Books

Gordon's argument is that the century from 1870 to 1970 was a one-time event. Electricity, running water, the internal combustion engine, antibiotics and the telephone transformed daily life in ways no later wave of invention has matched, and the productivity boom they drove was already fading before the computer arrived. The second half of the book asks whether the headwinds — inequality, education, demographics, debt — mean the slowdown since 1970 is the new normal. The evidence is dense and the conclusion is uncomfortable, which is why the book is argued with so much. The question underneath is what it would take to do it again.

The collected talks of Charlie Munger, Warren Buffett's partner at Berkshire Hathaway, edited by Peter Kaufman. Munger's subject is how to think: a latticework of mental models borrowed from psychology, physics, biology and mathematics, applied to investing and to living. The centrepiece is the talk on the psychology of human misjudgment — twenty-five tendencies that lead smart people to do dumb things. The rest is aphorism, anecdote and a running case for multidisciplinary common sense. Stripe Press reissued the book in 2023, a few weeks before Munger died at ninety-nine.

A history of personal computing told through J.C.R. Licklider, the psychologist who ran ARPA's computing office in the early sixties and used its budget to fund the people who would build time-sharing, interactive graphics and the ARPANET. Waldrop follows the whole community — Engelbart, McCarthy, Taylor, Kay, Xerox PARC — but the through-line is Licklider's 1960 idea of man-computer symbiosis, and how a government programme manager with taste turned it into an industry. Long, patient and the best single account of where the modern computer came from.

Frank Oppenheimer was the younger brother of the man who ran Los Alamos, a physicist in his own right, and a Communist Party member in the thirties. Blacklisted in 1949, he spent a decade ranching in Colorado and teaching high-school science before founding the Exploratorium in San Francisco in 1969. K.C. Cole, who worked for him, writes about the museum as an argument: that science is something you do with your hands, that exhibits should be built by artists and tinkerers, and that curiosity is the point. A biography of an institution as much as a man.

Richard Hamming's lectures from the course he taught at the Naval Postgraduate School, reissued by Stripe Press. The subtitle is the thesis: the content of the course — coding theory, digital filters, simulation — is a vehicle for teaching how to learn, and how to pick problems worth a career. The final chapter, You and Your Research, is the famous one: why some scientists do important work and most don't, and what to do about it. Opinionated, occasionally cranky, and the rare technical book that is really about style.

Simon Winchester's history of the Pacific since 1950, told in ten episodes: the Bikini atomic tests, the Sony transistor radio, surfing, the typhoon, the coral reef, the Korean peninsula, the rise of China as a naval power. Each chapter is a set piece, and together they make the case that the centre of the world has moved from the Atlantic to an ocean most of its neighbours still think of as an edge.

Rick Perlstein's account of America between Lyndon Johnson's landslide in 1964 and Richard Nixon's in 1972 — how a country that seemed to agree on almost everything split into two that agreed on nothing, and how Nixon, who had spent his life resenting the people who looked down on him, made that resentment a political method. Enormous, vivid and written like a novel. The cast runs from Reagan and Buckley to the Weathermen and the Chicago police.

Stewart Brand, founder of the Whole Earth Catalog, on what a pragmatic environmentalism would look like: cities are green, nuclear power is green, genetically engineered crops are green, and geoengineering will have to be on the table. Each chapter is an argument with the movement he helped start. The book opens with the line he has been repeating since 1968 — we are as gods and have to get good at it.

Nadia Eghbal spent two years talking to open-source maintainers and came back with a description of how the software that runs everything is actually made: not by communities but, increasingly, by one or two overworked people fielding an endless stream of demands from strangers. The book reframes open source as a problem of attention rather than code, and looks at what GitHub, sponsorship and the platform economy have done to it.

Donald Braben ran BP's Venture Research programme from 1980 to 1990, funding scientists on the strength of their questions rather than their proposals and without peer review. Several of the programmes it backed went on to transformative results. His argument is that the system of competitive, peer-reviewed grants that replaced this kind of patronage after the 1970s is very good at incremental work and structurally incapable of the other kind — and that a small number of researchers, freed from it, would be enough to change that. Reissued by Stripe Press in 2020.

The future promised in the 1960s — flying cars, cheap nuclear power, nanotechnology — did not arrive, and J. Storrs Hall wants to know why. His answer is that energy use per person, which had grown steadily for two centuries, flattened out around 1970, and that the flattening was a choice: regulation, litigation and a loss of nerve. The second half is a survey of what the technology could still do, written by an engineer who has clearly been waiting a long time.

Tyler Cowen's short argument that sustained economic growth is a moral imperative: compounding is the most powerful force in the world, so a slightly higher growth rate, held for a century, does more for human welfare than almost any redistribution. Growth is bounded by a small set of rights that should not be traded away, and by a refusal to discount the future. The rest follows. A hundred and sixty pages, and most of the interesting disagreements with it are about the first ten.

A wordless graphic novel. A man leaves his family and his country for a city where he cannot read the signs, the animals are strange and the food is unfamiliar, and slowly makes a life there. Shaun Tan drew the whole thing in sepia pencil, in the register of old photographs, so that an entirely invented world reads as memory. The absence of words is the point: it is what arriving feels like.

The standard English selection of Jorge Luis Borges: twenty-three stories, ten essays and a handful of parables. The Library of Babel, The Garden of Forking Paths, Pierre Menard, Funes the Memorious, Tlön, Uqbar, Orbis Tertius. Most of the stories are short, and most of them contain an infinity. Borges wrote about libraries, mirrors, maps, labyrinths and forgeries because they were the cheapest way he knew to build a whole universe in eight pages.

Jordan Mechner kept a journal while making Prince of Persia at Broderbund between 1985 and 1993 — the rotoscoping of his brother running and jumping in a car park, the deadlines, the doubt about whether to stay in games or go to film school. Stripe Press published the journals in a new edition in 2020. It is one of the few honest records of what shipping something looks like from the inside, written before anyone knew how it would turn out.

Will Larson's handbook for engineering managers, organised around the problems that actually come up: how big a team should be, when to reorganise, how to run a migration, how to hire and how to decide what to work on. It reads like a set of operating notes from someone who has done the job at several companies and kept track of what worked.

Vannevar Bush's memoir of running American science during the Second World War. As head of the Office of Scientific Research and Development he oversaw radar, the proximity fuze, penicillin production and the early Manhattan Project, and afterwards wrote the report that created the modern research university. The book is less about the science than about getting things done inside large organisations — how to work with generals, presidents and committees without being eaten by them. Reissued by Stripe Press in 2022.

Elad Gil's guide to the part of a startup's life that comes after product-market fit: hiring executives, managing a board, raising late-stage money, buying companies, and turning a team of thirty into a company of a thousand. Interspersed with interviews — Reid Hoffman, Marc Andreessen, Patrick Collison, Claire Hughes Johnson — and written as reference rather than narrative.

Douglas Hofstadter's book about how meaning and selfhood can arise from formal systems that have neither. Gödel's incompleteness theorem, Escher's impossible staircases and Bach's fugues are the recurring examples of strange loops — structures that reach up through their own levels and refer back to themselves. Chapters alternate with dialogues between Achilles and the Tortoise that enact what the chapters explain. Won the Pulitzer in 1980 and has been converting teenagers into computer scientists ever since.

Thomas Kuhn's account of how science actually changes: long stretches of normal science, in which a community works inside a shared paradigm, punctuated by crises in which anomalies pile up until a new paradigm replaces the old one. The book gave the word paradigm its modern meaning and started an argument about whether scientific progress is cumulative that has not finished. Short, clear and much more careful than its reputation.