🧵 View Thread
🧵 Thread (209 tweets)
Think a lot about someone figuring out that birds are bad at eating seeds and ends up creating a virus that helps birds better eat seeds. Then birds die because dropping seeds is actually a feature that helps edible plants grow around safe eating spaces.
Think a lot about someone looking at captive bear poop and thinking that the undigested seeds going thru the digestion tract are a waste of calories, and then use efficient digestion as a metric to optimize gut bacteria for. Then that bacteria leaking into the wild.
Like, what if all these scientists wasting their time creating bad psychology studies has some unknown secondary effect having to do with occupying elite overproduction, funding bubbles that redirect funds job making tech startups, or providing funding for military endeavors.
A lot of times things that look like absolute wastes of time are often a form of status signaling. A large country that can afford to have entire fields of bullshit science could be acting as a sort of posturing behavior. Like a Ritualized Fighting proxy ala costly signaling.
I keep hearing weird stories of researchers getting funding to do one thing and then actually going and doing something else with the money. I think a lot of fabricated science is a symptom of this. It would be a great way to hide off-the-books research.
"What is heartening is that the vast majority of research is fundamentally honest and, although corners may be cut, the intention is to add to knowledge not deceive the scientific readership or grant providers."https://t.co/qreodGeG49 https://t.co/DERd7mHcET
Incentives in this space are asinine. Fraud and manufactured science isn't the only thing that's a problem. Most of the time it ends up just being something someone works on to get status and grants, and then they shelve it. Costly signaling to get tenure.https://t.co/GUvme9XBcU
But trying to force scientists into making viable products for industry is also not ideal. It swings in the other direction. Years of exploration over a theme is what I see happening in labs. Sometimes basic science is needed to be done, and accumulated to be built upon later.
We have weird distortions; somewhere along the way public funding of science became the only viable source of funding, and the public demands results. It mirrors all the expectations we have of colleges to produce jobs. Many focusing on Job Placement Rate to advertise themselves.
I've been thinking about this for years.https://t.co/ZEz9ZsDFgR
Something that resonated with me recently is the idea of "Coase was a mistake".Is that was I was uncovering when I wrote this thread? https://t.co/ATur9Wc8PP
It was in the context of the collapse of institutions and how that interacts with society at large. It was what inspired me to write that top tweet that seemed to resonate so well.https://t.co/56kvo4cXMx
This, but it's about how people use scientism to justify their deeply held beliefs. https://t.co/7fHYU19Tvb
The yelling will continue until the balance of epistemology improves https://t.co/b2o8G7vhtS
"That misconduct happens isn’t shocking. What is: When the FDA finds scientific fraud or misconduct, the agency doesn’t notify the public, the medical establishment, or even the scientific community that the results [...] are not to be trusted."https://t.co/5qqRE0hAQ9Wat
"the FDA has no systematic method of communicating these findings to the scientific community, leaving open the possibility that research misconduct detected by a government agency goes unremarked in the peer-reviewed literature."https://t.co/etBsH16kFDWell, damn.
"non-trivial number of key discoveries have been at some point rejected, mocked, or ignored by leading scientists and expert commissions. Sure, eventually they succeeded, this is why we know about these cases. But they could have succeeded earlier, faster"https://t.co/SdDnkfIf1G
I don't actually wanna kill any mice.https://t.co/KSCAPApctv
The WHELP heard round the world.“You can see that the idea of what is respirable, what stays airborne, and what is infectious are all being flattened into this 5-micron phenomenon,”https://t.co/p5XqKXQZk9¯\_(ツ)_/¯ https://t.co/bOzvBVd4xC
I like my ideas raw. I don't even wait for them to die.https://t.co/EF3Vjar3tU
I must feed.https://t.co/XwpAI2gonk
"a meta-analysis of prior studies on questionable research practices and misconduct published in the journal Science and Engineering Ethics reported that more than 15 percent of researchers had witnessed others who had committed at least one instance" https://t.co/Fb7YXpTXeq
"We have long known that peer review is ineffective at detecting fraud, especially if the reviewers start, as most have until now, by assuming that the research is honestly reported."https://t.co/L2R3KCN10P
"A disturbingly large portion of papers—about 2%—contain "problematic" scientific images that experts readily identified as deliberately manipulated, according to a study of 20,000 papers published in mBio in 2016 by Elisabeth Bik"https://t.co/tHZTu8tw9p
Fool Journal Once, Shame on you.Fool Journal Twice, Express Concern.https://t.co/WGwVsYrsSFI just make a note to doubt everything tied to their work and taint entire subtrees of research in my mental web.
"It will therefore not come as a major surprise that peer reviewers are also not very successful in uncovering scientific fraud." [2012]https://t.co/xKFg5P85bG https://t.co/VmC2X4rP0x
Can nudge lying on surveys by words you use? 😹"Meta-regression showed that self reports surveys, surveys using the words “falsification” or “fabrication”, and mailed surveys yielded lower percentages of misconduct." [2009]https://t.co/X5NEEE9Jzvhttps://t.co/7hIBxPyLOE
"process for correcting or withdrawing a paper tends to be alarmingly long. Late last year, for example, David Cox, the IBM director of the MIT-IBM Watson AI Lab, discovered that his name was included as an author on two papers he had never written."https://t.co/0GxCnwampK
"presents the findings and methodology from a team of investigative journalists, hackers and data scientists who delved into the parallel universe of fraudulent pseudo-academic conferences and journals; Fake science factories"https://t.co/1NJVpTtpLthttps://t.co/ZAxMHqItzA
"As the NBER researchers conclude: "[T]he idiosyncratic stances of individual scientists can do much to alter, or at least delay, the course of scientific advance.""https://t.co/aD7lkqKo00
This clip is out of it's context, but still a very powerful framing of breakthrus in science.Found a copy online.Segment starts 85m 29s, Clip is from 92m 0shttps://t.co/GqWDVqFHR0https://t.co/GUvSe7YN8A https://t.co/NUNYPozQpM
"We are going to kill ourselves because of stupidity"Evergreen quote.https://t.co/6evnXp0BiE
We're all going to die because nobody knows how anything works and the only mechanism we have to stop all this bullshit is misguided mechanisms of shame based on tribal politicking. Most of us worried about gene editing babies is because of folkore and myths from pop culture.
"Everything you know is wrongJust forget the words and sing alongAll you need to understand isEverything you know is wrong" 🎶https://t.co/JF0eLHHn4C
"New science needs to be judged on its merits, not by the disciplinary credentials of the people doing it—particularly in fast-moving interdisciplinary areas where any formal training may be outdated anyway."https://t.co/p3af4nNgEBhttps://t.co/4Bl9KaBdof https://t.co/fIhwRalm0r
"Does it strike you as odd that so many people tuned in to hear about a doctored speedrun of a children’s video game, while barely a ripple was made—even among scientists—by the discovery of more than 80 fake scientific papers?"https://t.co/sRRq5ltdyh
Science as it currently exists isn't entertainment. Its not developing scenes on video platforms, nor forming industries of people making a living by showing it off like a sort of sports game on live streams.And it doesn't incentivize policing fake scientists who steal status.
It makes me wonder what it was like to be a scientist in the 1700s, just at the cusp of a new high status scene where people showed off curiosities. How were the status and trust networks splayed out at the time?https://t.co/ibA4ZcDkAChttps://t.co/w8RXhLNEMc https://t.co/3TouuTtUPl
"Robert Hooke, an English scientist, architect, and polymath, who, using a microscope, was the first to visualize a micro-organism, was able to see the nodal patterns associated with the modes of vibration of glass plates as he ran a bow along the edge"https://t.co/OoEtMB9NoC
Was Robert Hooke just cheating by plagarizing ancient manuscripts. Passing off discoveries as his own, to impress a bunch of people with too much time on their hands?I often wonder how many early scientific discoveries were just rediscoveries.https://t.co/VKQ4GJah1D
"from polished crystal, often quartz, and have been dated as early as 750 BC for Assyrian lenses such as the Nimrud / Layard [...] lenses from ancient Egypt, Greece and Babylon. [...] Romans and Greeks filled glass spheres with water to make lenses."https://t.co/xhVATgxzmJ
History is littered with people that seem to just be riffing off of the past.https://t.co/ofQyqry1Od
I even went so far as to reject the "doubt as far as possible, all things" quote.René Descartes "seeker after truth" quote was a memetic compression of an idea he probably read or heard second hand from a 10th century mathematical scholar. https://t.co/oT06R6iTGZ
And scenes developing around cool ideas, as people celebrate the rediscovery.https://t.co/TjJbv6NhGe
I am just rediscovering the wisdom of the past.https://t.co/b5CEoGiSrx
I wish this weren't true.https://t.co/dnXd1yuZYu
I'm not qualified to opine on the nature of scientific discovery.https://t.co/PvjO5XDBPm
If we believe Kuhn is correct, we'd expect there to be higher value in studying social systems around the nature of science, to improve things somewhat.https://t.co/j6LPbE2DcW
Maybe we can improve science somewhat?https://t.co/KSCAPApctv
What would we expect a new science to look like?https://t.co/i05vv2N3uK
if we want to improve the situation, we have to have a more accurate view of the phenomena and not this stupid myth of an idea that is little more than a long winded version of collective unconscious bullshit psychology Kuhn tried to ram into sociological explanations of science.
Maybe we should be leveraging network contagion models, and mapping how ideas spread among networks.https://t.co/5fW7nGoTAg
Trying to understand how ideas get fertilized, grow, and from complex structures in the agar that is our collective mindspace?https://t.co/BZYaVNJo9V
I find myself limited in this domain. The more I study how we model social systems, the less I find congruence. Even the very foundations of sociology seem to be ignoring the very field I thought would have the most application.https://t.co/AT3knwtXVf
(The 'Cybernetics' of Social Control and Power)"sociologists are generally unfamiliar with the control-theory perspective, and because it challenges many of our usual ways of thinking about human behavior, I start by describing the theory in some detail"https://t.co/3CaFknetPg
The places that seem to be working on the cybernetics space seem to be less about modeling social systems and understanding how ideas spread, and more concerned with military applications and machines. https://t.co/WYu18DUk3e
Looking at these spaces from a lens of social networks is so fruitful though. Thinking about how social networks almost necessarily bring about an emergence of an idea, and modeling hunches lets me learn a space of ideas way faster than a dry text book. https://t.co/lFn8xdholc
Brian Eno's scenius, but it's Norbert Wiener's Network Celebrity being used to engineer a collective steering behavior by resonating ideas from 1950's cybernetics research, tied to the hip with rocketry, engineering a form of weaponized autism. https://t.co/UIQuShw8kA
Even more abstract concepts—like the way that incentives within the collective end up guiding the short term future—just falls out of looking at this space under a lens of these ideas. We should be able to apply this to science as an institution, no?https://t.co/kDUhp1qkm2
2nd level Cybernetics."there's nothing very deep about deep learning. The technology will have far-reaching social and economic consequences, in large part because industry will steer economic activity toward the things that algorithms do well."https://t.co/N29IYvQ1mW
It's there if you look for it, but..."To the extent that cybernetics permeates the human sciences and our culture at large, it remains opaque – an only partially visible legacy often deemed too complex to form a simple object of historical narrative."https://t.co/es9wqiZbGJ
"...the history of science tells us that many concepts we consider as failed today were never explicitly disproven. Routinely, these concepts merely faded in their relevance or heuristic power, replaced by alternative explanations." https://t.co/qqvgNnUY74https://t.co/QZDFanIlUY
All studies are wrong. Some are useful.Deciding between which is which?That doesn't put food on the table.https://t.co/qduzjUKTkP https://t.co/ldnva6Py6C
"Good judgment requires that you turn knowledge into understanding. [...] Many leaders rush to bad judgments because they unconsciously filter the information they receive or are not sufficiently critical of what they hear or read."https://t.co/DLDg4g4pDV
> "In this environment, no sinister conspiracy is needed to allow for the construction of an irreparably flawed body of literature. In fact, the suspect quality of the [...] literature may be alarmingly commonplace."https://t.co/wHpsFwQxRr
The funny thing about playing battleship, is even a 'miss' can help you see the playing field better. The problem with using battleship under a model of Bayesian probability of shooting in the dark is that we don't bake in accuracy. What happens if you assume noisy samples?
Why assume people are useful idiots when we already have seen the entire system is rotting under the weight of bog standard bungling?https://t.co/ocN2mODENc
Feynman's "pleasure of finding things out" is tied to his "Science is the belief in the ignorance of experts".When you love finding things out, you eventually start butting heads with "experts" whom are famous not for being right, but for being useful.https://t.co/sucjrEaKhC
Someone told me I should get a PhD recently.I now have a 3rd party twitter thread I can use to respond to that with.https://t.co/3mk07ADIa9 https://t.co/zxmnwJbZMw
There's another reason I would never get a PhD. I wouldn't want things I make to be locked up behind patents owned by an academic institution, who would push me thru an incubator program and try to attract investors for a tool that has no market viability.
A friend shared this awesome tool with me recently and I recognized it. It's a good example of how the incentives in this space end up leading to hugely important ideas being shelved because the institution had no idea how to get a product to market. https://t.co/TOCqzVY1E9
Last I checked, it would take about $7000 to make a reliable personal backup of all of sci-hub + libgen.I know this because I wanted to do something like it myself.https://t.co/ZMeQ5CHWnR
I think about what it would take to store copies of it distributed in such a way that it was fault tolerant and resistant to confiscation. This is why I know about how to create vibration safe storage arrays for storing in the back of a van or a boat.https://t.co/JW27AolIZI
"if you're a software engineer at FAAAANG, then you can afford a 75+45 TB (scihub+libgen) hard drive array. [...] a $5-10k hobby project to store distilled human progress is something that you could make financially possible for yourself."https://t.co/5Bi1izPZ2L https://t.co/XFFXrNNDbr
My dreams are measured in library of congresses.https://t.co/h93vjwSMN5
I dream a lot. I slept in bed for 28 hours yesterday.https://t.co/nl5bvTCe0D
Deep breakdown of replication markets and how to make quick judgements on reliability of papers.https://t.co/G3bA2i7Pz7I tend to focus more on reasoning & models behind a paper (seeing errors like pictured). Need to pay more attention to process errors.https://t.co/JXWsf9rkgy https://t.co/TEy1caEGpI
That piece is a gem and is well worth a read. This part made me giggle. Echos a sentiment I hadn't been able to describe with my annoyance of reading evopsych and subtle feeling of just being angry when I do.https://t.co/U00QPRpTFn https://t.co/NTFcobj3q1
Sorry, but if two ideas in my head conflict, and one stems from Von Neuman's notions on the origin of life and self-replicating systems and one stems from someone making up just-so stories about plausible evolutionary forces to support a world view... https://t.co/sCdkgZDJit
"surveillance software like ThreatMetrix in their research databases, and data analytics companies like Clarivate are trying to acquire [...] to exploit library patrons' data to create more academic metrics to sell grant funders and research institutions."https://t.co/IEMOgppebj
Saw that in response to this thread breaking down the aggressive click-tracking on Elsevier's "enhanced pdf viewer" https://t.co/TszvHFpfX7
Someone also linked to an english copy of a review by @dfg_public (loosely translated is the "German Research Foundation") that explores the privacy implications. https://t.co/T5qR0Z3m2h Published October 2021. (see pdf) https://t.co/ht9KdCCDfW
"Long-term economics cannot support a system forced to endure such wide-scale theft and negative shifts in behavioral norms — and continuing on this path will produce negative outcomes for all." 🤔https://t.co/YTb09EgiH1
I stumbled upon an older tweet thinking about active manipulations of publications.I can't seem find attempts to spot active subterfuge. It looks like the main thing focused on is simple career interests /monetary gain. https://t.co/XZFcBQH92dhttps://t.co/82klhU8gPj
How much do you think they pay people at Internet Research Agency to manipulate western social science research to seed scissor statements? Would be a great way to do OSINT operations to target marks by watching how popsci articles spread among networks. https://t.co/3kdxQUVTUY
That thread is about taking mTurk surveys using crowds of people to manipulate results. Even there I could find little exploring large scale manipulations. Anti-fraud only focuses on people gaming it for money.It's missing an entire incentive angle.https://t.co/Lq1heiXCqQ
@JayTSack Someone trying to copy what internet research agency did for money - content farming. Spend $10 on ads, then rake in 1000x more by having stories with hundreds of ads on the page.Suspect an automated racket. They likely farm out to cheap writers and stuff like fiver for logos.
Excited to see movement against data manipulations in science publications."Level three manipulations include selective reporting or cropping of images so that they don’t represent the original data, and mislabelling or duplication of parts of images."https://t.co/QPNIxkuRAc
I'm struggling to even know what to look for. The best I can come up with is exploring research secrecy and privacy concerns around research. This 2006 paper has a good summary of that space.https://t.co/JWUAoUqhSi
Seeing all the data that is being collected via publications running click tracking and analytics of consumers-of-science-pubs raised my hackles. It made me wonder if any of this data is being sold to government entities to do reverse espionage.
And a corollary to that is thinking about other spy-work, like planting of false info and misdirection.Given how powerful it seems to have your thumb on the scale, I'd be surprised if larger political entities and NGOs aren't seeing it as an opportunity.https://t.co/z6jLQtbSVY
This thread here has made it abundantly clear that the scientific community is largely unable to spot or police cases of fraud. Further, initiatives like detecting of photomanipulation or data disparities seems to be fascinating because it is so novel. This alarms me.
"But this long history of learning how to not fool ourselves—of having utter scientific integrity—is, I’m sorry to say, something that we haven’t specifically included in any particular course that I know of. We just hope you’ve caught on by osmosis."https://t.co/g79rYZOytI
"A bigger obstacle to reliable research, though, is that scientists often simply cannot get at the things they need to measure."https://t.co/rjL4DiW3tFhttps://t.co/uTaodIdb4p
"Examples of how the streetlight effect sends studies off track are ubiquitous. In many cases it is painfully obvious that scientists are stuck with surrogate measures in place of what they really want to quantify."Oh hey, I know this one...https://t.co/YlaeKm1kqL
Press F(eynman) to doubt.https://t.co/2hQST6FMSR
https://t.co/GXAnGYxM9c https://t.co/gDt33NmLfk
"Ooh, Interesting, too bad you're totally wrong. In that way that the best science is often a long, slow, passive aggressive argument..."https://t.co/u43lC9Jlk3https://t.co/ozhJF38BOu https://t.co/SNr3T9vQvh
"The current [...] pandemic not only accelerated the speed of research but also brought to light some severe shortcomings of the scientific publication process, such as failures to quickly address errors or to catch and prevent scientific misconduct."https://t.co/LcqWXmOdE2 https://t.co/kgKVNRdZd1
"sobering and similar to those from US studies, which “suggests that the fundamental problem of conflict-of-interest non-disclosure is a persistent one” in clinical research — evident across journals, across countries and over time" https://t.co/3eGtgF5Ajlhttps://t.co/E15esKa3go https://t.co/1j965RZnuy
"Among the more than 5,000 trials [...] 934 were published by the New England Journal of Medicine. He flagged 11. One was the 2013 study showing the Mediterranean diet was responsible for the reduced rates of heart attack and stroke."https://t.co/e7u2MZqmVF (h/t @ladymcscope)
It bothers me to think that people don't do this kind of network analysis thinking on their own as they spider out thru papers to satiate curiosity.https://t.co/KPCD6TdREvhttps://t.co/hZeupLbRJP
Community detection to find clusters of research is an application of Conway's law in reverse, but using citations as a proxy for social networks.Why not just skip that step and go straight to the actual social networks? https://t.co/DIEFTsyzzI
When you scout around publications long enough you can start to get a feel for the people who write them. Maybe this is unique to me?https://t.co/MvM5zcA1ds
I figured out how to reverse a life long Prosopagnosia because I wanted to be able to recognize faces of famous scientists do speed up my ability to link ideas together.https://t.co/tw6FpbjW6wI am am not smart enough to explain how dumb we are all collectively being right now.
Heck, I can basically taste when I'm reading something from Wharton school for business because it's dripping with creepy ideas about how to manipulate people.https://t.co/ZnPgCrLDbj
I wanna rewrite Marshall Brain's "manna" using stuff I learn from Wharton & DARPA“You are the eyes and hands for this robot. And all so that Joe Garcia can make $20 million per year. Do you know what will happen if this spreads?” https://t.co/aectCRtcCvhttps://t.co/QFq7tiUMxc
If, as Afeyan suggests, the reason why they were able to do what they do is b/c they were unconstrained by academic demands. What can you do if you are unconstrainted by patent law and the need to turn a profit on your inventions?https://t.co/swVoCHZjVehttps://t.co/4YkoNXYYlg https://t.co/I8S4EPQ5Wp
"I think the more that people are aware of that, fewer and fewer people are going to be willing to take that risk—especially because there’s no guarantee of a tenure-track position at the end of it."https://t.co/gZHCLwLRWgWith good commentary in thread:https://t.co/zqcvsjdFPo
Scihub is goodhttps://t.co/NC0aFXXgSq
Scihub is goodhttps://t.co/SQ3OhCyUGP
Scihub is goodhttps://t.co/fP81POZP4T
Scihub is goodhttps://t.co/xTXEfQcPMs
Scihub is goodhttps://t.co/BR9fs6uFtj
This is a good worry. I hope more people worry about this. <3https://t.co/LxO21IykkC
"my frustration with the slow process of correcting the literature that has integrity issues and to demonstrate the potential adverse consequences if it is not corrected promptly."https://t.co/OrTcYjfcRyNeat.https://t.co/lF2jcxD5Qm https://t.co/VnDVfuudZV
"this isn’t an engineering problem that can be brute-forced or designed away, nor a political one that can be compromised out; the demarcation problem is a beast that’s never been solved, even by the best philosophers of science."https://t.co/G6UXHFHiUIhttps://t.co/28OvvPBrEr https://t.co/FGUh6UsXu3
"The NIH’s [...] process requires that many reviewers approve an application. This consensus-oriented style can be a check against novelty—what if one scientist sees extraordinary promise in a wacky idea but the rest of the board sees only its wackiness?"https://t.co/xoTpTIpere
'Impact' my butt."Moreover, by analyzing properties of over 2600 research fields, we observed that citation-based metrics are not beneficial for comparing researchers in different fields, or even in the same department."https://t.co/clICKv4N0yhttps://t.co/rHC8mVddOQ
"He found few scientific reports of children with autism improving while on antibiotics or ill with fevers. But when he spoke with parents in internet groups, he heard many such stories."https://t.co/HaJljo8Aw1
"immediate, obvious damage is wasted NIH funding and wasted thinking in the field because people are using these results as a starting point for their own experiments."https://t.co/3Xe8OsEQzBI'm a big fan of @MicrobiomDigest BTW.https://t.co/RfFcbCcv1L https://t.co/T3zYglllDj
the Alzheimer's stuff is really bad but this happens a lot in bio/med apparently more than people think just photoshopping results there are people out there who track this stuff, impression is there's a huge scandal every couple years that takes down a superstar
Gonna go wash out the instant pot so I can make more PEA for my GF.https://t.co/phl3LYKs3S
I also am a big fan of Rev.If that study was faked, this is an utter travesty on a massive scale.https://t.co/AuPZeUDHlQ
I wonder how much of ARPA-H's 6.5B went toward potentially skewed Alzheimer's research. https://t.co/2m1WLTjxpv
"total includes $6.5 billion for a new entity, the Advanced Research Projects Agency-Health (ARPA-H), that would initially focus on research into cancer and other diseases such as diabetes and Alzheimer's." https://t.co/O7Jb0xEgY1Different from HARPA? https://t.co/BPGja1jXCx
At least > $1 B given that.https://t.co/rL1M3Zhdh6
Jump in losers. Lets get them JORBS!https://t.co/i1IgLWnXUV
"So why have clinical trials of anti-amyloid therapies not provided clear-cut benefits to patients with AD? [...] These central questions in research on AD are being urgently addressed." https://t.co/QI0qNVFJcnhttps://t.co/p5X8j69NBN
"Something is rotten in the state of Denmark."https://t.co/sjcOpHyK1U
Hamlet on being betrayed."look you, this brave o'erhanging firmament, this majestical roof fretted with golden fire, why, it appears no other thing to me than a foul and pestilent congregation of vapours. What a piece of work"https://t.co/5O6egSm5BN