- Andy Greenberg has a long and wonderful Wired story about Mirai, the botnet responsible for record-setting denial-of-service attacks and a widespread outage in 2016 that took down several different $100bn+ market cap companies' sites in 2016. It’s full of useful details on how the web's underbelly functions; the worst outcomes seem to be when actively sociopathic people team up with bright and curious youngsters who want to understand how the Internet works and who view things like 1Tb/s DDoS attacks as a reasonable test of their skills. (The piece is also one more datapoint in favor of the thesis that The Social Network had the biggest economic impact of any movie in history, though the movie with the biggest historical impact overall is probably The Shield and the Sword.)
- Tyler Cowen recently noted that public intellectuals are getting more religious, and pointed out offhandedly that their affiliation "could be Catholic or Jewish or LDS or Eastern Orthodox, with some Protestant thrown into the mix, but Protestants coming in last." Two good quick pieces exploring why Protestants come in last, from Fergus McCullough and Aaron Renn. It's an interesting question, because Protestants are about twice as numerous as Catholics in the US, and at least in the first few centuries after the Reformation, Protestant parts of Germany outperformed Catholic ones on measures of human capital.
One possibility is that it's a bit like this argument for why the Lisp programming language never became dominant, but has always been influential: a more fragmented faith tradition has a lower cost of exit, so it's easier for people to find the exact denomination that matches their preferences, but that means no one denomination reaches critical mass and that everyone who has made a considered choice in the matter has lots of specific reasons to argue with someone who made a slightly different choice. Whereas in more centralized belief systems, those differences need to be worked out within the faith. (One interesting running theme in Huston Smith's The World's Religions is that sufficiently old religions end up having different tracks that make sense within that context, i.e. they've had to find ways to make the faith attractive to kids, families, the elderly, introverts, extroverts, nerds, jocks, ascetics, epicures, etc. If a religion is newer and hasn't had to deal with that problem, it can lead to a schism. Whereas with an older tradition, the niche interests get incorporated—"Oh, your favorite thing in life is deadlifting? There's a patron saint of that! And you really, really love pierogis? You're not going to believe this, but...")
- Steve Randy Waldman on the internal contradictions of the social and moral aspects of the postwar order. These kinds of critiques get written all the time, but they're typically right- or far-left-coded; this is a left-but-not-extremely-so perspective. A good line: "Entropy is a fact of social affairs even more than it is a fact of physics. If we want many different things, we will have to expend work and effort towards each of those things, and navigate tradeoffs between our various ends and the means required to achieve them."
- Nathan Tankus got the Fed to release its internal memo from 2011 on how to get around the debt ceiling. The upshot is that one of the tools they have is the ability to treat defaulted treasury bonds as equivalent to non-defaulted ones, whether for the purpose of repurchase agreements or outright purchases. And that's all you really need to ignore the debt ceiling almost entirely—there will be a population of normal bonds that live on private-sector balance sheets, and as long as the Fed has room to buy the ones that default, the market can operate fairly normally. Which means it's a somewhat dangerous memo to release: debt ceiling negotiations are a periodic opportunity for theatrics, a sort of political passion play where we all remind ourselves that borrowing has consequences by pretending those consequences are more immediate than they actually are. But the whole event only works if the risk feels real for all sides; if there's actually more room to maneuver, then debt ceiling negotiations are only meaningful when they push much closer to the edge.
- Jerry Neumann on why the right discount rate startups should use in net present value calculations is probably 40%, not the typical 10-15% most companies use. Discount rates implicitly determine every investment, even when they're unconscious—if you buy a house rather than rent, or vice-versa, you're implicitly using a discount rate that estimates the present value of future housing, for example. But they're very tricky to reason about, in part because of the risk of double-counting or under-counting risks. The approach Neumann argues for in this piece is to bake the risk into the required rate of return: if you were in a boring industry, you could aim for a 15% return because a mistake might get you to 12%, which is still good. But in a new industry, there's a wide margin of error; the internal rate of return on your new AI product is probably negative, but if it's positive it's a lot higher than 15%! Since founders are naturally optimistic, and since they have a high cost of capital, they should set a high threshold. A high discount rate is just a rigorous, analytical way to implement Derek Sivers' directive that if the answer isn't "HELL YEAH!" it's "No."
- In Capital Gains this week, we cover the most important chart in finance: when a trading strategy gets more crowded, risk-adjusted returns momentarily go up, because more new traders are making the bet that you already had on. But that's a temporary blip. This is the chart behind many promising careers that ended surprisingly early, and it generalizes quite well to other domains.
- By popular demand, there is now a Diff podcast, The Riff, produced with Turpentine Media. You can listen to The Riff on Spotify, Apple Podcasts, or YouTube. You can listen to Episode 1 on SBF, Cryto, Influences (Spotify / Apple / YouTube) or Episode 2 on OpenAI, Investing Bets, and Social Media Paradigms (Spotify / Apple / YouTube)
- Romney: A Reckoning: A fun feature of the US political system is that, like many other fields, it's gotten much more professionalized and standardized over time. The Presidency is a demanding job, inasmuch as it involves running the largest organization on earth (by budget) and one of the most complex in terms of responsibilities. Running for office has also gotten more competitive, to the point that someone who wants to do it well really needs to specialize. Which means we select for a demanding job through a method that is also extremely demanding, but requires an entirely different set of skills; it's as if a comedian who did a really killer standup special for Netflix immediately got promoted to their VP of Infrastructure.
In the book, Mitt Romney comes across as someone who is good at running nearly anything, with the one exception of "for office." If America had a board of directors rather than an electoral college, Romney probably would’ve been in charge at some point. But the other thread that runs through the book is the continuous evolution of his willingness and ability to draw a moral line. He was quite well-known as a principled Trump critic, for example, but ran for office in 2012 attacking Obamacare, a program whose progenitor had been launched by Romney himself as governor of Massachusetts. His evolution on social issues shows a similar drift; as a Republican running in a Democrat-leaning state, his social views evolved towards the center; as a Republican running in Republican presidential primaries, he experienced another sudden bout of enlightenment on social issues and tacked right. Which is the nature of the game; the real measure of a politician's principles is probably the margin of their defeat.
- Drop in any links or comments of interest to Diff readers.
- What are everyone's day-after thoughts on the situation with OpenAI? There's still a lot of uncertainty about what happened and what's going to—obviously things were still in flux hours after the press release came out. I would note that when the claim is that Altman was "not consistently candid in his communications with the board," there's a wide range of unknown unknowns.
Companies in the Diff network are actively looking for talent. A sampling of current open roles:
- A company building the new pension of the 21st century and building universal basic capital is looking for a GTM / growth lead. (NYC)
- A diversified prop trading firm with a uniquely collaborative team structure is looking for experienced traders and PMs. (Singapore or Austin, TX preferred)
- An early-stage startup aiming to reduce labor costs by over 80% in a $100bn+ industry is looking for a part-time technical advisor with robotics experience; this has the potential to evolve into a full-time role. (NYC)
- A concentrated crossover fund is looking for an intellectually curious data scientist with demonstrated mastery in analytics. Experience with alt data, web scraping, and financial modeling preferred. (SF)
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.