Links & What I've Been Reading Q4 2021
1. Investigating the replicability of preclinical cancer biology: "50 experiments from 23 papers were repeated, generating data about the replicability of a total of 158 effects [...] for positive effects, the median effect size in the replications was 85% smaller than the median effect size in the original experiments"
2. A catastrophic failure of peer review in obstetrics and gynaecology: "I estimate that across these 46 articles, 346 (64%) of the 542 parametric tests (unpaired t tests, or, occasionally, ANOVA) and 151 (61%) of the 247 contingency table test (Pearson's Χ² or Fisher's exact test) that I was able to check were incorrectly reported."
3. The Business of Extracting Knowledge from Academic Publications: "Close to nothing of what makes science actually work is published as text on the web."
4. A large replication project in marketing, with fairly catastrophic results. Amusingly the abstract doesn't mention the rate of successful replication.
5. Increasing Politicization and Homogeneity in Scientific Funding: An Analysis of NSF Grants, 1990-2020. The methodology is somewhat questionable, but insteresting nonetheless.
7. Zvi's latest.
You are probably going to get Omicron, if you haven’t had it already. The level of precaution necessary to change this assessment is very high, and you probably don’t want to pay that price.
8. ADS on the Zvi-Holden bet and taking ideas seriously.
Making a blockchain game might genuinely be the best use of Zvi’s time, and he might be acting both rationality and ethically in choosing to pursue it. And so this situation is Good, but only in a very limited and local sense. The tragedy isn’t Zvi’s decision, it’s that a scenario even exists where this is the decision he has to make.
9. Omicron spreading faster than delta because of immune evasion? SARS-CoV-2 Omicron VOC Transmission in Danish Households. Plus twitter thread.
10. Forecasting in the Field: academics and non-experts try to predict the effects of development interventions.
the average correlation between predicted and observed effects is 0.75. Recipient types are less accurate than academics on average, but are at least as accurate for interventions and outcomes that are likely to be more familiar to them. The mean forecast of each group outperforms more than 75% of the comprising individuals, and averaging just five forecasts substantially reduces error, indicating strong “wisdom-of-crowds” effects. Three measures of academic expertise (rank, citations, and conducting research in East Africa) and two measures of confidence do not correlate with accuracy. Among recipient-types, high-accuracy “superforecasters” can be identified using observables. Small groups of these superforecasters are as accurate as academic respondents.
11. The United Fruit Company? Good, Actually.
Using administrative census data with census-block geo-references from 1973 to 2011, we implement a geographic regression discontinuity design that exploits a land assignment that is orthogonal to our outcomes of interest. We find that the firm had a positive and persistent effect on living standards. Company documents explain that a key concern at the time was to attract and maintain a sizable workforce, which induced the firm to invest heavily in local amenities that can account for our result.
12. Reviews of Moby Dick from 1851. "This is an odd book, professing to be a novel; wantonly eccentric; outrageously bombastic; in places charmingly and vividly descriptive." I love it when modern editions of old books include their contemporary reviews, unfortunately it's not done very often.
14. Bloomberg report on Tether, including the story of how a French screenwriter ended up owning a Bahamian bank.
15. Vitalik Buterin on Crypto Cities.
This monster was watching Ethereum for an obscure mistake deep in the process of creating a transaction: the reuse of a number while signing a transaction. I went searching for this creature, laid bait, saw it in the wild, and found unexplained tracks. To understand how this bot works, we need to begin by reviewing ECDSA and digital signatures.
17. Some answers to my questions about Borges, Browne, and Quevedo: On Borges and Quevedo. "The (sad) irony in Tlon’s ending is, therefore, not in a contrast Quevedo vs Browne, then, but in the contrast (Borges + Quevedo + Browne) vs Tlon. Or, maybe, grecolatin tradition versus modernity. With a tinge of sad resignation for the slow but unstoppable victory of the second over the first."
19. SMTM wrap up the Chemical Hunger series on the causes of obesity after 20 posts.
20. On the NIH and the challenges of funding alcohol consumption RCTs. The big alcohol study that didn't happen: My primal scream of rage.
21. RCT of health insurance in India finds few positive effects: Effect of Health Insurance in India: A Randomized Controlled Trial.
23. An interesting ACX comment on reversals in artistic "progress".
it's a pattern that has repeated throughout history and around the world, one of naturalist art executed with great skill being deliberately replaced with highly abstract art not requiring as much skill.
The cave paintings of Chauvet Cave in France ca 30,000 BP (before present) are more natural and technically much more sophisticated than any cave or rock paintings found after 20,000 BP (some of which are quite abstract and stylized).
24. The Browser interview with QNTM.
26. Razib Khan: Out of Africa's midlife crisis
two San from different groups both living in Namibia’s Northern Kalahari desert, and speaking click languages from the same family, are more genetically distinct from one another, by a solid 20%, than a person from Stockholm is from a person from Shanghai.
27. Don't take psychedelics. "Results revealed significant shifts away from ‘physicalist’ or ‘materialist’ views, and towards panpsychism and fatalism, post use."
29. Interface | Part II, cool animation project.
30. A project that made 999 forgeries of a Warhol drawing, then randomly mixed in the original, and sold them.
32. And here's a cool remix of Hugh Masekela's Stimela.
The Man from the Future: The Visionary Life of John von Neumann by Ananyo Bhattacharya. Bhattacharya approaches his subject by focusing on ideas. The first chapter takes care of JvN's early life, and the rest of the book is split up based on the subjects he worked on: mathematics, quantum mechanics, the nuclear bomb, computing, game theory, RAND, and artificial life. Large parts of the book (I'd say about a third) are dedicated not to von Neumann but rather the work other people did based on his ideas. The game theory chapter, for example, covers Nash, Schelling, Aumann, etc. in economics, and John Maynard Smith, Price, Hamilton, etc. in evolutionary game theory. Bhattacharya is good at making all these technical subjects accessible without dumbing them down too much. JvN's personality, personal life, professional relationships, etc. on the other hand are given scant attention.
Overall it felt a bit too short. In less than 300 pages we get such a wide array of ideas, and the story of how they influenced so many people, that it often feels like we're just skimming the surface in a speedboat. I'd like to take a deeper, more ponderous ride in a submarine some day.
Meetings with Remarkable Manuscripts by Christopher de Hamel. Fantastically gorgeous book, filled with high-quality prints of medieval manuscripts. Pleasant conversational style. Just lovely all around. Not just about the manuscripts themselves, but also who owned them, their condition, where they're housed, the librarians taking care of them, etc.
The Rings of Saturn by W. G. Sebald. A book of digressions. The frame is a walking tour of England, and on it are bolted various musings on Sir Thomas Browne, Joseph Konrad, silk manufacture, the Taiping rebellion, and so on. The subjects flow into each other so you don't know where one digression begins and the other ends. However, Sebald kind of undersells how interesting his subjects are; comparing his notes on FitzGerald to the famous Borges essay, for example, makes me wonder how Sebald managed to turn such a fascinating subject into such a dull essay.
Conquistador: Hernán Cortés, King Montezuma, and the Last Stand of the Aztecs by Buddy Levy. I didn't love the book (it felt a bit sloppy, and the style isn't great), but Cortes is an incredible character. The determination, the ingenuity, the absolute ruthlesness. When he murders his wife at the end of the book, all you can think is "well of course he did". And self-aware too: "I and my companions suffer from a disease of the heart that can be cured only with gold"! Perhaps it is the contrast against the Aztecs that, in a way, softens his image? Going to try Prescott's History of the Conquest of Mexico next.
Over the Edge of the World: Magellan's Terrifying Circumnavigation of the Globe by Laurence Bergreen. Solid narrative pop history. Feels a bit rushed after the point Magellan's death. Exciting, adventurous stuff as you'd expect from the age of exploration.
A Man on the Moon: The Voyages of the Apollo Astronauts by Andrew Chaikin. Covers the entire thing plus a ton of backstory, very thorough (within its scope). Focused on the astronauts, and much of it is the preoduct of interviews with those astronauts, which is kind of obvious at many points as you're only getting one person's perspective on certain events. It would have been better with a broader, more objective view, in my opinion. The latter parts (after the first moon landing) include a surprising amount of geology! I read three books on the early space program this year and none of them was completely satisfying, I'm still trying to find the Richard Rhodes of Apollo...