i do love the color of the sky

Casual reminder that genetics exists and is sociologically relevant.

(Denialism about this is one of the major points driving me away from today's leftist coalitions. Truth must come before justice, all else ends up in insanity.)

I thought long and hard about this statistic. Is it meaningful without base rates and absolute numbers? Is it meaningful without randomised controlled adoptions?

I mean I don't know either what this might mean specifically, it's simply the kind of a result that's clearly not complete coincidence that you can sweep under a carpet and declare to be nothing-to-see-here, everyone move along now.

If you're already a member of Tumblr Free Speech Fanclub or for that matter anywhere firmly on the right, you probably don't need to be told that merely asking questions about this should be permissible and people trying to suppress all discussion of it should not be treated as allies… but the interesting issue is, even quite a lot of the center-left do in fact think crime per se is a valid issue and I would like it to get thru to them that any radical anti-hereditarian worldview does not really help on understanding it.

Roughly how many people have you sent a direct message to in the past 24 hours (including for work)?

This includes text, email, tumblr chat, Discord DMs to one person, Messenger chat, Signal, Telegram.

This does not include tumblr asks, Discord servers, or other 4+ person group chats.

For 2-4 person group chats, please consider that entire group chat a separate entity that is one person. (So, if you're in a GC with A and B, and you message that GC as well as A and B separately, that's 3 people you've sent a DM to.)

For anything I've forgotten about, please count it in a way that feels right to you.

How many people have you sent a message to in the past 24h

Seriously, what happened in the middle of the 20th century that made Americans go “Actually, names that begin with vowels are terrible”?

A

E

I

O

U

(source)

"If you do [preventative intervention], you'll just overcompensate with rashness and make everything worse."

At this point, in circles I travel in, this is uncontroversially considered a bad argument against condoms, masks, and seatbelts.

I hear more disagreement around bicycle helmets and quantitative reasoning.

How often is the argument actually correct? Can I just dismiss new versions of it by its track record? If not, are there easy broadly-applicable questions to answer to settle this point in any particular instance?

In the bicycle helmet case - I’m curious if the effect of helmets is to increase total collisions while reducing fatal collisions, which would be a good tradeoff.

iirc in the case of bicycle helmets, the relevant compensatory behavior is supposed to be that of nearby drivers

The thing with the bike helmets is that there’s data backing it up, and multiple other reasons why bike helmets are bad.

From the last time this came up:

1. Bike helmets overall provide a decrease in safety, partially direct (compensatory behavior by drivers) and partially indirect (the hassle leads to fewer cyclists which lead to drivers having less experience with cyclists)

(source)

2. If drivers or pedestrians wear helmets, it would decrease their injury chance by much more than if cyclists wear helmets – so if you don’t wear a helmet while driving, there’s no reason to wear a helmet while cycling

A 1996 Australian study found:

Risk of head injury per million hours travelled:

(source)

Uh, was this supposed to be a chart providing evidence for a connection between wearing bike helmets and bike accidents? I see mainly a correlation between biking being rare and bike accidents, and I would think even quite a lot of that comes down to a lack of bike lanes or similar infrastructure. The first linked article seems to agree: “We already have the world's best cycling safety statistics, second only to the Netherlands, thanks to our separated bike infrastructure.”

…As far as the rest of it goes though, the heavy rhetorical push on branding helmets as a “religious symbol” does not do any favors on just trusting them on their main claim, and I see no evidence presented whatsoever for the nonsensical claim that helmets would decrease safety.

Article 2 seems like an isolated study as a part of an ongoing discussion. For that matter, I’d be interested in knowing if “motor vehicle occupant” in fact includes motorbikists who might be driving the accident rate (already known to have many times the accident rate of car drivers). Also the claim that pedestrians have a high risk of head injury really does not make sense to me at all and I wonder how this is supposed to work?

I would appreaciate some link to “the last time”, whereever and whenever this was.

Nassim Taleb is sassy af.

I admit I don’t have an intuition on what a dataset would look like where A positively correlates with B and B positively with C but C negatively with A (or what kind of a causal relationship would produce such a pattern).

cool but also: let’s see you transform a star into an X into a dinosaur while having any kind of a statistically significant correlation going on

fewer than a million people live in sf and i. i never knew that. my worldview has been shattered this morning.

Isn’t that true of most cities? In the US at least I think there are only maybe 10 cities with more than a million people in the city proper.

I was having dinner one time with a colleague who just moved to North Carolina from Karachi, Pakistan. I asked him how he was adjusting and he said “it’s hard being in such a small town, I’m used to big cities. In America, there are no big cities.” It kind of surprised me, but it’s true. Other than NYC, which isn’t even in the top ten worldwide, we have basically no big cities.

Oh wow, yeah. I was just looking at https://worldpopulationreview.com/world-cities, and NYC, which is #42 worldwide, is apparently the only US city even in the top 100 worldwide (LA barely misses the cut at #103). I knew we had smaller cities on average than many countries, but I don’t think I ever realized by just how much.

This is currently true, but this is also very recent really.

The first modern city to reach 1 million people was apparently Beijing around 1800. (Some places, most prominently Rome circa 1 CE, Baghdad circa 1000 CE and a rotating succession of Chinese capitals from the mid-1st millennium on have reached there earlier but have not sustained that for longer than a century or two.) The first to reach 5 million was London arond 1890. The NYC metropolitan area reached 10 million people around 1930; I’m not sure what would have been the first city proper to do so, perhaps Tokyo (1963). For most of world history, anything over 100,000 would have been a “big city”.

Karachi meanwhile broke 100k people in 1900, 1M around 1950 and 10M around 2000.

New York might be no longer one of the absolute biggest, but using some kind of an “integrated population” metric: take the population each year and sum these together — it might be one of the only few cities world history to have maintained more than 500M people-years per century. Plenty of other cities in the US would also rank much better by this metric than by their current population.

But yes, the 21st century will probably change this quite a bit…

From the Economist:

I am not familiar enough with the happiness literature to say whether this is consistent with it, but the connection between our resources and our happiness has always seemed straightforward to me.

Top 25 Coffee Consuming Nations

I keep wondering how much of this is because it’s in units of kilograms per capita… A thing about coffee consumption is that in Northern Europe in general and Finland in particular we prefer lightly roasted coffee, which weighs more than dark-roasted coffee (roasting drives water out); counting instead bean for bean, the chart could well look more even.

(According to folk wisdom, this goes back to how way back when households used to roast their own coffee, Finnish households developed a meme “you get more coffee out of your coffee if you roast it less” — and we then developed a taste for the kind of weaker and sourer coffee this yields)

[caption]

The 2010s — Ten Reasons It Was The Best Decade Ever

Sources: Overall: Johan Norberg, “Progress: Ten Reasons to Look Forward to the Future”. 1) World Bank, 2) World Bank, 3) UN IGME, 4) UN, 5) ILGA, 6) World Bank, 7) Our World in Data, 9) McAfee, “More from Less” 2019, 10) Freedom House. Data collecting takes time, so some numbers refer to 2007–2017, 2008–2018 or 2009–2019.

[/caption]

For all of China’s growing global influence, I’ve seen very little demand for its modern cultural exports (in my small anglosphere perspective, anyway)

Japan has been running up the score for decades, and has a lot of positive affect in the West even when its Imperial atrocities are still in living memory; Korea has been catching up fast, and it can field both slick pop groups and beloved auteurs like Bong Joon-Ho

But most of the Chinese artists acclaimed in the west are people the government detests, like Ai Weiwei

I’m unsure if this’ll even be a big factor in the long run, but it does seem like the CCP is missing out on a vital way to build international sympathy, to their peril

I’m not sure you can become a major cultural exporter without also having some semblence of a liberal democracy, honestly--it’s hard to produce media that can compete internationally if it’s chafing under the demands of the censors. Compare the cultural reach of the US and the Soviet Union during the Cold War, or the success of India’s film industry abroad compared to China’s. Japan only became a big cultural exporter post-war, and K-Drama and K-Pop only went big after SK’s dictatorship fell. As much as the CCP might like to get the world hooked on Chinese cultural exports, it’s difficult to see a path to that that doesn’t involve the CCP ceasing to hold power in the way it currently does.

(OTOH, I’ve heard that Egypt is a pretty important cultural exporter when it comes to Arabic-language media, but that’s a playing field in which free societies are pretty thin on the ground--if there were more Arab democracies around, perhaps they’d dominate the media landscape there. Makes me wonder how influential PRC media is in other Chinese-speaking countries.)

Possibly related: I recently ran a little analysis of the top album charts at Rate Your Music (which skew a lot more esoteric than best-selling charts etc.) per country and the distribution is preeeetty uneven even beyond the degree of democracy.

The top 100 has albums from just five countries:

The top 500 adds a bit more variety, though mostly from continental Europe:

Russia debuts at #951, Korea at #1657, India at #1963, Mexico at #2003, Iran at #2506, Indonesia at #3979, Turkey at #4000, Thailand at #8076…; there is still nothing from China within the entire numbered top 10000.

But then this obviously cannot be read as anything like a measure of global influence anyway! Some places like Iceland, Sweden or Jamaica clearly in general punch above their size in music (some of these are just cases of an exceptionally popular musician from a relatively obscure country); and we can then ask if the UK and the US are not indeed also doing that on top of being great powers. After all the US is obviously doing this in cinema. But if Russia or France can be major players in world politics while not exporting all that much culture, China can surely do the same.

This is surely going to be bad news for the international adoption of Chinese of course; media is where people learn English from these days.

TIL Chicago has more people than Paris, which feels wrong somehow.

The populations of administrative “cities proper” are often not very compareable at all between countries… China has some particularly egregious examples where these actually reach higher than the real urban population, e.g. Chongqing has about 30M people administratively, but only about 18M within its own urban area; Hefei has 7,8M+ people administratively, but only a bit over 3,7M within its own urban area.