Hacker News Reader

Comment by sashank_1509 | Hacker News Reader

This makes me extremely bullish about Varda Space, they have engineers of high quality and expertise that they could do a decent replication very quickly (if they got access to raw materials earlier, they might have been first to do so, comparable with the Chinese Groups) while working on their off time (from what I’ve seen of their twitter) and not a single big academic lab has managed to complete any sort of replication to date!

View on HN

The first pre-print came out on 22 July (if I am not mistaken). Not enough time has passed pure and simple.

If I was an academic working on this, I would be following the Sagan Standard and triple checking my results before claiming anything.

I think we are all suffering from 24 hour news cycles. The work in China and now Varda is very exciting but they are warp speed developments.

This is simply what post-peer-review looks like.

Instead of a few anonymous peer reviews, you get the entire planet to replicate. Moment you get something meaningful, publish it on arxiv. Quantity, not quality is king. Anomalies are drowned out by large samples. Questionable methods get mass debated in public.

Scientists are dependent on public funding, why squander the biggest moment in pop science for literal decades?

Now this won't work for 99.99% of research. But top tier research will probably get published on youtube and arxiv in the future...

> Questionable methods get mass debated in public.

By people who don't know what they're talking about.

You're seriously underestimating the level of expertise that it takes to professionally assess most research. It wouldn't be unusual for there to be a small handful of people with that ability in the world.

What matters is the discovery of new things that are true, that we can do engineering with. Publication by youtube & arxiv doesn't have to be oriented at public consumption - it works just as well when it's by-researchers, for-researchers. Except without gatekeeping, people like McCalip can participate too. Who cares who's sitting on the sidelines? In post-peer-review, industry-funded academics rush a paper to a preprint server, a space engineer replicates it, and someone else figures out how to mass-produce it, and we get room-temp ambient-pressure superconductors, and we know we did because the engineering works.

This works just as well with less exciting tech, because the point is that we collaboratively learned how to build something.

Agreed. What I don't understand is the gatekeeping. Why would the actions of outsiders have any bearing on "real" research? Are they worried that grant money will get redirected toward shallower, more glamorous groups?

I think the gatekeeping is relevant because almost every time in recent years that the general completely unvetted, often actively ignorant public has gotten an in on the scientific/R&D process, it has been a disaster for the field.

NASA can't take risks and seriously innovate anymore because even intentional explosions will be passed off as failures, we all experienced the sheer madness of the covid years, similarly anything nuclear brings complete randos out of the woodworks claiming absurd things about its dangers.

Not having some amount of gatekeeping just lowers the bar too much and introduces so much more noise.

> Not having some amount of gatekeeping just lowers the bar too much and introduces so much more noise.

I can see how this would be a very controversial subject. Discouraging scientific participation? That's a tough pill to swallow without a lot of justification.

It isn't discouraging participation, it's requiring that you meet some minimum level of proficiency to be considered credible. Kind of like how you need to have some experience in programming to have any credibility when talking about how a programming language should be designed.

> it's requiring that you meet some minimum level of proficiency to be considered credible

That's not really the case though is it? Doubting someone's credibility would simply mean ignoring their results. But withholding information is something else entirely.

Maybe I'm misunderstanding, where did withholding information come into this?

> Maybe I'm misunderstanding, where did withholding information come into this?

When I mentioned gatekeeping I was specifically referring to how there's this formula that was kept secret since 1999. At some point the potential significance surely became apparent, and one of the only speculations I can come up with to justify that would be gatekeeping; keeping the information out of the hands of the filthy commoners for reasons.

Maybe it wasn't gatekeeping. If not I'd sure as heck like to know what it was.

In 1999, I was still a student in solid state physics, not far past my bachelor's degree, and happened to notice the publication of something that, to this day, I suspect might be (note, load bearing "might") a critical breakthrough for desalination. Nothing came of it. Nobody has published any follow-up for it. Nobody has done any research that would contradict it either. It's just sitting out there, unnoticed, decades later.

Was it withheld from the public? Eh... yes? Kind of? Because it was only ever published in a journal - probably Materials Letters, but this was a long time ago - that you would only have access to if you were a graduate (or ambitious undergraduate) student with plenty of time to read in a university library, or a specialist in the very narrow field that journal was for, somebody with a membership included subscription...

This isn't gatekeeping, this is, to be blunt, profit motivated hoarding by a handful of publishers in the academic journals business. And this is the thing that arXiv.org and other pre-print venues are actually having an impact on. I, for one, am strongly in favor of academic information being widely available. I'm not as much in favor of people who have not demonstrated their competencies in something that requires a great deal of both knowledge and skill feeling entitled to being given a platform for their (usually laughably clueless) proposals. If you've done enough work to have a tested hypothesis, that's different.

At least post the citation for the desal thing?

Ah, then I completely misunderstood you. Yeah I'm not in favor of outright withholding information.

However, what I think happened here is that they weren't able to work on it for a while, and if say, they didn't feel confident in their findings back in 1999 and only got around to being able to look at it again in 2017, it would make sense that their intent wasn't to gatekeep, it had just become one of those "we'll get around to testing it one day" type of projects everyone has. In this case it just potentially was a pretty massive thing to put off.

They didn't keep it, they simply did not work on it and only came back to it 20 years later (somebody posted a bit of history on the other thread currently on the front page, I'm on mobile so it takes to long to find it atm). They didn't know about it's superconducting properties either.

So no withholding information at all. But on a more general level if everyone would be posting out every little bit of information they produce during their research instead of publishing articles it would be a disaster. The reason we do science is because we are trying to understand things just posting your random results would just give everyone information overload, because you would know what to focus on. Lk99 is actually a great example, the material would have completely been ignored by most (maybe even has been, not sure if they published about it previously?) if not for these recent results. Moreover because the results were published before the authors were confident about what they had, they have been subject to lots of ridicule and accusations of fraud. It might turn out alright in this case, but something like this can be career ending if it turns out to just be an experimental error.

> They didn't keep it, they simply did not work on it and only came back to it 20 years later (somebody posted a bit of history on the other thread currently on the front page, I'm on mobile so it takes to long to find it atm). They didn't know about it's superconducting properties either.

I really do hope it's something innocent like this.

> if everyone would be posting out every little bit of information they produce during their research instead of publishing articles it would be a disaster.

I'm not sure it would be any worse than it already is. I don't see how we could achieve any more bombastic headlines than we already get.

My personal opinion/reaction likely isn't going to affect anyone's grant status. I'm just a random asshole. But there are less random assholes whose opinion might count. If someone sounding authoritative, but is just a random asshole like me, shits on this work on social media it might affect a group's grant status if it taints the opinion of someone holding the purse strings.

So I don't think asking about credentials or asking about references is gatekeeping. It's putting a statement into context. If I as a random asshole say "this isn't a superconductor" you as a reader should want to know where I'm getting that opinion from rather than just taking my statement at face value.

I'm not suggesting we give random people credibility: I'm suggesting that when we make potentially-enormous breakthroughs that maybe we shouldn't keep that information to ourselves for decades.

Edit: According to this it was just an unfortunate 19-year hiatus. (Yikes.) https://twitter.com/8teAPi/status/1684385895565365248

> You're seriously underestimating the level of expertise that it takes to professionally assess most research. It wouldn't be unusual for there to be a small handful of people with that ability in the world.

Does it mean that grant money are distributed by people who are unable to "assess most research"?

Depends on the level. At the level of congressional appropriation then certainly. At the program manager level then hopefully the pm knows.

No, because funding agencies use external reviewers who are experts in the field of an application.

They say you don’t really understand something until you can teach it. I wonder to what extent “can I teach this to the general public, and do they care?” is an indicator of both understanding new results and also whether or not they really matter.

Of course, we’re often working on a tool to help make tools for other scientists, so that sort of thing is hard to explain to the general public, but ultimately it should terminate in some results that actually have obvious value to society in general.

> > Questionable methods get mass debated in public.

> By people who don't know what they're talking about.

I'm glad I'm not the only one who finds all these public mass debaters disgusting.

This kind of public interest and demonstrations is unique only to RTS and few other phenomena

- ordinary ppl more or less understand the importance of RTS

- a visible and understandable effect (Meissner) can be shown on successful replication

- the materials required are easier to understand, but the synthesis procedure is tricky, and purity is important

- The skills of chemists comes into play, leading to kind of a Breaking Bad feel

> top tier research will probably get published on youtube and arxiv in the future...

"Top tier science" is already published on arXiv.

The future happened back in 1995 or something like that.

> Quantity, not quality is king.

No, quality is still king. We don't need videos of flakes standing on their end we need someone who has been working with superconductors all their lives who is competent enough to measure zero resistance across the sample. If one of the wild theories is right we need someone who actually knows how to prove it and not just sound like they're intelligent by talking about 1-d superconductors.

Not all meaningful science is this level of groundbreaking for this model to work. Formal peer review is still very much necessary.

>This is simply what post-peer-review looks like.

No, this is a new face for peer review. Existing peer review processes have stagnated in the information age, possibly with good reason. There are a lot of advantages to more immediate review of papers and data, and also to bypassing the systems of qualifications presently used to choose who is a valid reviewer.

There is a new claim or two of room-temperature superconductors published on Arxiv every year: https://nitter.net/peternemes/status/1684483863999827973#m... and yet this is the only one that sets the Twitters afire.

The reaction I've seen from most of the scientists has been pretty negative on the papers. The originals aren't strong papers. The chemistry has issues: https://nitter.net/Robert_Palgrave/status/168545015680691814... (unbalanced equations, starting materials have pretty different element ratios to final products). As for the main claim, note that LK-99 seems to be a poorer conductor than copper wire at room temperature: https://nitter.net/bedoya_pinto/status/1686848392616423425#m... which is the kind of thing that should really cause a record scratch and make everyone go "um, I think you did something wrong there"--it's something that absolutely would be caught in peer review.

Your link sort of hints at the reason, I think. There's enough posted to make a serious run at replicating it, and it seems to be simple enough that you can do it in a kitchen. At the same time, it's not so easy that people have cleanly demonstrated it -- the constant rollercoaster of "we're so back" and "it's so over" I think drives some of the hype.

It also seems like it has at least pretty strong diamagnetism, which while not a demonstration of superconductivity is both somewhat interesting and gives a vivid demonstration that's amenable to social media.

I’m not sure I understand your point. Are you trying to say that people shouldn’t be paying attention to this? I would personally be horrified if I found out that the paper on the synthesis of a room temperature superconductor went through the normal peer review process instead of letting the world know they discovered something huge.

My point is why should this unreviewed claim of ambient condition superconductor be exalted as the holy grail and all of the other unreviewed claims of ambient condition superconductor languish in obscurity when they come out? It's definitely not for the strength of evidence!

It's pretty much due to the strength of evidence.

There is no conclusive proof that this is a superconductor on the replications, but the evidence has been on a different level from nearly everything else you'll find on Arxiv since the beginning.

AIUI, things that float in a magnetic field at ~room temperature are moderately common claims, and this is really the only evidence that LK-99 might be a room-temperature superconductor. The published ancillary evidence in the original paper actually raises red flags that isn't a superconductor--the "0 resistivity" domain in their temperature-dependent resistivity plot is (when you realize the units they're using) actually slightly less resistive than regular copper wire, which suggests that the sharp drop in resistivity may be nothing more than a pedestrian phase change in the material.

> things that float in a magnetic field at ~room temperature are moderately common claims

At monopolar magnetic fields weak enough to come from a permanent magnet is a lot rarer.

Almost all superconductor claims come from measuring magnetic resonance or direct resistance measurements. Both are noisy things to measure, and won't get you reliable results.

Only if you’re doing sexy science.

The peer review system is broken but I think we still need some sort of structured system in addition to this “hey world, check this out!” method.

> This is simply what post-peer-review looks like.

Peer review was never a substitute for replication in the first place. In this sense we've been "post-peer-review" the entire time.

> Quantity, not quality is king. Anomalies are drowned out by large samples.

Yeah, that's not how it works.

At practice, anomalies are hyped into everybody's attention, and large samples are drowned out by lack of novelty. You don't need questionable methods to get bad results, you just need quantity.

All the things that are happening are great, and this may be an astonishing discovery. But don't put more certainty on it than it's warranted.

> Instead of a few anonymous peer reviews, you get the entire planet to replicate.

Notably, peer reviews typically don't involve reviewers trying to replicate the results. So we've essentially skipped peer review (at least temporarily) and went full speed ahead with the post-peer-review process.

Speed is correlated with execution quality. All else being equal, A lab which is better equipped - with better access to materials, and motivated and competent researchers will deliver before the equivalent lab which is deficient in any of the above.

A twitter post isn't peer review, but it is a nice way of sharing part of the process. A video etc. will help others in their attempts to replicate.

> If I was an academic working on this, I would be following the Sagan Standard and triple checking my results before claiming anything.

Counterpoint: It's better that a lot of people know "Hey, we might be onto something interesting here" early on.

It's really disappointing that no other US lab has engaged with the public in the same manner. Regardless of the outcome, I think seeing things get built and experimented with in mostly real-time is a massive source of inspiration.

The National Labs have to be a lot more careful with putting out potentially incorrect results due to their notoriety, being public funded and the public's lack of respect for the trial and error involved in the scientific process.

So they'll quietly slog away trying and retrying experiments, checking and rechecking their data before finally saying anything.

As we saw with other previous big things, like the first gravitational wave detection, they took some time to make sure that they were right.

I think that if it turns out to be real and national lab experiments prove that out, we'll be hearing rumors circulating within the field some time before an official announcement, just like how there were rumors about LIGO's detection circulating for a month before the announcement.

I don’t know what’s going on here, but I can see some guys are trying to make a LK-99 bubble in here. politically, not scientifically. We should be aware that there is still no critical evidence of room-temp superconductors.

All I see from US academics on Twitter are debates about this and that. Borderline annoying. This is the first result from mighty US capitalist machine meanwhile 3 Chinese labs have published pre prints and Russian replicated it in her kitchen.

This, its utterly shameful. Hot takes here, hot takes there, skepticism here, tell people to calm down there.

Yet not a single replication! China has gone through multiple failed and successful replications so far. And the first US replication is from an amateur.

If LK-99 works out, expect some congressmen to use this as a bludgeoning tool against the state of US public science.

Nanjing University physics professor Wen Haihu said in an interview with The Paper that the news was extremely likely to be false and they only assigned a student to it because many groups internationally were working on replications https://www.thepaper.cn/newsDetail_forward_24024641

"A researcher from the Institute of Physics at the Chinese Academy of Sciences (CAS) declined to comment on LK-99, telling the Post that even taking the issue seriously would be ridiculous." https://www.scmp.com/news/china/science/article/3229778/coul... (It's unclear to me whether this was before or after the HUST team published their video.)

I don't think it's so much "China good at science, everyone else shamefully bad," but that some institutions (including in China) are more risk-averse, because labs who're already working on something they expect to succeed are less likely to benefit from dropping it to try something unproven.

Not go all conspiracy, but isn’t that also pretty much the playbook you’d run if you were trying to discredit it, to get as much of a jump on the rest of the world as possible? The Chinese government is not exactly known for playing fair.

That requires a government so competent they immediately notice the significance of a claimed breakthrough by a foreign lab and coordinate an information campaign where their leading scientists act all skeptical just like leading scientists in other countries, denying rumours that the Academy of Sciences is working on a replication etc., ... but completely fail to act on similar rumours sparked at the same time by screenshots of a HUST professor getting all excited about his university's replication work in a group chat.

Usually, when the Chinese government is confronted with an unexpected development they want to suppress, the opposite happens: official spokespeople clam up, nobody has a statement ready, related keywords get censored more or less effectively on social media, until an official narrative is revealed that doesn't withstand further scrutiny but survives anyway because such scrutiny is censored as well.

So no, it's pretty much not the playbook they'd run if they were trying to discredit it.

Individual lab heads might do this sua sponte. Think: while everyone else is busy playing replication games you can work on reliably making LK-99 and beat everyone else to the punch on the more valuable patents and engineering glory.

The other thing is that even if LK-99 is finally conclusively shown to be a room temp & ambient pressure SC, there will be lots of important challenges to work on surrounding LK-99, so not being the first to replicate is no tragedy:

  - challenges in practical and reliable
    manufacturing of LK-99
  - challenges in engineering w/ LK-99
  - challenges in finding better materials
    than LK-99 using similar approaches

If I were in a position where I could devote funding and researcher and lab time into LK-99 research, I might well quietly let others to do the replication work while having my team quietly research practical manufacturing of LK-99 so that we could a) beat everyone else to the punch on practical and reliable manufacturing, b) get the patents for (a).

To be fair, no one in the us has a grant to replicate something that came out this week. Its also the academic summer. A good amount of PIs are probably nowhere near their labs on vacation currently.

> To be fair, no one in the us has a grant to replicate something that came out this week

And yet researchers in other countries managed to find the money this week. That would seem to point to a real relative weakness that's worth thinking about.

Whether it is the result of gatekeeping, bureaucracy, academic squabbling, or something else, a cranky (and hilarious) Russian did better, faster work in her kitchen. Even putting aside professional scientific dysfunction, that seems to indicate a strategic cranky kitchen chemist gap.

Seems kind of stupid to do chemistry in your kitchen. What exactly is the rush here to come out with results right now?!

> What exactly is the rush here to come out with results right now?!

Well, the work is happening right now.

You seem to want to make it sound like impatient kids being unreasonable or something, which is a ridiculous read.

It is more a matter institutions becoming irrelevant because they literally can't keep up with their peers. Which brings me back to,

> Seems kind of stupid to do chemistry in your kitchen.

When the institutions in your country become incompetent, that's what you're left with.

Hmm I'm not convinced that the establishment is sitting it out. Isn't it more likely that they are working the conservative process of registering experimental intent and planning only then to be followed by the experiment and only then to be followed by press release? If a bunch of amateurs can spook the old guys into violating well established procedure then id be pretty concerned about their competence.

It's the same in Korea where the research originated.

A bunch of professors belonging to a hastily formed academic committee are trying to monopolize the public debate, quibbling about errors in the arxiv paper and demanding that Lee & Kim turn over samples of LK-99 because the big-name professors are obviously too busy to make their own. It seems that their first priority is to avoid a repeat of the Hwang scandal than to touch any novel research.

Any time the material exhibits a properties inconsistent with superconductors, the to-go explanation is that it's supposedly extremely difficult to manufacture. Not to mention, the original instructions include stuff like "The sealed tube containing the mixed powder was heated in a furnace at 925°C for 5-20 hours." which sounds a tad bit imprecise.

Don't you think in those conditions it's at least a little bit reasonable to demand samples?

>Don't you think in those conditions it's at least a little bit reasonable to demand samples?

No. Because ultimately they don't owe anyone anything outside of the standard scientific process. They want to get the peer reviewed article published, and have stated that if people still want samples after that they will provide them.

Half of twitter shitposting about FLOAT THE ROCK and a bunch of curious scientists and engineers attempting to replicate this doesn't fundamentally change this equation. If it's a superconductor now, it'll be a superconductor in 6 months.

None of the media hype or unwashed masses treating this as science entertainment should significantly impact the process they are going through. Assuming they've got an RTAPS, they'll get the Nobel Prize regardless of whether or not they satisfy everyone's curiosity right this moment or a year from now. Assuming they don't have one, well, that'll get found out too.

One would think so, if the rest of the world weren't already cooking up similarly half-levitating samples in their kitchens and garages.

This is very ambiguous. So, everyone can make samples in their kitchens, but unfortunately, the researchers don’t have enough samples to provide, right? Researchers have their own primary responsibility to prove it, not the public. And this was not a debatable issue before the LK-99 bomb.

They’ve said they will provide samples after peer review. Which seems fair doesn’t it? How can they complete their paper and peer review without the samples? And if the samples get damaged they will be regret ever handing them over. If you were in the same situation you’d make the same decision.

I and everybody else would love them to hand over the samples for external testing, but it’s their samples so they can do whatever they want with it, and the rest of the world just has to pound lanarkite while we wait.

But I heard that in the interview, Dr. Kim said they have various samples that have different magnetization properties. It doesn’t make sense that they have only a few samples, ironically. I’m sorry, but this makes me think that they are trying to move the public some other way.

Judging from the Korean-language interviews making the rounds today, there seems to be some sort of political tension between the researchers and the so-called verification committee. There's disagreement about whether anyone actually demanded samples, or whether it was just a polite request...

I have a hunch that this might have something to do with the late professor Chair (Choi?) who came up with the idea back in the 90s, and whose last wish his disciples claim to be following. Chair was a sort of outcast in the superconductor research scene, and the methodology his disciples have been using are also, um, unorthodox at best. Someone might stand to lose a lot of credibility if the verdict goes one way or another. We know there's been tension even inside the team, as one of the three men named in the first paper was dropped from the second paper. At this stage, I'd be more concerned about these internal tensions and personal grudges than any attempt to manipulate the public at large.

Now I understand what's going on in South Korea. Okay, this is more political than I thought. But the problem is, the researchers even didn't send the sample abroad to prove it. Even though they had many requests for a week. This doesn't make sense, and that kind of background story is not helpful in solving this.

It would not be a normal part of the process for people to be demanding samples just because there was a preprint posted on arxiv. It's a weird situation and they don't owe anyone any of these samples.

They say they want to share samples after they get a peer reviewed paper published. That's not a weird position to be in. If it wasn't for the hype around this, they wouldn't be in this position.

Let the process play out like it normally does, and if none of the replication attempts pan out, they can provide the samples after the peer review. We'll still care about RTAPS even if it takes another year for a sample to be provided.

This case is something different. Because everyone reviewed their Arxiv papers and found that there’s no meaningful figure for the resistivity. And even magnetization data was not matched between two Arxiv papers. Now, here is where we are. After these reviews, they never updated their data to solve this issue, but they decided to let other guys do their experiment instead of them. The easiest way to solve this is to plot the resistivity data again on a logarithmic scale. And this takes less than a minute because they have raw data. This is a very uncommon situation in the scientific community.

Both papers were significantly rushed for reasons unrelated to the science aspect of it.

Wait for the peer reviewed paper if you are this concerned about it.

I assume they have more than one sample, don't you? Hopefully, if they had only one, they'd say that explicitly on one of their papers.

> gone through multiple failed and successful replications so far

The "successful" replications are partial. No one knows how to make a large enough sample of LK-99 that can be tested to definitely show superconductivity (or not) at room temp and ambient pressure.

IMO the most bullish research done yet is the theoretical research.

> If LK-99 works out, expect some congressmen to use this as a bludgeoning tool against the state of US public science.

I hope for the same here in Germany. It's way past time we actually deal with the sorry state of science funding and academia in general across the Western world.

Yeah and it's not just the US but also France/UK/Germany/Canada, all the western countries usually held in high regards for science, the lack of enthusiasm (for lack of a better word) is rather disappointing. I don't think I've seen a single university lab in these countries announced "hey we're trying to replicate this as well".

They don't have the funds to do so any more. Everything is just project based grants, which often enough aren't even enough to run a project to completion. And professors have to chase grants and do other bullshit paperwork in an awful lot of their time.

Given the international reach of the US' War on Drugs, is red phosphorus also controlled in other nations? Being denied access to the precursors needed makes replication far harder.

They couldn't even source Phosphorus!

Yes, but our ESG scores are impeccable.

Naturalize the Russian, problem solved.

Seeing her tweets, I doubt she wants to exercise this option.

Yes, I laughed when I saw that comment. I believe she would be denied entry to the US based on the questions that screen out communists, although because the particular question mentions a party and I don't believe she's a CPRF member it's possible it wouldn't apply. Regardless, not a good fit on either end of the relationship.

I'm more amused by 'grab anyone we want' stance, along with 'everyone wants to be the US citizen'. Imagine some Chinese would said that about some US scientist.

> not a single big academic lab has managed to complete any sort of replication to date!

They are still busy getting approvals from three layers of management and one group of lawyers.

Doesn’t ring true for me at all, academia is many things but if your supervisor is onboard, you can generally do anything.

Who can do anything? Things cost money and money is usually already earmarked for other things. The labs that can engage with this have both some sort of a slush fund or another account appropriate to charge these fishing expeditions to, and also a post doc who somehow has plenty of time on their hands to drop everything they are doing and validate a dubious result.

Okay still stuck in peer review then

Nope you can just put a preprint out. Like it or not, academic labs are probably busy doing things like checking their data and doing multiple replicates varying conditions. You know, science.

Yes, exactly. And getting approvals from three layers of management. You know, science.

I'm pushing back against academia because I think the system is pretty much a mess. Academics believe that we got all the nice technological innovation because of academia. I think it is despite academia. Most money going there is just wasted on writing not what you think is right but what you think will be published, waiting for peer review, and other bureaucracy. The main problem is that academia works with the belief that science needs to be rigorous since bad science might damage their reputation. However, this is how governments think, but not how scientists should think! Would the world be a better place if any new website, company, or YouTuber should be reviewed by peers first? No. Just like science, these are strong link problems [1].

[1] https://www.experimental-history.com/p/science-is-a-strong-l...

I’m in an academic lab right now. I have never once had to get approval from three levels of management to do any experiment, nor have I ever heard of anyone having to do that. I’m not even sure who would be three levels of “management” above me.

Indeed. I've worked at two academic labs and one startup. The talent density at the startup is higher, but the academic labs were far scrappier and resourceful with limited resources. Venture capital makes people fat and lazy. They forget what it's like to operate on fumes.

I'm sorry but your facts don't comport with my preconceived notions so you are wrong prima facie.

For journal publish, yes but not for doing the actual work. If you find something interesting, you can go fast.

Don’t forget the DEA.

The reason the Varda team even took this long is because they couldn’t get red phosphorus.

You just have to be registered / approved with the DEA — most labs are and can freely order phosphorus and almost other chemicals with no further burden. Seems like Varda didn’t have any prior use for restricted chemicals so they hadn’t registered yet, a not uncommon situation that will be quickly resolved for their next attempts I’m sure.

Indeed, DEA-listed chemicals don't even require a DEA license, it's just a bit of paperwork for new buyers, so the DEA can later come and ask uncomfortable questions if they suspect your usage is not legitimate.

Currently out of office, will respond as soon as possible after Impostor Syndrome therapy session is over.

There is a large degree of luck involved. The original team, who have been working on this for significant time, have only attained a 1/10 success rate (one of the better samples came from a test tube being dropped on the floor part-way through annealing). So it's part talent, part luck.

World first replication was done by Iris Alexandra using her own accelerated process and there is no way Varda Space could have beaten her.

I don't understand why people are attaching any weight to Alexandra's claims. AFAIK, it's a pseudonymous account that has provided no evidence that couldn't be easily faked. You might object that the evidence provided by the non-pseudonymous scientists could also be faked, but the key distinction is that they're attaching real, professional identities to their claims. There are reputational consequences to making things up. They have skin in the game. A pseudonymous account doesn't. Why should we believe them?

Not quite pseudonymous - she says her name is Iris Alexandria and that she's affiliated with the IGB in Russia and is now collaborating with the MEPI in producing more and measuring samples and that this is the reason why she doesn't publish any data as it's not hers and is bound for a publication, and that some of the researchers there are familiar with alternative superconducting theories which LK99's late PI was basing off on.

Of course it could all be faked, but she is prima facie attaching real, professional identities to her claims.

> I don't understand why people are attaching any weight to Alexandra's claims.

Confirmation bias is a hell of a drug.

A worthless replication. Grainy photos of supposed samples floating. When asked for videos she just insulted the people asking for videos to no end and to date has not provided any. No measurements provided either.

It wasn't even a replication by their own admission!

https://nitter.net/iris_IGB/status/1685774956330635264#m

They won't publish a preprint on their version because they think it would be "unfair"!

https://nitter.net/iris_IGB/status/1687020011548770304#m

Every post this person makes throws up more red flags. I have no idea why people believe any of their claims besides "haha anime girl"

I think you're misinterpreting the second tweet, I read it as "I can't publish a preprint myself since I'm not the only one working on it now, and my team doesn't feel our work is ready for publication". Of course it can still be bullshit but it makes sense that a biochemist doesn't have the means to test this material and that a credible physics institute wouldn't want to rush publication if they aren't sure of their results.

I believe she said it wasn't a replication because she had changed the formula.

She seemed to have problems with the efficiency of the original formula, which is why I believe she doesn't believe it's replication what she produced may in the end not be LK-99.

It seems like she's in the process of getting X-ray diffraction analysis (XRD) done.

There has been several fake videos placed out there on the Internet.

There is a quote by jay-z I'll paraphrase

" Men lie

Women lie

But replicable science don't lie. "

The mixture of USSR tweets has patriotic Americans so ruffled that some can't see through their disbelief and rage.

>I believe she said it wasn't a replication because she had changed the formula. She seemed to have problems with the efficiency of the original formula, which is why I believe she doesn't believe it's replication what she produced may in the end not be LK-99.

Yes, it isn't replication, and the only proof that her modified substance does anything interesting is a photo of a single grain suspended inside a small tube. She hasn't written anything clarifying the process from her vague twitter thread, she hasn't posted a video showing the sample's behavior (which would be better than nothing), she hasn't shown any measurements, and when people ask for these things, she gives unconvincing excuses.

And yet every thread is full of people saying she was the first to replicate. The USSR tweets are a little eyerolling but not the main issue

_biggest eyeroll ever_

That person’s twitter feed reads like the howlings and scribblings of someone in a manic depressive cycle. I’m a little surprised _anyone_ would take them seriously.

Do we actually know who that is though? I thought it was pseudoanonymous. It’s a new twitter account at least.

It is pseudonymous, although the handle iris_IGB suggests she works at IGB, Institute of Gene Biology, Russian Academy of Sciences.

https://genebiology.ru/

I have to assume that charlatans have long since figured out how to select social media names that insinuate a relationship to a well known organization, when in fact no such relationship exists.

Her rock actually floats. Did it all in the kitchen too. She should rightly be credited as first one.

More than that, she calculated how high it will float, and how height will depend on temperature, before starting the experiment, and confirmed all her predictions.

Where did she do this? I haven't found anything in her twitter feed

Wow, that's hugely impressive! Where can I find more info about this? Is it all on twitter?

She has been alluding to old Soviet scientists having a theoretical foundation for the type of conductivity shown. She posted a link to this comment that explains some of the background: https://news.ycombinator.com/item?id=36996337

Check out her twitter for some real fun rants @iris_IGB

They use a magnet for movement. “When the magnet is placed parallel and flat to the bottom of the beaker, the sample stands at a 90deg angle to the field.”

>> World first replication was done by Iris Alexandra using her own accelerated process and there is no way Varda Space could have beaten her.

World first public replication

As much respect for her as I have & as much I’ve enjoyed watching this pan out, a U.S. DoE affiliated lab or other nation states’ equivalent likely did something in the same timeframe.

People working in those labs are unfathomably smart & talented.

Do you have any evidence or are you just speculating?

I don't doubt that the people working in these labs are very smart and talented, perfectly capable of their own replication attempt. But would they rush to try replicating it immediately after the paper's publication?

The word "bullish" and fundamental science should stay as far away from each other as possible.

I was impressed by this being an extra time endeavor.

It does show genuine scientific curiousity from Andrew and his team(they might just be along for the ride).

If more company bonding events could be live streamed/tweeted experimental replication attempts of novel compounds instead of happy hours.

The world would be a stranger place.

100% true and actually a very interesting statement. Can you imagine if he had tried to present this idea to the c-suite as a potential marketing or PR coup?!

Do we really care if they confirm it in a week or two ? Are we on a hurry ? I don't know anything bout supraconductors but given the number of failing attempts to repicalte (and the number succeeding), I would not be suprised if it'd take quite some time before we can be certain. Moreover, we'll need a consensus among all the people who understand the subject. So, a week or even a month won't make any difference.

We're not talking about a vaccine.

My point was not actually about the ultimate result of replication. This guy undoubtably performed an indirect marketing coup for his company by engaging competently on this tangentially related topic and then following through.

In my experience though, if he had made any effort to link those two things prior to doing this - e.g. by proposing this as an internally sponsored marketing stunt - it would probably have been rejected.

Eh, I have met the CEO of Varda. They are very much a startup and have that startup energy. One of the founders of Varda has been very vocally supportive of this on Twitter as well.

Of course, this is all Varda IP as they’re in California and anything a head of research performs that generates IP is owned by the employer.

Is that a law in California? My understanding of the labor law is that California protects employees, so that if they sign a contract assigning IP to the company (which is almost a sure bet today, but not a law afaik- please correct me otherwise) they are actually protected such that it only applies if:

- using company resources, such as time, hardware, or office space, unless otherwise stated.

Clearly the case here, but my understanding of this law doesn’t align with how you described it.

As engineers we are given building blocks that we can build things from. Iron, steel, copper, plastic. Every once in a while a new building block arrives. It takes a while for us to figure out how to use it. But when we do, the world changes.

- batteries + touch screens => iPhone

- batteries got better => Tesla

- reinforced concrete => pretty much all the worlds buildings and bridges

- neodymium magnets + gyros => drones

A room temperature superconductor... well, that changes things on a completely new level. Order of magnitude smaller losses plus order of magnitude higher magnetic fields? OMFG can we build SciFi things from this if it pans out.

So yeah - we are in a hurry. If this pans out, there will be a massive shift in how pretty much anything is made. And the reward for the people who can figure it out first are insane, in terms of potential value created.

We're in a hurry in the sense that we don't want to leave investigation of this on the back burner for a few months, but we're not in a hurry in the sense that it's going to make any difference in practice whether a lab group reports today that they've reproduced it, or takes a bit more time to be careful and reports more thoroughly in two weeks time. Any first mover advantages are likely to accrue to the groups who come up with a reliable reproduction method over the next couple of months, not to the ones who send out the first press release and twitter post this week.

The point is not whether it takes one week or two weeks.

But eastern competition is fierce, and the western world is mired in esg and bureaucracy.

Something is rotten at a general level, and if its not fixed soon, this trend of lagging behind will also start manifesting where it starts to hurt.

Ever felt schedule pressure / been on the critical path? This is that concept applied globally. The hurry is that delays in initial research measurably delays potential rollout. I'd like to see this material meaningfully used in my lifetime for its potential global society wide benefits / ease of human suffering / margin on total ecological collapse. The pace set now affects the mean and standard deviation of probability of that happening on the timeline.

From a scientific perspective, you are probably correct.

But think about it from a company/wealth creation perspective. Are you then really sure the correct course of action is to take it chill?

My point partly is that as outside observers we can't deduce "is taking it chill" merely because we haven't seen two press releases and three tweetstorms from any particular lab yet. They might be slogging away but not caring to report every step on the path to the outside world.

This is why Elon takes dumps on competitor companies.

He understands how to market in the social media age.

These are completely different activities... Doing some week long materials experiment has almost no relevance to production aerospace engineering with regulatory burdens and multi-year test schedules. I might as well say that any mathematician who can publish a paper quickly about boundary functions can also be a very effective SWE at Amazon working on cloud.

You forgot the Russian twitter biologist.

the difference between doers versus academics has never been demonstrated more beautifully

Oh yeah, the commercial "doers" are happy to have a go once academics have slogged away at the hard, risky, expensive, and often thankless trial-and-error before the discovery. Not to do down the people in this particular example, and I'm happy to assume they aren't claiming any credit they shouldn't, but the dichotomy really isn't as you make it out to be.

...without academics we wouldn't be here. at all. here an historical genealogy that leads to the LK99 discovery: https://hackmd.io/@lifthrasiir/lk-99-prehistory

The problem is they needed academics to show them what to do.

The story goes, some professor had a theory in 1999 that was rejected by the mainstream regarding creating superconductors. He passed away and as a dying wish asked his students to continue the research work and develop a theory for why it works. The students unable to get “academic” funding set up a company and got funding from industry to do their research in 2017 and the rest is history.

It’s very hard to redeem academia in this example.

it's kinda false: https://hackmd.io/@lifthrasiir/lk-99-prehistory people were working on it since soviet times, in academia. And those people influenced other people. Also in academia.

It's a slow moving process, but it works.

That's not entirely accurate. For my understanding, it is a private institution which employed the individuals who made the discovery. Some of whom have alternative backgrounds in academia, and some of whom have it in industry.

Private R&D has always towered above academics. People who can compete create significant innovations. People who can't compete stay as professors doing research because it's the easiest and most convenient pathway.

Maybe.

A better explanation for the slowness of the national laboratories is they their word will be taken more definitively on this issue than anyone else.

They are the ones with reputations and have to answer to critics in the private sector, amateur sector, and in Congress.

I think their slowness is an encouraging sign, it means LK99 isn't easily debunked.

Reply on news.ycombinator.com