Internal review finds falsified data in Stanford President’s Alzheimer research
stanforddaily.com
stanforddaily.com
Something that became really clear during COVID is the sheer extent to which universities seem to lack basic research ethics. The ivermectin debate could be summed up as "once we exclude obviously fraudulent papers, what do we have left" which is why it was so rancorous. Disease prediction as a field is pure fraud: they cannot predict epidemics, never could, aren't on track to be able to, don't use the scientific method and yet routinely make predictions with absolute confidence whilst claiming to be scientists. Microbiology is flooded with papers that are quite obviously being manufactured by professional fraudsters, logical fallacies and false statements are easy to find across many fields, every other paper falls apart the moment you poke it yet nobody cares about any of this. Universities don't care, journals don't care, peers don't care, nobody within the academic system cares.
Why don't they care? Why don't universities actually fire professors who are found to have been writing BS papers? The only viable explanation is that the people at the top know full well that this behavior is at saturation levels and trying to clean it up would destroy their institutions. How could they know that? If they got to the top by doing it themselves, that would definitely be one explanation.
Surgisphere: mass audit of papers linked to firm behind hydroxychloroquine Lancet study scandal
https://www.theguardian.com/world/2020/jun/10/surgisphere-sa...
The clorox society police is out in full force on HN.
The reasoning Daszak and his co-authors (who failed to declare conflict of interest related to GoF work) employed wouldn't pass a first-year exam.
... But no one spoke up. What use are any of them.
I agree. Fire them all and replace them with scientists that signed the Great Barrington Declaration, with priority to those that signed earliest.
Almost all of them, but the prime ones to target are the ones whose leading journals do not require open data to be published along with their paper.
Drug trials with pre-registered hypotheses are actually among the more rigorous approaches to research.
> _nobody_ cares about any of this. Universities don't care, journals don't care, peers don't care, nobody within the academic system cares.
Please be mindful about generalising over such a large number of people. Maybe those who care have little power to do anything about it.
Maybe somebody cares, but an awful lot don't.
At the same time, we (outsiders) see academics set up massive political and pressure campaigns against the public, politicians and each other - that aren't directly sanctioned by the management - and this seems to happen all the time. Academics are some of the most powerful members of society. They are one of the few groups whose collective beliefs are actively dangerous to criticize in public. So the idea that they're all powerless victims in this regard is just so hard to take seriously. If the academy really wanted to, it could hold its own staff to account just like it does over the ideology of the day.
But the most remarkable thing is the incredibly high quality of the journalism. Unbelievably thorough investigative reporting, outstanding exposition of highly technical items. From a college newspaper!
> Schrag, the Alzheimer’s researcher without knowledge of the internal review, told The Daily that “to his credit, Dr. Tessier-Lavigne authored several of the later studies which revised the findings of his 2009 paper.”
Some smart kids over there in Palo Alto. I hope they keep at it when they graduate.
It is lot more likely to be able to get to spend countless hours required than a large commercial publication for an investigative piece ?
Student journalists get paid far less if anything at all and will definitely be motivated in a story about their college and put in the extra hours than a professional journalist also easily tap into subject matter expertise right there in their college
Journalism quality is a primarily function of how motivated the journalist (and the publisher) is about the story, there is no shortage of talent in university like Stanford.
https://www.palladiummag.com/2022/06/13/stanfords-war-on-soc...
Another very high profile article related to Alzheimer’s and plaque formation was also recently retracted…weird. I’ve become very skeptical these days, of my fellow scientists, which is both a good and a bad thing.
I think that is a broad generalization. I never had to take an ethics in research course as a PhD student.
> falsifying data, plagiarism, how to ethically work in animal models ... image manipulation.
My take is that this may be a "field"-dependent class. My PhD is in a non-experimental field (math); thus, we cannot have these sorts of issues (minus plagiarism).
What often happens is that a given program is required to include an ethics portion or course, so the university has some mandatory class whose syllabus reads "this class will include an ethics portion".
Then there just never seems to be the time needed to get to that material.
At least, that's how my computer science degree went.
A lot of the class consisted of videos, the contents of which were mostly irrelevant to mathematics (dealing with things like falsification of data). The people assigned to teach it (it was someone different each week) would try to talk about what was in there, but there wasn't much to talk about because so little of it was relevant to math (mostly just plagiarism, as you say).
In an attempt to do something useful with the remaining time, most of the professors ended up turning it into something of a career-and-publishing Q&A, answering questions about various issues (though not really ethical issues) that can come up in writing and publishing papers. <shrug>
(I seem to recall one of the professors tried more seriously to actually stick to the material, but he was an exception...)
Unfortunately, I think that these sorts of classes mainly serve as either CYA or to check a box for a funder.
Most of the folks who do unethical stuff (in general) either think that what they are specifically doing is not unethical, or they just don’t care (“playing the game” is something I’ve heard quite a bit).
In certain fields, especially if you include anything that’s in a gray area of ethics, unethical behavior is more the norm than the exception.
The presentation was two slides. The title slide was presented, and then a content slide was presented. Professor's lecture on ethics was, in its entirety, "You guys know enough not to be unethical. Don't be unethical." We pretended to be learning from the second slide while someone took a picture to present to administrators, and the ethics class was dismissed. I had 55 minutes of my hour left.
Unethically completing an ethics class was one of the most ironic experiences I've ever had.
Professor is not getting grants to make research ethics slides, so he's not going to do it. The group needs a volunteer, but everyone else is already busy with assigned tasks. Would you volunteer to run the ethics training and do it properly?
Yea but I doubt they'd have me. I'm just a disruptive individual. With regards to the question of why I would make a scene in the lecture, it's because of the contradiction of agreeing to be ethical and at the same time walking out of the ethics lecture knowing it was meaningless. It's the complicity in society that leads us to being led by a bunch of fucking jellyfish.
For example, look at the claims that the Genentech internal review was routine. I don't believe it at all; I read it as 'we believe this claim will stand because you can't prove otherwise'. That is not honesty at all, but a normalized protocol for lying. I'll bet you didn't think twice about the integrity of doing that.
But my key point is that, if that claim is disproven, what is the result? Do we say Genentech and the individuals making these claims are liars, and not to be trusted? Are they shamed and shunned, their reputations damaged? No, we've normalized accepting these lies; it will just be viewed as losing a game, as the normal result of the protocol - it was disproven so the claim falls. Honesty is not implicated.
We are lied to because we accept it, we normalize it, we don't even notice it.
"You" however is "readers on HN" and they very very likely did think about the integrity.
Your point stands, of course, that the sleight of hand/speech is an attempt to pull a fast on one readers in general.
* Many follow the crowd in attacking whoever is the target, and sometimes that includes raising questions about integrity, etc.
* Many attack the integrity, etc. of everything, which isn't thinking critically - it's the opposite.
It seems to me lying is a great strategy in today's society as you rarely face harsh consequences and you about never lose more than you have already gained thanks to it.
(Covered at length in the book "Why Fish Don't Exist")
Remember Joi Ito at MIT Media Lab? You can deny and claim ignorance and shift responsibility but once multiple real witnesses start coming forward your position is untenable and you’re out. See https://news.ycombinator.com/item?id=20905905
Kudos to the Stanford Daily editorial team for following the story despite the official denials and probable warnings. Good journalism matters.
which is another way of saying... "it's not a lie if you believe it"
To give them the benefit of the doubt though, if they realized the data was wrong they should have requested a retraction. That they didn't after being requested to do so speaks a lot, to me at least.
It doesn't invalidate the entire story in any significant way, but after all these years, even after an hour or a day or two, wouldn't they all have been aware of each other's accounts and points of view? They may have explicitly reconciled them years ago - 'holy cow, what are we going to say?' Genentech may have coached them - in fact, I would almost expect it.
That's practically the canonical form of saying, "I did it, but I'm going to demur and blame it on other unspecified people."
People who test the veracity of others call this an "exclusion qualifier." [1]
[1] https://jacksonzheng.com/the-cia-uses-these-methods-to-tell-...
Did they Friday news dump (before a 3-day weekend) their own big story?
[EDIT:] Let me put my reply into some further perspective:
1) It's well known that job advertisements blatantly lie about the job description and requirements. The standard advice software engineers give to each other is to apply to jobs even though you don't meet all of the requirements, because the requirements often turn out to be optional, more of a unicorn wish list.
2) Ask yourself, which companies and which jobs specifically do lying software engineers apply to? Do those companies perhaps have a culture rewarding this behavior? And could honest software engineers be averse to applying there, instead seeking places or means of employment that are more amenable to their personal ethics?
I literally did that: "It's well known" "The standard advice software engineers give to each other". Is anonymous "kodah" well known? No. Do software engineers give each other standard advice about getting hired by kodah? No. Also, I talked about "companies" plural, not kodah's company singular.
The only one who insinuated about kodah, and about me, is you. You ought to read the guidelines yourself with respect to my comment and your own reply.
The intended implication of my previous comment was that the behavior of software engineers in our industry is driven largely by how companies are run, by their management practices, how their hiring works, how their promotions work, etc. The engineers themselves are not really in the driver's seat, they're in the passenger's seats, and the incentives for current and potential employees are often perverse.
1. I would say that our listing was accurate other than the fact that we actually worked in 4-5 major languages, not just the one advertised. That was never much of an issue though, we screened for a certain level of adaptability. To me there's a very big integrity difference between lying and applying for a role which you don't meet all the requirements. If you tell me, "I haven't worked on Kubernetes before, but I've worked on distributed systems and have messed with a lot of REST apis." I'd be understanding, no one can have all the right experience at once. Contrast that with someone having Kubernetes all over their resume only to find out they've pushed a button on an automated deploy pipeline and don't know the difference between a Deployment and a StatefulSet. The latter I consider a lie.
2. No, I know that company didn't have a culture of rewarding lies. In order to claim things for promotions you had to have multiple peers back up your claims.
What preempted the lies was probably discovering that we primarily did software development kit work, which most engineers have not done or don't have a ton of experience doing. Mix that in a bowl with the fact that we were easier to get hired at than Google and paid slightly above average (at times) it was fertile ground to attract candidates willing to espouse some aspirational stuff as if it were an accomplishment. This was especially relevant around Senior hiring when we'd ask about leading initiatives. A lot of SWEs have delivered software in a team, but few SWEs have led those initiatives, or only owned a very small part which they worked with (usually) one other engineer to deliver.
You could only screen who applied, and the job posting inaccurately described the language(s) used. How could that not have an effect on the applicant pool, and how the applicants approached the application?
> No, I know that company didn't have a culture of rewarding lies.
Banking and financial services? Hmm...
This is reaching imo. Most of the candidates never even got to a point where we were talking that deeply about the languages we worked on primarily. Second to that, candidates were allowed to pick whatever language they wanted for the interviews.
> Banking and financial services? Hmm...
This is a really odd take. We didn't do banking and we're heavily regulated. Additionally, it was a security product development team. Anyway, this line of conversation from you has been mostly unenjoyable and unhelpful.
I would imagine, because you seem to be missing the point at every level.
The larger point is that when I talk about "the biggest liars and cheats rise to the top", I'm not referring to something like your software development team, which is nowhere near the top.
Another point is that people who are nowhere near the top are mostly just reacting to the perverse incentives created by a system designed by the biggest liars and cheats at the top, a system that naturally tends to reward people like them. The less powerful have little choice but to operate within that system and deal with it as it is, flawed as it may be.
A small point is that when you talk about "the candidates", these are already the self-selected result of an inaccurate job posting, and thus what you do after you start talking to them doesn't make up for the false impression that was given before you talked to them.
> we're heavily regulated
So?
You flat out told me that the job listing was inaccurate:
> I would say that our listing was accurate other than the fact that we actually worked in 4-5 major languages, not just the one advertised.
"accurate other than" is a euphemism for inaccurate.
Anyway, you're still missing the point, which is that it's irrelevant whether your company's specific job posting is accurate, because job postings in general are inaccurate, with exaggerated requirements, causing job applicants to behave accordingly, which is to largely ignore the specific requirements listed in job postings and also exaggerate their own qualifications. And then companies scratch their heads densely and wonder why, somehow ignorant of the perverse incentives the industry has already created.
The rest of your post is just rambling. That's not why people lie, and there's a big difference between exaggeration and outright lying. You seem to have zero delineation in your head between this, which is frankly unsurprising given the interactions I've had with you here.
You: "When I was in charge of hiring I dealt with this every day. A huge swath of software engineers lie about their skills, involvement in projects, and just general accomplishments."
Also you: "Contrast that with someone having Kubernetes all over their resume only to find out they've pushed a button on an automated deploy pipeline and don't know the difference between a Deployment and a StatefulSet. The latter I consider a lie."
You later: "there's a big difference between exaggeration and outright lying. You seem to have zero delineation in your head between this"
The part of the quote you chopped off:
> If you tell me, "I haven't worked on Kubernetes before, but I've worked on distributed systems and have messed with a lot of REST apis." I'd be understanding, no one can have all the right experience at once.
If you have Kubernetes on your resume and this is what I find, that's an exaggeration. This is what would be "the former".
Anyway, I now realize reading through your comments why this conversation was a waste of my time. Have a nice day.
And the part of your own quote that you chopped off:
"To me there's a very big integrity difference between lying and applying for a role which you don't meet all the requirements."
That's not a difference between exaggeration and lying, it's a difference between honesty and lying. Where exactly is the exaggeration?
> I now realize reading through your comments that you're the kind of person that tries to vouch for flame bait and then complains about it. I've clearly wasted my time here.
Heh. Are you referring to this? https://news.ycombinator.com/item?id=34849006
The commenter posted a link: "see his rebuttal here: https://tessier-lavigne-lab.stanford.edu/news/false-allegati..."
I said: "I've vouched for this comment. Not sure why it was flagged. Regardless of whether you believe Tessier-Lavigne's denial, it's directly relevant to the story."
The link was not flame bait, it was simply the accused person's response to the accusation.
Now, in retrospect it does turn out to the case that the commenter has engaged in flamebait in other comments. I did not vouch for those other comments, nor was I even aware of them beforehand. I didn't look at the commenter's bio or other comments before I vouched for that one particular comment, which was simply a link, a very relevant link.
Thus, your criticism of me is totally unwarranted and ridiculous. You're grasping at straws here.
[EDIT:] I noticed that you've edited your comment to remove the part about vouching. Have a nice day to you too!
Right, because the person for whom the goal is to get a great job / promotion, will leap through the hoops required. Another type of person might care more about what they are working on, or their professional goals or their values and less about playing the game.
Obsessed with how they stack up compared to others... Perhaps coming from a place of insecurity.
How did you filter for the lies/bs? Did you drill them to see if they actually knew the topics at hand?
Did you make this word up? I can't seem to find it anywhere.
:)
Most Greek prefixes don't have an o. Instead, the o is used to connect the consonant at the end of the prefix to the consonant at the beginning of the root.
Where the prefix ends in a vowel, that vowel is used directly. Compare anabasis, catabasis, analyze, paralyze, metamorphosis, polygon, etc etc etc.
None of this works here, because "mendax" is not a Greek word. (Which doesn't help "mendacracy" - the epenthetic vowel in Latin is i.) The Greek word for liar is ψεύστης pseustes; you probably want pseustocracy ("rule by liars") or perhaps pseudocracy ("rule by lies", but seems more likely to be interpreted as "false rule").
It's not clear where the rest of the root of "mendax" is supposed to have gone in "mendacracy"; if you really wanted to jam it onto the -crat ending, you should end up with "mendacicrat".
/\w+(?: \w+)+ R-[ACDFGIK-PS-W][ACDEH-LNORSTVXYZ]/Then as I worked for a few years it became apparent this isn't close to being true. I ran into a huge number of people who reached the top by various forms of lying, whether it was little lies like taking credit for others' work or bigger lies like scam websites, or even bigger ones like crime. It's actually interesting how people will just tell you these things if you are even a bit friendly with them.
What tends to be lost in these stories is what you're not hearing. For every fake study, there's probably many more things going on with many more people.
There was some paper posted the last couple of days about estimating rates of corruption in financial firms. It would be interesting to apply similar methods to academics.
So wild reading that. Would be very funny if it wasn’t so deadly serious.
Is it just availability bias on my part?
A giant ball of corruption and deceit. Sometimes, a cigar is just a cigar.
There are also ethical people and companies you hear nothing about, but in a all but the smallest orginizations it takes extreme effort to not have at least some