The Diff - Longreads + Open Thread

Today's issue of The Diff is brought to you by our sponsor, 2 Hour Learning.

Longreads

• In The New Yorker, Charles Duhigg covers the rise of tech industry lobbying. The thrust of the piece is that in the good old days, tech companies made products and made money, and political operators made issues salient and mobilized voters. But tech companies have done something wildly unfair that upset the equilibrium: they've started trying to make issues salient! They're even mobilizing voters! From this perspective, it's actually quite a funny piece: lots of Washington insiders acting utterly shocked that someone might, for example, organize a protest and then choose camera angles that make it look much bigger than it is, or that a PAC might go after a candidate who was already going to lose in order to take credit for it. Anyone who has read the Washington Post or Politico on two consecutive days knows that DC people are constantly shifting their alliances, priorities, and aims, and it's not especially surprising that tech companies would decide to get involved and be able to afford talented operators of their own. (While it's not central to the piece, there is a factual error that's important to call out: "Most fiat currencies—that is, those issued by governments—are used primarily to buy such things as food and clothing, rather than to gamble on the rise and fall of exchange rates." This is flatly untrue. The Bank of International Settlements says that FX turnover is about 30x global GDP, so while it's true that crypto is mostly used for speculation, it's also true that 97% of fiat transactions are just bets on relative prices of tokens.)
• From the archives: this interview between Patrick McKenzie and Ramit Sethi is a great course in pricing. Much of this has become received wisdom in the decade-plus since the interview came out, but that's a good reason to revisit it: the point of Chesterton's Fence is not that you should never change the rules, but that you should try to appreciate why they made sense before you try to fix them.
• Eric Schwitzgebel on the dominance of the Silent Generation in academic philosophy. In manufacturing, Toyota likes the idea of asking "Why?" five times in a row to get to the root cause of a problem; three-year-olds like even further recursive depth. But in social sciences, compounding "Why"s mean compounding uncertainty. This piece is a great two-why summary for the phenomenon it observes. First, after dismissing a few other hypotheses, it argues that these philosophers get cited more because they did objectively more important work than either their immediate predecessors or the following generation, and the reason for that is they were demographically fortunate to start their careers when philosophy professor jobs were plentiful, and to reach their peak prominence when there were many more grad students and early-career professors desperately publishing to have a better shot at tenure. Luck is a form of leverage: if you end up in fortunate circumstances, it magnifies the upside from existing talents. (And that works at multiple levels: when jobs are more competitive, there's more randomness in who gets the best ones, since selection mechanisms can't be fine-grained enough to capture every meaningful difference between highly-skilled applicants. There is luck involved when a profession is lower-status, too, but it's the luck of being genuinely interested in something before it gets more popular.)
• Richard Ngo on limits of the Bayesian worldview. One of the ironies of Bayesian statistics is that it's both a very practical way to think about uncertainty and a very impractical approach to actually implement outside of stylized scenarios. Once you move from thought experiments about urns with multicolored balls to messier real-world questions, it's hard to exhaustively document the potential hypotheses in order to accurately update your view on which one is explanatory. One partial way to rescue it is to remember that whether or not there is some underlying truth that we're trying to assess, all we have to work with are our observations and our hypotheses about what explains them. There's a difference between saying "1 + 1 will always and everywhere be precisely equal to 2" and saying that because of this, any computation that can be reduced to addition is 100% reliable. It's the difference between believing in truth and believing that you've identified something that's 100% true, and it's useful to have ways to think about both.
• Katherine Dee argues that culture is not stuck, it's just evolving fast in places that aren't viewed as high culture. Which is true—one thing teachers often emphasize is that Shakespeare, Beethoven, etc. were, if not strictly popular entertainment, at least a lot less formal and stuffy to contemporaries than they seem to be today. (If they read as stuffy, one reason for that is that very stuffy people are reluctant to take the risk of genuinely enjoying something that hasn't been canonized as high culture yet.) If English Lit 101 still exists in a hundred years, I still think they'll spend more time on The Bard than on TikTok, but it's also true that cultural evolution is easy to miss if you have a definition of culture that defines away the most dynamic parts.
• In Capital Gains this week, we looked at video games that simulate the economy. As a few readers pointed out, Eve Online is an obvious omission here, for the very good reason that I don't have the time to get really into Eve. Simulations in general can be a great educational tool, but there's a big gap between designing a game with an open-ended simulation to see what you learn, and designing a simulation around ensuring that players learn some specific lesson that isn't a natural emergent property of a well-designed simulation.
• In this week's episode of The Riff, we're discussing money, legitimacy, the surprising rarity of young entrepreneurs, disrupting education, and more. Listen with Twitter/Spotify/Apple/YouTube.

Books

God's Crucible: Islam and the Making of Europe, 570-1215: Stories about conquest are often stories about collapse. Alexander the Great invaded an unstable Persia, Cortés showed up in Mexico right when the Aztecs’ neighbors had gotten well and truly fed up with them, Russian communism took advantage of the ineffective late empire, etc. And, in the span of less than a century, the Caliphate conquered a sizable fraction of what had until recently been the Roman empire, and also managed to outdo the Romans in Persia and grow beyond that.

American students in history classes tend to encounter medieval Europe as a general concept, without much reference to what was happening in specific countries. When we encounter Spain, it's a very Catholic country (especially because that first classroom encounter is likely to be in reference to the context in which the US became an independent country). But the Spain that financed voyages to the Americas and conquered swathes of North and South America had actually spent much of its recent history being ruled mostly by Muslims, and the Muslims who took over conquered it from Visigoths, who were Christian but not Catholic (specifically, for any prospective time travelers in the audience who wish to avoid insulting their Visigothic hosts, immediately pre-Islamic Spain was ruled by Arians—as in Arius, not the other kind—so it was a somewhat newly-Catholic country. (This comes up in books about the Reformation, like the one reviewed here: Spain was more serious about Catholicism because it had more recent experience with religious diversity.)

There's a surprising tendency for legendary conquerors who won against staggering odds to be fairly tolerant of cultural and religious differences once they're in charge—the Ottomans, the Mongols, al-Andalus, sometimes the Normans (the original Normans were so tolerant that, within a few generations of extorting a chunk of France from the king of France, they were much more French than Norse). But it's actually a likely outcome: being organized enough to win battles is very different from having the numbers necessary to impose entirely new behaviors on a much larger host population, so tolerance was the winning move.

Open Thread

• Drop in any links or comments of interest to Diff readers.
• Is it harder to start new religions than it used to be? Or has the category of "religion" solidified enough that new quasi-religious movements wind up getting different descriptors?

A Word From Our Sponsors

Can this really be true?

Diff Jobs

Companies in the Diff network are actively looking for talent. See a sampling of current open roles below:

• A number of companies (large crossover funds, startups, etc.) in The Diff network are looking for data scientists that love wrangling and analyzing alt. data to help inform investment decisions. (Remote)
• An AI startup building tools to help automate compliance for companies in highly regulated industries is looking for a director of information security and compliance with 5+ years of info sec related experience at a software company. Experience with HIPAA, FedRAMP a plus. (NYC)
• A company building the new pension of the 21st century and enabling universal basic capital is looking for a general counsel to help lead capital markets and regulatory efforts. (NYC)
• A hyper-growth startup that’s turning customers’ sales and marketing data into revenue is looking for a product engineer with a track record of building, shipping, and owning customer delivery at high velocity. (NYC)
• A well funded startup founded by two SpaceX engineers that’s building the software stack for hardware companies is looking for a new grad / early career software engineers. (LA, Hybrid)

Even if you don't see an exact match for your skills and interests right now, we're happy to talk early so we can let you know if a good opportunity comes up.

If you’re at a company that's looking for talent, we should talk! Diff Jobs works with companies across fintech, hard tech, consumer software, enterprise software, and other areas—any company where finding unusually effective people is a top priority.