Mark Pontin — Writer, Editor, & Consultant

Kim’s H-Bomb?

When Pyongyang claimed its first H-bomb test in 2016, many scoffed at the small blast. But no nation-state that’s seriously tried has ever failed to build a basic fission bomb. The real leap is to thermonuclear and boosted fission weapons: far harder, but the only path to warheads physically small enough to place atop an ICBM. In that context, a small yield may even be evidence of engineering precision—the North Koreans wasting as little tritium as possible.

TIA Lives On

In this 2006 piece, I reported what Edward Snowden would prove in 2013: NSA had to be mirroring telecom carriers’ fiber-optic switches and harvesting everyone’s data before they could surveil any specific individual targets. Also, when Congress killed the Total Information Awareness program, it killed only the privacy-protecting data-anonymizing technology, Genisys, moving the rest to NSA, then to private cutouts such as Palantir, for further deniability.

Removing the police from Joshua Epstein’s computer model of ethnic conflict, and lowering one group’s perceived legitimacy by just 20 percent, is enough to trigger ethnic cleansing every time. Add police back, and the genocide can still happen—just slower. Epstein’s simple rule-based agents, initially built to simulate a hunter-gatherer economy, can reproduce real historical patterns with precision: run against the actual Anasazi landscape, his artificial settlers colonized the site where the real civilization had once stood.

Project Orion was Cold War America’s most Faustian technological scheme: a spaceship driven by detonating nuclear bombs behind it, meant to carry forty astronauts around the solar system by 1970. Arthur C. Clarke, asked what he’d never expected, replied: “That we would go to the moon, then stop.” But one big reason for that is the inherently limiting physics of chemical rockets, which burn 16 tons of fuel per orbited ton and 1,000 for a Lunar round trip — a dead end, which Orion at least tried to vault over.

SpaceX promises to cut launch costs tenfold, to $1,000 per kilogram by 2010—yet its Falcon 1 rocket keeps failing, most recently catching fire and dying 30 seconds after liftoff. The optimism behind NASA’s 2006 push toward commercial spaceflight rests on a shaky assumption: that cheaper fuel and lighter materials will drive down costs, when the real expense is labor, which eats 60 percent or more of any rocket program’s budget regardless of launch frequency.

Two British subs penned in the whole Argentinian navy during the Falklands War, while a single Exocet missile proved a “slow-moving capital ship” could burn to the waterline. John Arquilla, a Pentagon advisor and advocate of the doctrine of military “transformation,” argues the lesson still hasn’t landed: in 2006, the US Navy keeps building supercarriers bigger than the Nimitz class and insists on fighting against enemies armed with smart, cheap, long-range weapons.

At the National Atomic Museum in Albuquerque, a placard next to a 16th-century tapestry of war elephants compares them to nuclear bombs. Both, we’re supposed to believe, were weapons that “society always placed limits on.” The comparison is dubious and dishonest: dubious because elephants saw centuries of unrestrained battlefield use before and after Rome disarmed Carthage of them; dishonest because its purpose is to pretend that nuclear weapons’ vast destructive powers are ‘thinkable’ in any conventional frame at all.

In 2006 I got asked to profile UC Berkeley’s Nuclear Engineering Department and a genuinely new reactor design, the Economic Simplified Boiling Water Reactor. It had no pumps, no emergency generators, almost no moving parts beyond its control rods — gravity alone drove the cooling. Engineer Per Peterson’s team spent 15 years perfecting the design and it was licensed — and then never built in the US. It took Canada, breaking ground in 2022, to actually build its smaller descendant, the BWRX-300, which will start supplying power in 2030.

Critics call human genetic enhancement playing God; Francis Fukuyama warns that it could fracture humanity into genetic haves and have-nots. SF writer Bruce Sterling’s rejoinder is that any superbaby will be only a beta release, “a hack job” obsolete by adulthood, anyway. Gregory Stock proposes technology to sidestep that problem: an add-on human artificial chromosome, or HAC, upgradable at any time for version 2.0., which leaves that baby’s natural genes alone.

In the 1950s, Seymour Benzer breeds mutant fruit flies whose internal clocks, mating rituals, and memory falter or disappear depending on which single gene he’s disabled—early proof that genes, not just environment, write behavior. A half-century, biotech is trying to monetize that insight: a new drug now costs $500 million and takes 15 years to get to market, and yet 140 are moving through the pipeline anyway.

Frankenfoods Redux

By 2002, nearly a quarter of American corn was genetically modified. And yet proponents’ insect- and herbicide-resistant crops had made Monsanto only a few hundred million a year, nowhere near the world-conquering business its critics had feared—even as GM crops proved not to be unstoppable “superweeds,” but engineered dependents that couldn’t survive without the Monsanto products they were bred to require.

Jeremy Rifkin’s The Biotech Century compares the race to patent the human genome to Europe’s land enclosures. E.O. Wilson‘s Consilience argues that our religious impulses are genetically derived: primate dominance structures which humans, never satisfied with “raw apish feeling,” extended into abstraction, inventing deities — “hyperdominant if invisible members of the human group.”

Inject additional testosterone into a low-ranking male monkey and he doesn’t turn alpha but just gets meaner to monkeys below him, as the hormone raises aggression only where dominance struggles already exist, not where status is fixed. That’s one neurobiologically rooted view of primate sexual behavior. Another, cruder one imagines cheap paternity testing ending marriage as we know it, as men pay women only for sex, and women pay men only for sperm.

In 2000, as dot-coms rushed to fix healthcare’s paperwork with the internet, one real-world fact undercut the optimism: Blue Cross and Blue Shield of Massachusetts employed more staff to administer coverage for 2.5 million New Englanders than Canada employed to run single-payer care for 27 million people. The real bottleneck wasn’t technology and prices weren’t set by patients, but by HMOs and employers.

At 12, Walter Pitts hid from street toughs in a Detroit library and emerged three days later having taught himself Bertrand Russell’s Principia Mathematica. Talking Nets gathers the oral history of the pioneers who followed him into neural networks, among them a young Geoff Hinton, whose stubborn faith in backpropagation helped lay the groundwork for today’s LLMs after two decades in which the field was starved of funding.

A manned spaceship powered by nuclear bombs, colonized comets, trees genetically reprogrammed to grow a hundred miles tall — Freeman Dyson’s visions make Silicon Valley’s virtual malls look small. In Imagined Worlds, the scientist who reconciled quantum mechanics’ mathematics turns to civilizations engineering the universe itself, insisting: “For any speculation which does not at first glance look crazy, there is no hope.”

At a converted glider factory in St. Paul, a nondescript WWII veteran named Seymour Cray began tinkering and ended up building 50 of the Western world’s 70 supercomputers by 1983. The Supermen traces his machines’ escalating speed, from a 1951 computer strung together with fishing line to the Cray-2, a metal tank of processors cooled by boiling fluorocarbon — a Cold War achievement Cray himself later gave away for nothing.

Norbert Weiner: Father of Cybernetics

Short, stout, myopic, and so socially maladroit he could pick his nose mid-lecture — but by 14, Norbert Wiener had already graduated college. He invented the science of cybernetics, taught WWII anti-aircraft guns to predict a plane’s flight path, and warned early on that automation carried real social costs. Decades on, his mid-century worries about intelligent machines now look less eccentric than prescient.

ENIAC had 1.8 billion chances of failing each second, and still beat Howard Aiken’s Mark I a thousandfold. Yet its creators, Mauchly and Eckert, hadn’t even heard of Babbage, let alone Alan Turing. John von Neumann knew Turing had proved that a machine that could read and write symbols on an infinite tape could simulate any other machine whatsoever. He wanted to build a Turing machine that ran at electronic speeds and walked off with the credit.

Danny Hillis has built disk arrays, forgery-proof currency, a clock designed to run 10,000 years, a working computer made entirely of Tinkertoys, and pioneered massively parallel computing at Thinking Machines. Now, at his secretive new venture Applied Minds, he argues intelligent machines won’t be engineered at all — we’ll evolve them, understanding how minds work only after we’ve built ones that do.

“Our fundamental response is: hogwash. We look out there like kids in a candy store,” scoffed Microsoft’s Steve Ballmer at Nicholas Carr, the analyst who declared IT a mere commodity in “IT Doesn’t Matter.” The Big Switch doubles down, predicting computing will become a utility piped in like electricity — while warning that the shift could gut IT jobs and fracture civil society along the way.

Einstein dismissed it as “spooky action at a distance.” Now Caltech physicists have manipulated entanglement itself, halting photons within clouds of cesium atoms so as to build the first working nodes of a quantum network. Between teleported photon states and superconducting loops, the pieces for a quantum internet — and uncounterfeitable money — may be starting to fall into place.

A single square centimeter of light hitting your retina carries more than a million points of data, all arriving together — your optic nerve’s data rate rivals a hard drive’s. Mind at Light Speed argues silicon can’t keep pace: as transistors shrink to single electrons by 2020, David Nolte foresees computers made from photons instead, which will rebuild themselves at light speed, nanosecond by nanosecond.

“Software is a gas — it expands to fill its container,” warns Microsoft’s CTO, describing the arms race between bloated code and ever-hungrier chips. The Invisible Computer argues the fix is networked, smart appliances. We Were Burning is the story of the Japanese engineers who, two decades ahead of Silicon Valley, created the consumer electronics business.

At 17, unable to write fast enough to keep up in class, Konosuke Matsushita dropped out — orphaned, penniless, and destined to build a $65 billion empire. Matsushita Leadership traces his rise; Do Lunch or Be Lunch supplies the theory: since most deals repeat, keeping your word beats grabbing the money and running — game theory’s case for why good guys can finish first.

“Buy a karaoke cassette of ‘Sing a Song for the Party’ and you’ll get a free picture of Chairman Mao” — a Beijing music store’s pitch captures the surreal hybrid that is modern China. In a culture where vacuum-cleaner ads quote the Great Helmsman and insecticide slogans borrow his war cries, slackers need to be supremely determined.

Surgical vans wait outside China’s labor camps as prisoners are executed, then rush away with harvested organs — one of the “creepier benefits” as the laogai‘s commandants privatize profit-sharing. Harry Wu, twice imprisoned for reporting on it, maintains capitalism and democracy aren’t the same thing: newfound Chinese markets, he argues, are producing a pitiless hybrid of totalitarianism and profit.

“Two hundred million children are sleeping in the streets tonight. None are Cuban” — the propaganda line that captures globalization’s uneasy conscience. The Lexus and the Olive Tree casts America as benign free-market hegemon; Juárez answers with photos from the border’s maquiladoras, where workers vanish into the desert and free trade’s triumph looks like a laboratory for a horror-show future.

Two businessmen shake hands in an airport lounge; a mile away, armed guards separate starving people from food. Manuel Castells spent twelve years and three volumes mapping the global “network society,” where informational capitalism excludes whole populations into what he calls the “Black Holes” of the new economy — documenting a new world order that both transcends nations and abandons them.

Cathedrals, pyramids, Moon landings, the Internet: humans require large, visionary projects. From Apollo to SAGE’s continent-spanning radar to the Atlas missile program, Cold War megaprojects created Silicon Valley. Free market advocates must face the fact that the Internet’s bottom-up growth was enabled by ARPANET’s decentralized design, bequeathed to us by government planners and funded by federal monies.

“By 2025, Greenland’s main export will be legends and fairy tales” — so predicts The Dream Society, imagining CEOs as storytellers-in-chief peddling myth over strategy. The Entertainment Economy tracks the same shift already underway: Los Angeles’s defense assembly lines replaced by soundstages, consumers everywhere paying premium not for the product, but for the story wrapped around it.

“Is this building really necessary?” — the question E-topia insists we’ll soon ask of everything, as wall-size smart Sheetrock replaces whole structures with electronic substitutes. Celebration, U.S.A. looks inside Disney’s real-world utopian town, where front porches went unused, construction was shoddy, and the residents who lived there planned to leave, while dreaming instead of a town merely “a lot like Celebration.”

In 2005 a young woman let her dog foul a South Korean subway car and refused to clean it up. Someone snapped a photo; within days she was known to millions as gae-ttong-nyue — dog poop girl — and dropped out of her university. Daniel Solove’s Future of Reputation argues that the internet has revived village-level social punishment, on a scale villages never had.

Joel Kotkin coins names for the digital economy’s winners: “nerdistans,” the high-end suburbs untouched by commuter traffic, and “Valhallas,” scenic rural retreats like Vail where the rich keep ranch estates. The losers get no such poetry — just old manufacturing cities and rural towns that experience decline and every more degraded infrastructure.

In a legendary hand of liar’s poker, Salomon Brothers’ chairman proposed a $1 million bet, no tears. John Meriwether shot back: “I’d rather play for real money — $10 million, no tears.” He walked away and later founded LTCM, a hedge fund staffed with physicists and two Nobel laureates that leveraged $1.3 trillion in derivatives and nearly broke the global economy.

With the global currency market now exceeding the combined GDP of every major economy, governments get regular lessons in how little control they actually retain over money. William Greider fears where this leads; forty essayists in The Case Against the Global Economy want to stop it outright. Future currencies, one contributor speculates, may form around shared identity rather than geography

Being the Aztec equivalent of a good Rotarian demanded paying for a human sacrifice that cost as much as 40 woven cloaks. The victim’s still-beating heart was then ripped out by an appropriately pious priest. Those corpses that the merchants didn’t sponsor had their choicest cuts sold for chocolate beans. Jack Weatherford traces the history of money from ancient times to today’s electronic currencies.

It’s not just the economy, stupid. Niall Ferguson spends 380 pages proving the opposite of America’s favorite political cliché: there’s no free market without an effective state, economic growth doesn’t guarantee democracy, and — as Al Gore learned the hard way — voters don’t reliably reward good times. Money financed wars and empires, but power dictates to money as much as money dictates back.

When Margaret Thatcher was still just Britain’s opposition leader, she held up high economist Friedrich Hayek’s book and declared, “This is what we believe.” The Commanding Heights traces how that conviction conquered the world — while Paradise Lost shows what it cost California, where voters gutted their own taxes in 1978 so that today Californians are still living in the wreckage. 

A Future Perfect argues that we resemble the Victorians more than we admit, then dares ask globalization’s rudest question: isn’t it the developing world’s turn? Even Keynes abandoned free trade for protectionism once, before Bretton Woods brought him back. If poorer nations can do our jobs more efficiently today, our real objection was never economic, only that we don’t want to lose.      

A Korean businessman changes his first name to Drucker. A billboard in downtown Beirut, amid pockmarked buildings and Hezbollah’s yellow banners, summons the faithful to a Total Quality Management seminar. In an Asian newspaper, a full-page advertisment announces “the world’s most competitive man”—management guru Michael Porter is coming to town. Management theory is big business.

Peter Drucker fled Nazi Germany after they burned his pamphlet, got banned from Westinghouse’s offices as “some kind of radical,” and was fired from Sarah Lawrence for refusing to sign a Communist condemnation. He went on to invent management theory outright. As one editor puts it: reading Drucker is depressing, because he’s as good as this stuff gets.

Peter Drucker, aged 91, casually claimed to me that Franz Kafka was a leading expert on workers’ compensation in Vienna, then that he himself was the sole survivor of an Eisenhower health-care commission no one else now recalls. He also personally knew Freud, Keynes, Buckminster Fuller, and McLuhan. Whether or nor every fact holds up, the perceptions do.

Asked whether he’s “a dishonest, scheming little shit,” Michael Wolff doesn’t hesitate: “Oh yes, that’s exactly what I am.” His memoir of the dot-com gold rush skewers nearly everyone he met — a financier described as “a wire-thin, almost feminized JFK Jr. look-alike,” Wired’s founder as a “lost soul” chasing his own myth. Nobody has sued him yet, at least.

Paul Krugman’s teenage fantasy was to become a psychohistorian, predicting society’s future like Asimov’s Hari Seldon did. Efficient-markets theory assumes something just as unrealistic: that though a third of Americans believe the government talks to aliens, rational actors run the economy. After Robert Shiller talked to the Fed, Greenspan accepted the idea of market irrationality.

Forget programming — teach your kid to cut hair. Krugman’s contrarian advice: today’s safest professionals, from economists to coders, may be this century’s version of 19th-century hand-weavers, whose wages soared right up till the arrival of the machines. What will actually survive automation, he argues, will be the jobs a robot still can’t do — haircuts, housecleaning — because they require human hands.

A colleague once called Georges Doriot “the most schizophrenic Frenchman I’ve ever met” — devoted to wine and French sophistication, yet driven by a pro-entrepreneurial zeal most Americans couldn’t match. In 1957 he staked $70,000 on a fledgling company called Digital Equipment Corporation. It returned over $400 million. He’d just invented modern venture capital.

Benchmark’s venture partners average six-foot-five and $100 million in new net worth a year, but Randall Stross’s eBoys can’t decide if it’s profiling geniuses or frat boys — its opening pages read like a locker room. Better is Done Deals, an oral history where VCs like Bob Kagle actually sound wise, flatly declaring biotech off-limits with a slide showing the medical caduceus crossed out.

How many surrealists does it take to screw in a lightbulb? Three — and Robert Reid’s survey of the Web’s founding entrepreneurs opens with a 32-page introduction nearly that absurd, so overwrought it reads like self-parody. Strip away the hype, however, and the profiles — Marc Andreessen, Halsey Minor, Mark Pesce among them — hold up as sharp, useful journalism.

Cyberspace was never destined to be free — it’s built from code, which does whatever it’s designed to do. Government and commerce, Lawrence Lessig warns, are building a surveillance state. Neal Stephenson, meanwhile, sees operating systems as car dealerships: Windows sells station wagons, Apple sedans, and Linux just hands out tanks for free.

Two CEOs, two very different origin stories. Larry Ellison talked his way from being an unwanted Brooklyn teenager into Silicon Valley’s fifth-richest man. Andy Grove escaped the Nazis, scavenged after the Soviets invaded, and runs Intel by demanding two lawsuits a quarter; one ex-employee lives armed, terrified that Intel may try to kill him.

Why is the night sky dark? By pure logic it shouldn’t be — an infinite, static universe should blaze white with the light of an infinite number of stars. That it doesn’t is proof the cosmos is expanding, finite, and young. In 1998 astronomers found something stranger still: the expansion is accelerating, reviving the very idea Einstein once called his “greatest blunder.”

As China grabs the headlines with its taikonauts, India is quietly designing something stranger: Avatar, a reusable scramjet spaceplane that takes off from a runway, skims its own oxygen from the air mid-flight, and could reach orbit for $67 a kilogram — against $4,300 for a Russian rocket launch. Yet some people suspect, alternatively, that it’s really just a missile in disguise.

Cell phones allowed India and China to leapfrog landlines straight into the information age. K.R. Sridhar wants electricity to do the same — junking Edison’s century-old grid for minivan-sized fuel cells generating power on-site. And he’s also building an oxygen generator for Mars, eight inches tall, to enable humans to live off the red planet’s own atmosphere.

Heat, it turns out, behaves like light — a wave-particle duality Einstein anticipated but nobody could exploit until now. Arunava Majumdar explains how stacking materials 20 molecules thick could yield super-insulators, super-conductors, and batteries the size of a lighter cartridge which will be able to outlast today’s by a factor of ten.

At fourteen, plowing an Idaho field, Philo Farnsworth looks at the furrows and sees rows of electrons scanning a picture — the idea that becomes television. He’s self-taught, obsessive, and about to be crushed. RCA’s David Sarnoff wants the technology without the man, and the eccentric loners who actually invented TV lose everything but the credit.

The celebration of black athletic prowess in America is derived from the same pseudoscience that white Americans once used to brand black people as mentally inferior. Simultaneously, today’s “cult of black athleticism” quietly fuels anti-intellectualism among black youth. So at least runs the argument of one white college professor. Could he be right?

A future of AI agents reshaping e-commerce, negotiating on our behalf, screening telemarketers, and personalizing products in real time was already visible in 1998. So was the cost — personal data becoming currency, privacy turning into a luxury good, biometric surveillance tracking your every move, and identity itself up for sale.