Live data from Hacker News

Nobel Prize winner Gregg Semenza tallies tenth retraction (2023)

retractionwatch.com

51–60 of 84 posts

Re: Nobel Prize winner Gregg Semenza tallies tenth retraction (2023)

#51
post #26

Earlier quoted context omitted.

And William Shockley with eugenics

Eugenics are "merely" deeply political incorrect/blasphemic/unethical, but not a wrong crackpot theory per se. Obviously what works for plants and animals would work for humans too.

"Eugenics" is not simply a synonym for the selective breeding of humans, but for a certain type of selective breeding - one that depends very much on who gets to decide what "good" means in the Greek prefix "eu-".

For example, people do not use the term "eugenics" to describe the unintentional selective breeding of humans who maintain the childhood ability to produce lactase and thus can digest milk as an adult.

The term is also tainted by a long history of proponents and advocates wanting to apply the principles of selective breeding to non-inheritable traits, like "criminals", due to a naive understanding of genetics (eg, https://en.wikipedia.org/wiki/Eugenics#Objections_to_scienti...).

The phrase "what works for plants and animals" seems to be hiding a lot under the covers, like the number of gene lines killed off because they weren't profitable enough for the people doing the breeding. Even in domesticated plants there is an ongoing debate about decline and extinction of heirloom/landrace varieties, and consequential loss of genetic diversity, in the face of "what works" for industrialized agriculture in short-term/decade time-scales.

The "new eugenics" movement emphasizes how choice should be left to the parents, as an alternative to the history of the state deciding these matters. However, if we really are looking at "what works for plants and animals", we can see that optimizing for an individual doesn't always result in what benefits the group.

For example, you can breed an aggressive chicken, more willing to attack other chickens and take their food. The aggressive chicken will be bigger and heavier than the others, but if they are all that aggressive then they end up fighting each other, to the overall detriment of all.

Some of these options will result in a Prisoner's Dilemma, where if some parents choose trait X for their children than those children benefit (as defined by both the parents and children), but if all parents choose X then things becomes worse (again, as defined by both the parents and children.)

Which leads us back to how politics and ethics are inherently connected to the "eu-"/"good" in the deliberate selective breeding of humans, and not simply "merely" connected.

Re: Nobel Prize winner Gregg Semenza tallies tenth retraction (2023)

#52
Keeping track of retractions is good, but I'm on the fence as to whether the community should be making it a shameful act (like committing a crime), or a positive one (like apologising, and admitting mistakes).

We want scientists to have incentives to not deliberately commit fraud, but also feel that they can rectify mistakes.

Having retraction be shameful might do the former, but will harm the latter.

I don't have an answer off the top of my head, this is just my observation on this.

Re: Nobel Prize winner Gregg Semenza tallies tenth retraction (2023)

#53
post #13

How do you accidentally have 3 images that are exactly the same?

With multiple people working on a paper, some junior, some senior, tight deadlines, and last minute changes, honest mistakes are certainly possible.

Paper figures are usually hand assembled using Adobe Illustrator or similar, which can be error prone.

For instance, placeholder images might have been added during drafting, and they were not later replaced.

Re: Nobel Prize winner Gregg Semenza tallies tenth retraction (2023)

#54

Earlier quoted context omitted.

Einstein's discovery of relativity is something everybody takes for granted now, but think about how exceptionally ridiculous the notion that light travels at the same rate of speed from all perspectives is; or how the universe will bend over backwards, including having people 'advance through time' at different rates of speed or literally physically change the distance between things, in order to maintain this perce…

Einstein did not discover that the speed of light was fixed. This was a consequence of Maxwell's research on electromagnetism in the late 19th C. It is entirely reasonable for any physicist (indeed, Maxwell himself could've done it with enough time), to work through the consequences of a non-relativistic equation for the speed of light. This approach is routine in physics, and there's nothing in it that would appear…

Special relativity was not surprising. Physics overall was converging toward it.

General relativity, on the other hand, was definitely ahead of its time.

Re: Nobel Prize winner Gregg Semenza tallies tenth retraction (2023)

#55
post #53
post #13

How do you accidentally have 3 images that are exactly the same?

With multiple people working on a paper, some junior, some senior, tight deadlines, and last minute changes, honest mistakes are certainly possible. Paper figures are usually hand assembled using Adobe Illustrator or similar, which can be error prone. For instance, placeholder images might have been added during drafting, and they were not later replaced.

So my assumption is the figures must be vital data filled figures. If they're just diagramatic ones (fig 1 type) that are similar amongst articles from the same group on thr same line of work then it's probably plagiarism by the letter of the law but not the spirit of the law.

Re: Nobel Prize winner Gregg Semenza tallies tenth retraction (2023)

#56
post #52

Keeping track of retractions is good, but I'm on the fence as to whether the community should be making it a shameful act (like committing a crime), or a positive one (like apologising, and admitting mistakes). We want scientists to have incentives to not deliberately commit fraud, but also feel that they can rectify mistakes. Having retraction be shameful might do the former, but will harm the latter. I don't have a…

Honorable retractions, as a result of honest errors, happen all the time in science and, while unfortunate, are not a huge deal.

Much different are the retractions based on research fraud. retractionwatch provides a valuable service in tracking all types of retractions, and in warning the community about unethical or predatory journals.

Re: Nobel Prize winner Gregg Semenza tallies tenth retraction (2023)

#57
post #8

Worth remembering that Linus Pauling went off the deep end on Vitamin C. Not that he faked stuff, but that even very smart people get sucked into wierd science sometimes. So leaping from this, to Semanza is emperor's-new-clothes is pushing too far. If he's moved into a space where marginal behaviour around research is normalised thats really bad but it doesn't have to mean his Nobel wasn't warranted, that depends on…

Einstein's discovery of relativity is something everybody takes for granted now, but think about how exceptionally ridiculous the notion that light travels at the same rate of speed from all perspectives is; or how the universe will bend over backwards, including having people 'advance through time' at different rates of speed or literally physically change the distance between things, in order to maintain this perce…

> We always have, and probably always will have, an arrogance of the present. It's explained by a tautology - what we think is right, is what we think is right.

I think it's culture + semantics/semiotics. I see it occurring most frequently in the form "What is true, is what we believe (~'consensus') is true" (the word belief is usually not explicit though).

In practice, the word "is" has multiple meanings, one of them being "[It is my opinion that] X is true", and pointing out the presence of the implicit opinion part is culturally unacceptable, depending on the scenario. It's a lack of cultural distinction between the universe and reality, they are typically conflated in this era, in our geographical region. Personally, I think LLM's offer some compelling insight into the process underlying this phenomenon, and I bet this can be proven out.

Re: Nobel Prize winner Gregg Semenza tallies tenth retraction (2023)

#58
there are a lot of "bad science" problems and people tend to get them mixed up.

Here's a list of all the things I can think of, roughly ordered by my opinion of seriousness:

1. Fraudulent data in real publications in real journals, even sometimes the most prestigious journals like Science, Nature, Cell. (cases in article are this situation.) Worst of all is when it has a real impact on policy, or leads to bad investment in a new technology that can't work.

2. P-hacking, questionable research practices, etc. without fraud, by "legitimate" academics. This happens everywhere but seems to be a bigger problem in social sciences especially branches of psychology. Real-world impact has been less severe but does cause problems and wasted money sometimes. The worst impact is probably that it causes some young researchers to waste their careers on dead-end lines of research.

3. Really badly done studies to legitimate dubious supplements. Subcase of 2. I actually rate these as less severe, because the average consumer of dubious supplements is not reading these pointless studies, and supplements are basically unregulated, so I don't think getting rid of these studies would make much difference.

4. Completely fake garbage papers in paper mill journals. Not just fake data but outright total nonsense. These are ignored by most scientists, so less severe, but are a concern for the people who fund scientists, and they can cause real trouble if they end up in a meta-analysis.

I only write this list because so much of the discussion of fraud and bad research seems to conflate all of them together. But they're all distinct.

I had mentally accepted p-hacking and paper mill garbage, but seeing Nobel-prize-winning labs publish photoshopped blots in top journals has shaken my faith in our scientific institutions far more.

Re: Nobel Prize winner Gregg Semenza tallies tenth retraction (2023)

#59

Remember the Sokal Hoax and Sokal Squared? Turns out that STEM also has all those problems. For profit universities pushing academics to "publish or perish" are the real enemies here, for all disciplines.

This is true but a critical difference is that STEM is a lot more falsifiable.

Postmodern deconstruction lends itself to a million interpretations, but biological testing either works or doesn't, and inconclusive generally means it doesn't.

Re: Nobel Prize winner Gregg Semenza tallies tenth retraction (2023)

#60
post #15
post #8

Worth remembering that Linus Pauling went off the deep end on Vitamin C. Not that he faked stuff, but that even very smart people get sucked into wierd science sometimes. So leaping from this, to Semanza is emperor's-new-clothes is pushing too far. If he's moved into a space where marginal behaviour around research is normalised thats really bad but it doesn't have to mean his Nobel wasn't warranted, that depends on…

It's a pretty common thing, too [1]. [1]: https://en.wikipedia.org/wiki/Nobel_disease

Feels like a case of Berkson's bias[1] to me.

[1]: https://en.wikipedia.org/wiki/Berkson%27s_paradox

Post reply on HN