Hacker News Reader

Comment by misja111 | Hacker News Reader

Why is it necessary for other labs to produce the material themselves to verify the claims? The original authors claim they produced some superconductive material, can't they simply offer their samples to other labs to verify?

View on HN

>Why is it _necessary_ for other labs to produce the material themselves to _verify_ the claims?

I highlighted "necessary" and "verify" because your wording of your question somewhat hides the motivations of the labs.

Multiple labs around the world are not just doing a "homework assignment" to academically verify someone else's work. Instead, the tantalizing idea of a "room-temperature superconductor" would be a miracle material and That Scientific Leap In Progress is why independent labs are racing to reproduce it.

Same excitement of so-called "replication" would happen if somebody put out a plausible "cold fusion" paper or a workable synthesis of a new "drug molecule" to cure cancer. Multiple labs around the world would be motivated to implement their own cold fusion machine and cancer drug. It isn't "necessary" that they replicate it. Instead, they're genuinely excited about making their own miracle devices/drugs and possibly extending the scientific discoveries beyond the original authors' findings.

> or a workable synthesis of a new "drug molecule" to cure cancer.

You mean like this? https://twitter.com/gigaj0ule/status/1686814611658715136

To my knowledge the original authors promised they would send out samples in the future but want to wait till they have a finalized paper approved for publication. Whether that's reasonable or not is very hard to tell.

IF they are correct and they have an RT superconductor, then it's a highly political decision on who gets sent some samples. Like, all their former colleagues and collaborators will be pissed if they don't get any. But you also would want to send some samples to someone who you know has 1) all the necessary equipment to confirm your claim, 2) has the necessary publication history to make him/her a trustworthy expert on high temperature superconductivity that others would believe, 3) would not want to slow down your progress just so he/she can gain an advantage.

And with something with such huge potential for applications as RT superconductivity, you then also get vested national interests where your national physics society or research ministry might be pissed off if you don't involve enough researchers from the country which funded your research.

Honestly, if I was in their shoes I'd be delaying me sending out my samples into the world as well.

Of course this all hinges on whether they actually have a room-temperature superconductor or not. If they don't actually have one then the delay tactic could also just be in hopes for this all to blow over without them having to prove their findings.

This is of a magnitude that finding the funding to get one of the original researchers to be custodian of a sample while they tour to some well respect other lab whose researchers then do the verification is well within the realm of the possible. That way you can have verification without losing control over where the samples end up, you could take it back when they're done with it.

Reproducibly is kinda a cornerstone of academia. This isn’t a product sneak peak.

This is about material that didn't exist before.

The existence and effect of it should suffice.

How it is produced is a later step.

Well, in the case of LK-99 the production method itself seems to be part of the puzzle, and the original authors aren't noticeably better at it than anyone else. Even their best sample doesn't fully levitate.

We need higher quality samples in order to determine whether the thing actually superconducts, because it's really hard to measure the resistance of small grains of material buried inside a mostly useless rock.

Based on your statement, the original author would have claimed something untrue since they also cannot really produce LK-99 that fully levitates.

It is certainly not a part of the puzzle here.

The puzzle is to reproduce the experiment, but you claim the original experiment itself is bogus...

What do you think is easier, making a possible superconductor, or popping it on top of a magnet and seeing if it floats? The latter is an "experiment" only in the sense of an elementary school science class.

Why are we so obsessed with finding out whether the original authors did their homework? Science is not about crediting or discrediting people, it's about collaborating and standing on one another's shoulders to find out truths about the world. If LK-99 turns out to be a superconductor, it doesn't matter if the original paper was glaringly incomplete. If LK-99 doesn't pan out, it matters even less.

An outside lab reproducing the alleged superconducting properties of a given sample of LK-99 would be a good reproduction.

That would prove that they'd done it, but not that anyone knew how (including the original authors). If this is going to be useful, we need to know how to make it.

It could be reverse-engineered.

If they can prove this one sample is a room temperature semiconductor that would surely be relevant in deciding how much time to spend on trying to create more of the same material.

It’s like the 4 minute mile tbh. If we know it is possible, we can and will find a way.

It's technically not necessary, but it's also not necessary that the authors give away their samples while they're still running tests (apparently they never even wanted to go public at this stage). The claims will be checked against very high standards, and we will know eventually. But we'll need a bit of patience, the authors haven't even submitted a paper to peer review yet. Science doesn't generally doesn't progress at the instant-gratification speed of social media.

Rarely do researchers get an opportunity to get 15 minutes of fame like right now. Anyone successfully recreating LK99 can

* Potentially do it better/purer

* Get a few news articles, and a mention in a very high traffic Wikipedia page

* Again if they do it well, get 8 figures in funding pretty easily. While during any other day they might be fighting for $50,000 here and $100,000 there

Hell, even if they can't do it better, they might simply be able to document their process better.

Partly to ensure that the whole process is well understood, and that there aren't some (unstated) aspects specific to the original lab. By way of example, when multiple labs conduct genetics experiments on mice, even the food that the mice were fed has to be standardised to ensure that it isn't driving variation between the results.

And now even the sex of the researcher because mice were reacting differently surprisingly.

Not sure that's "weird" because I have certainly seen pets who were more comfortable around women than men. Maybe from bad experiences in the past before the pet was adopted.

They did send some of the material to other labs

Source?

The Korea Institute of Energy Technology (KENTECH) received samples about a month ago and they analyzing it for possible applications. https://www.edaily.co.kr/news/read?newsId=02646966635704040&...

If this material is real. And the mechanism producing it and materials like it could be understood starting process now would give head start on discovering materials that are like it.

It is pretty big pot to get part of if you discover way to refine the process, or find adjacent material with similar or better characteristics.

Perhaps I have a cynical take, but I would guess that the labs are primarily verifying the claims to themselves to identify ways to make profitable research or production lines. They want to fund their own labs.

Publishing verified claims to help others would be secondary.

That’s a cynical tale, but once you realize that profit is a useful way to signal to technologists what the world finds useful in aggregate, it means there is more likelihood that the science gets out of the lab to the real world.

tldr; for good or bad, humanity hasn’t found an incentive mechanism better than capitalism for deploying tech at scale.

Isn't this one of those game theory scenarios? Some labs could profit by not announcing replication, but others could by announcing it as quickly as possible? Thus not all labs will behave the same way, unless they are all colluding which seems unlikely.

It’s not strictly necessary, but knowing how to reliably synthesize the material is also important.

It's not necessary and they could simply offer it to other labs. The reason they have not done so (except to one small lab in South Korea that doesn't appear to specialize in superconductivity and claims they need six months to verify) is pretty obvious by now: they don't actually have a superconductor.

Reply on news.ycombinator.com