pithymaxim

I think it's more important to look at all-cause mortality, because treatment can also harm you. In this case it's a fairly tight confidence interval: "The risk of death from any cause was 11.03% in the invited group and 11.04% in the usual-care group (risk ratio, 0.99; 95% CI, 0.96 to 1.04)."

Mostly agreed. I think everyone (including myself) was ready for a huge pop science book to be bad, and boosted the post based just on its title.

In defense of the post, Walker's called-out claim that "Routinely sleeping less than six or seven hours...double[s] your risk of cancer" does seem pretty out there. Even if that correlation were well-established it would be hard to interpret.

Based on that last entry about the SF bus line, I would really love to see https://patrickcollison.com/slow

I always thought this was just price discrimination, extracting more surplus from readers by bifurcating them into two groups. Hardbacks are released first. Everyone who really wants it now needs to shell out more, even if they don't really care about the physical properties. Then the paperback is released to attract the more reluctant readers.

Slide summary from the authors: https://bcerp.org/wp-content/uploads/Session3-AlexandraWhite...

For those interested in the actual policy:

> [C]hecked-out books automatically will renew as many as 15 times, as long as no one else places a hold on them...Items will be marked as "lost" and accounts will be charged a replacement fee one week after the last due date, but the charge will be cleared if the item is returned

https://www.chicagotribune.com/news/breaking/ct-chicago-libr...

Curious to see some library data guru openly analyze the consequences...

>nearly 40 percent of Americans, a Federal Reserve report found, are in such a financially precarious state that they say they would have trouble finding $400 for an unexpected expense like a car repair or a medical bill.

This has got to be the most repeated statistic in reporting on hardship, but the actual responses seem way less dire: https://twitter.com/p_millerd/status/1118071142311288838?lan...

> Not worried enough? Imagine an attacker who manipulates road signs in a way such that self-driving cars will break traffic rules.

If someone was manipulating a bunch of road signs for bad, human drivers would probably be in trouble too..I guess the point is that these attacks could be more subtle/scalable?

Is the headline figure here actually kind of small?

| The IEA finds that global offshore wind capacity may increase 15-fold and attract around $1 trillion of <<cumulative>> investment by 2040.

Averaged over the next 20 years, that's $50 billion per year, or 2.5 percent of oil and gas revenues in 2017[0], or 7 percent of what was invested in oil and gas supply in 2016[1]. I could be thinking about this wrong but it seems like there numbers could have been much more encouraging.

[0] https://www.investopedia.com/ask/answers/030915/what-percent... [1] https://www.reuters.com/article/us-iea-energy-investment/ele...

In case it's useful I immediately got a dead link https://argdown.org/guide/a-first-example/ from this page https://argdown.org/guide/

Checklists are probably good but looking at the actual paper[0] I don't think we can conclude "a third" with much confidence. The mortality rate for surgeries was already declining--the argument is that it declined faster during the implementation period (Fig 2; 2008-2010). Also it appears the types of admissions changed a lot during the same period based on the stark difference between Fig 1 (unadjusted) and Fig 2 (adjusted).

I'm surprised that the researchers didn't use better data or variation (e.g., hospital-level mortality or the timing of the rollout) to strengthen the case. So much can be lurking beneath annual country averages...

[0]: https://onlinelibrary.wiley.com/doi/full/10.1002/bjs.11151

Yep. Also this is all based on stock price data, so it could be that businesses typically hire the person that the market "expects"...which would predict zero alpha even if counterfactually hiring a bad CEO would've tanked the stock price.

There's some cool evidence that even this system led to non-negligible "misvoting"

> minor candidates' vote shares almost double when their names are adjacent to the names of major candidates

https://users.nber.org/~luttmer/misvotes.pdf

>Instacart and DoorDash...have figured out a clever way to use customer tips to lower their labor expenses.

Isn't this how restaurants work..?

Worth mentioning that their summary measure of well-being is more comprehensive than what's typically asked of adults and probably does correlate with learning and development:

"Data from 19,957 telephone interviews with parents of 2‐ to 5‐year‐olds assessed their children's digital screen use and psychological well‐being in terms of caregiver attachment, resilience, curiosity, and positive affect in the past month."

https://onlinelibrary.wiley.com/doi/full/10.1111/cdev.13007

Still plenty to quibble with in the study though.

This is super cool although one thing that I would do differently--after running--is to grab a beer and act a little more casual. Alcohol can help you relax if you're an occasional user, and there's the concern that you can make events like this traumatic by calling them traumatic.

Maybe, but the regression they use to argue this is pretty bonkers...they need to assume that once they control for income, population, education, and employment, the coefficient on the Airbnb share captures the causal effect of Airbnb on prices. But obviously AirBnB share will be correlated with many other things that make a neighborhood nice, which will also be correlated with prices. So this just says that desirable neighborhoods are desirable.

Similar argument in this paper but with better methods and data--2 million LA middle schoolers! http://home.uchicago.edu/~npope/morning_afternoon.pdf

Not an Uber fan but this just seems like price discrimination with better data, which in principle could subsidize rides for poorer drivers. Student rush tickets to the opera could be framed as price gouging of non-students (by probing into their personal lives, no less). But in reality more students get to go to the opera this way. Hard to judge without knowing the tradeoffs.

The headline is misleading. 22 percent of Germans chose to fully equalize the endowments (or "help others") compared to 17 percent of Americans. This seems striking for how similar the proportions are, not how selfish the Americans are.

Didn't mean to make that claim, but I'd guess that chess is better for cognitive development because its demands on working memory are limitless. This is speculative, but it's also speculative to claim that 2048 is any better for learning than Candy Crush.

I think #5 is wrong. 2048 is based on numbers but has very little to do with math--you can play the game just as easily using only colors. At best students are learning the powers of two, and at worst they're wasting time they should've spent on chess or checkers, or basketball.