote:
| Nature very specifically wants to be the venue where the most
| important and impactful discoveries are first reported, and
| they tolerate a very high false positive rate to achieve it.
| Quick, light peer review is part of that process.
|
| We always joked that if you read it in Nature, it was certain
| to be wrong, but realistically, it meant that the topic was
| likely an important one with a small number of competitive
| groups who could all evaluate each other's papers and will
| loudly complain about every inconsistency or irreproducible
| detail.
| araes wrote:
| Rather common question on Academia StackExchange for exactly
| the scary reasons mentioned. Almost every time, the rational is
| something close to "my professor will ruin my life if I don't
| do what they say."
|
| My own research is creepy and I want to leave (except my
| professor will ruin me):
| https://academia.stackexchange.com/questions/67897/i-dont-wa...
|
| My professor is using us to rig the IEEE elections (except
| afraid of losing scholarship and grad courses):
| https://academia.stackexchange.com/questions/69486/what-shou...
|
| My professor is rigging nanotechnology data and plagiarizing
| (except I don't stand a chance against him):
| https://academia.stackexchange.com/questions/31265/my-profes...
| paulddraper wrote:
| That's unfortunate.
|
| There are bad professors in academia just like there are bad
| bosses in industry.
|
| Hopefully anyone with bad ones will find better ones.
| halfdan wrote:
| The big problem in academia very often is that you cannot
| simply take your research and finish your PhD elsewhere.
| With jobs you mostly can switch without losing much.
| Obviously depends on the situation.
| paulddraper wrote:
| > With jobs you mostly can switch without losing much
|
| Seniority, bonuses, equity vesting, resume damage
| BeetleB wrote:
| > Seniority, bonuses, equity vesting, resume damage
|
| Relatively minor. Most senior people I know who leave
| with bad blood end up with equally senior roles, and
| often "fail upwards" to even more senior roles in another
| company.
|
| Bonuses/Vesting: Very common to get good vesting at a new
| company to make up for what you're losing. Of the people
| I know who've moved, it's the exception that didn't get
| decent replacement for vesting.
|
| Resume damage: What? Person spent 4 years at a company
| and left. What resume damage? Your experience still
| counts.
|
| Contrast with a PhD who leaves in the middle of a
| program: The years he spent are 0. Generally, the
| experience/knowledge accumulated is not factored into it
| when he applies elsewhere for a PhD - unless he has some
| fantastic publication.
|
| Still, this is the _good_ outcome - I 've often seen
| students leave an advisor or school and succeed
| elsewhere. The real danger is when you get your PhD and
| the advisor will not write letters of recommendation.
| Unless you've made some really special connections, it
| means your academic career is over. You won't get a post
| doc. You won't get an academic position. You won't get
| into a national lab. And depending on your field, you
| won't get a research position in industry either.
|
| In the industry, no boss has that power.
| bsder wrote:
| Yeah, that doesn't work.
|
| Once you are in a PhD program you are locked in for 5+
| years. It's almost as bad as the green card timeline.
| worik wrote:
| Those are terrifying
|
| What a mess.
|
| After the "metrics" revolution we started to measure and
| incentivise everything. It has not worked well for science.
| stared wrote:
| People in the industry think that doctoral advisors are
| something like bosses in companies. They aren't. A Ph.D. is an
| all-or-nothing experience. When leaving on bad terms, you not
| only lose a few years of your life, but it is likely to ruin
| your career. With very competitive hires, an anti-
| recommendation might be a kiss of death.
|
| Sometimes, this power balance pushes people over the edge.
| There is a case of a PhD student who, after a struggle with his
| PI wanting him to publish fraudulent results, had a breakdown
| and took his own life. https://huixiangvoice.medium.com/the-
| hidden-story-behind-the...
| cma wrote:
| For foreign PhD students their kids can be ripped out of
| school and they and their family deported if they can't get
| another advisor within 30-60 days depending on the visa.
| BeetleB wrote:
| Nope. A student's ability to stay in the US (along with his
| dependents) is by remaining a student at the university
| (i.e. registering for a full time load each semester).
| They're not dependent on whether they have
| funding/assistantship, or whether they have an advisor.
| rwmj wrote:
| Surely with something as concrete as a substance, you accept the
| paper to give it a priority date, and then you wait on
| publication until an independent lab has verified the claimed
| properties? It either is or it isn't a room-temperature
| superconductor.
|
| Of all the problems in science, this seems like one of the easier
| ones to fix.
| dotnet00 wrote:
| Who's going to pay for the independent replication, especially
| if the result hasn't been published? Peer review is done for
| free because it's relatively less effort compared to a
| replication, and even then it's only barely tolerable.
| rwmj wrote:
| There's a big dispute about who pays the cost of publishing
| papers these days, so throw it in that bucket.
| dotnet00 wrote:
| It costs a lot less to publish a paper compared to
| replicating a study.
| rwmj wrote:
| I didn't say replicate the study, I said test if the
| material is a superconductor, which isn't nearly as
| difficult.
| dotnet00 wrote:
| That is replicating the study when the study is about the
| superconducting properties of a material.
|
| I think the LK-99 thing made it pretty clear that it is
| pretty difficult to do, there are lots of variables and
| lots of things that need to be ruled out to prove that
| something is superconducting.
| rwmj wrote:
| What are you talking about? Dias claimed to have measured
| the properties of the material he had made. The material
| exists, who needs to replicate anything? Send the
| material that was made to a lab where they'll probe it
| with a multimeter and wave a magnet at it. (I bet the
| reviewers of the paper could do this actually.)
| fabian2k wrote:
| Showing your co-authors the manuscript for the first time a few
| hours before you submit it to Nature is not how it's done, not at
| all. That is really unusual and a giant red flag, though the PhD
| students were probably not experienced enough to judge just how
| weird this was. And I could imagine that in that field the PI
| could convince them e.g. by stating they wanted to avoid getting
| scooped or something like that.
| bluenose69 wrote:
| Also, an analysis shows that Dias' thesis was largely (21
| percent) plagiarized [1, 2].
|
| 1. https://www.science.org/content/article/plagiarism-
| allegatio...
|
| 2.
| https://www.science.org/do/10.1126/science.adi2603/full/dias...
| lykahb wrote:
| What is the point of fabricating data here? An extraordinary
| claim about room-temperature superconductivity is going to
| receive a lot of scrutiny that it wouldn't stand. It's not an
| average publication made to pad up the H index and meet the dept
| requirements.
| WalterBright wrote:
| How did anyone think the deception wouldn't be promptly found
| out? Too many people would rush to replicate it.
___________________________________________________________________
(page generated 2024-03-08 23:00 UTC)