Reading Roundup - March 2025
A monthly roundup of random readings from around the web.
Double DOGE, jet engines, accounting fraud, litigation finance, ASML EUV lithography, MicroStrategy’s “Perpetual Strife” securities, and more. Enjoy!
This is the most thoughtful review of DOGE I’ve read. Two good critiques of DOGE are: (1) the focus on measurable business efficiency metrics—headcount and dollars saved—is avoiding the bigger cost of government in implicit costs (e.g., bureaucracy that costs 20+ hours per year, per person, on filing taxes); and (2) his "turn everything off" playbook from Twitter/SpaceX has cut state capacity by eliminating competent employees from programs like USAID. Two reasonable responses to those critiques are: (1) his “shoot first, ask questions later" approach might be the only way to defeat entrenched resistance in a limited timeframe; and (2) programs like USAID have real accountability problems that need addressing. The article contains other thoughtful points on the pros and cons of DOGE, but the conclusion is a good point to keep in mind: "I still find it hard to evaluate DOGE in toto. There are things I hate and things I very much appreciate, but it still feels too early to pass judgment."
(2) Is Elon Musk remaking government or breaking it?
“This does not mean DOGE has failed—yet. There are three possible outcomes. First, that just as rivals laughed at Tesla and SpaceX in their early days, DOGE will come around in time. Second, that Musk will break the government. The third, likeliest scenario is that DOGE becomes snarled up in court; many good civil servants are fired or quit; fewer talented people see government as an appealing career; and America is left with a stronger president and a weaker Congress. This would be a huge missed opportunity. Imagine the Musk of the early 2010s, the genius-builder, in charge of procurement at the Pentagon or federal infrastructure projects. Instead, America has got the late-era Musk, radicalized by his own social-media platform, flirting with authoritarian movements, stuck in the same mind-numbing partisan thinking as millions of less talented folk.”
(3) Why It’s So Hard To Build A Jet Engine
Building a jet engine is an engineering task of similar difficulty to building a leading edge semiconductor fab (TSMC), a wide body airplane (Boeing), or a reusable rocket (SpaceX). One challenge is compressor efficiency at higher speeds: over 50% of the generated energy goes to compress incoming air (the "back work ratio”), and adding efficiency requires more compressor stages that add weight until the thrust gains are offset. Another challenge is finding exotic materials that handle higher temperatures and pressures (titanium was already being replaced by exotic superalloys in the early 1960s), and measuring how those novel materials behave in high temperature and pressure environments for many years. The physics is hard, but the economics are even harder: “It’s not that building a jet engine that “works” is so difficult or expensive […] The difficulty is building an engine that meets performance targets—thrust, fuel consumption, maintenance costs, [etc.].There’s no point in designing a new engine if it doesn’t significantly improve on the state of the art. [That] means engine development [must push] technological boundaries: higher compression ratios, hotter temperatures, lighter weight, larger fans, etc.”
(4) Panthera Resources - Multi-Bagger Legal Claim?
Panthera Resources is a UK-listed nanocap ($24M GBP market cap) with a $1B+ litigation claim against India for the expropriation of a gold mine. Many bilateral trade treaties include investor-state dispute settlement (ISDS) clauses that facilitate binding arbitration of legal claims outside of corrupt local courts. Panthera, and its litigation finance backers, are using this method to get some value from its long-dormant legal claim. The story has lessons for (1) creative methods of litigation finance; (2) potential market inefficiencies in pricing inactive litigation assets inside of smaller companies; and (3) the barbell (venture-like) nature of litigation finance, although with more probabilistic precision. That said, “anyone interested in Panthera Resources should keep the old German saying: "On high seas and in courts of law, you are in God's hands" in mind.”
(5) A16Z and Benchmark-Backed 11x Has Been Claiming Customers It Doesn’t Have
11x is (was?) a hot AI startup that raised over $80 million in the past two years to create “AI sales reps.” In practice, their biggest innovation was a novel approach to revenue accounting. 11x signed new customers to an annual contract (say, for $1,000) with 3-month free trials, then booked $1,000 in monthly revenue, which investors annualized into “ARR.” In other words, cash accounting led to annualizing already-annual contracts. Many such cases! Even worse, most free trial customers cancelled before making a single payment because “their products barely work.” Yet investors are still defending them? Sometimes, the real product is the stock...
(6) Something Is Rotten in the State of Cupertino
Brutal criticism of Apple’s sclerotic culture in the context of its AI feature delays: “[Siri’s new AI features are] vaporware. These were features Apple said existed, which they claimed would be shipping in the next year. [The] fiasco is not that Apple is late on AI, [but] that Apple pitched a story that wasn’t true, one that some people within the company surely understood wasn’t true, and they set the course based on that. […] Tim Cook should have already held a meeting to […] rectify this Siri and Apple Intelligence debacle. If such a meeting hasn’t yet occurred or doesn’t happen soon, then, I fear, that’s all she wrote. The ride is over. When mediocrity, excuses, and bullshit take root, they take over. A culture of excellence, accountability, and integrity cannot abide [accepting] those things, and will collapse upon itself with the acceptance of all three.”
(7) Tyler Cowen, the Man Who Wants to Know Everything
“Cowen is famous not for a single theory but for the broad scope of his intellect. Put simply, he seems to know something about everything: machine learning, Icelandic sagas and where to eat in Bergen, Norway. “You can have specific and detailed discussions with him about 17th-century Irish economic thinkers, or trends in African music, or the history of nominal GDP targeting,” said Patrick Collison, co-founder of Stripe […] Cowen’s emotional life remains a mystery. He told me he did not experience regret. “I don’t know what the function of it is,” he said. “Is it to signal thoughtfulness? To stop you making further mistakes? It’s like revenge. I don’t understand it.” [He] also said he didn’t understand envy or anger. He didn’t know what he should be envious of. He didn’t get lonely, by himself or in company. He actually said, “Why bother?”” Teutonic energy, big time.
(8) The Race Is On To Build The World’s Most Complex Machine
The explanation of ASML’s EUV lithography machines never gets old: “ASML’s most advanced machine is mind-boggling. It works by firing 50,000 droplets of molten tin in a vacuum chamber. Each droplet takes a double hit—first from a weak laser pulse that flattens it into a tiny pancake, then from a powerful laser that vaporizes it [and] turns each droplet into hot plasma, reaching nearly 220,000°C, roughly 40x hotter than the surface of the Sun, and emits light of extremely short wavelength (extreme ultraviolet, or EUV). This light is then reflected by a series of mirrors so smooth that imperfections are measured in trillionths of a meter. The mirrors focus the light onto a mask or template that contains blueprints of the chip’s circuits. Finally the rays bounce from the mask on a silicon wafer coated with light-sensitive chemicals, imprinting the design.”
(9) Strategy Announces Proposed STRF Perpetual Preferred Stock Offering
“Strategy” (formerly MicroStrategy) is a bitcoin holding company that is getting innovative with preferred stock. I’ve never seen anything like this before… They are issuing perpetual preferred stock (not perpetual strike, but perpetual strife… never heard of that before) with paid-in-kind (PIK) features. The preferred stock cannot convert to common stock, it cannot be redeemed, and (at Strategy’s election) it could also forego payment of the cash dividend in perpetuity. In other words, they are issuing a security that will likely never return a dollar to investors. On the other hand, that could also describe bitcoin and most startups? Sounds lucrative for the bankers.
(10) The Horrors of Shared Docs
Quality content from the Economist: “Shared docs tend to get shared, and they frequently end up in the browsers of total strangers. You don’t know why they are in it, and they probably don’t either. Some people may not be identifiable at all: they have usernames like “anonymous tapir”. They might be the boss; they might be one of the interns. They are probably not a tapir. When enough people are involved, chaos ensues. Someone adds a paragraph, causing someone else to make alterations in the wrong place. [One] anonymous user spends so long on rewriting a sentence that people start to wonder whether it may actually be a tapir after all. [Countless] emails arrive to tell you that there has been “activity” in the document. [One] person is always there, day and night, but never does anything.” For even more horror, try using Notion.
(0) Miscellaneous
Feudalism as a contested concept. Hilarious review of Meghan Markle’s new cooking show. The political implications of Manus AI. Making a better AI accounting system. Why are so many unusual skyscrapers being built in Tirana, Albania? The importance of repetition in the workplace. Some theories on the OpenAI non-profit buyout. Data integration at the IRS. Using Grok’s voice mode as a reading companion. Rippling’s corporate counter-espionage. Consciousness in babies. Booming market for second hand clothing. “I wonder if the fall of Rome was this stupid.” What is an AppLovin?