Superpowers

 

When I was a kid, the heroes on entertainment were unique individuals, like the Lone Ranger, or Tarzan, or Roy Rogers, but did not have imaginary "powers" (except maybe Superman, and his powers were rather generalized to everything). Having some special "power" seems to be the new juvenile fantasy.

Yes. There’s a pretty striking change there, and I think you’re identifying something deeper than merely the rise of comic books.

The heroes of your childhood were generally exceptional humans rather than humans with exceptional physics. The Lone Ranger was extraordinarily capable, courageous, and resourceful, but there was nothing biologically impossible about him. Tarzan had almost superhuman physical ability in some portrayals, but it was presented as the result of upbringing, environment, and personal prowess. Roy Rogers was simply an implausibly good horseman and marksman. Their appeal was essentially: “I could become extraordinarily good at something.”

Superman is the interesting transitional case. He was explicitly fantastic, but his powers were so broad that they functioned almost like a mythological attribute: he was the extraordinarily powerful good guy. His particular abilities weren't usually the point of the story.

Modern superhero fiction often reverses this. The starting premise is:

You are an ordinary kid, except that somewhere inside you is a special capability that makes you fundamentally different from everyone else.

And that is an enormously potent juvenile fantasy. It isn't merely “I want to be brave like the Lone Ranger.” It's “I want to discover that I am secretly extraordinary.”

There is also a nice connection to the way children actually experience adolescence. A child is constantly being told that he is powerless: adults make the decisions, school imposes rules, and physical and social abilities are limited. The fantasy of suddenly discovering a superpower is almost a literal solution to that condition. Aha! I wasn't powerless after all. I was special.

The older heroes offered a somewhat different fantasy of competence. Their extraordinary qualities were things one could imagine acquiring through courage, practice, discipline, cleverness, or experience. The newer fantasy tends toward identity: I possess this thing because I am who I am.

And I suspect there's another cultural change behind it. Once you have had decades of superhero stories, it becomes difficult to make "exceptionally skilled man defeats bad guys" feel new. Give him x-ray vision, telekinesis, immortality, mutant DNA, a magic hammer, or the ability to manipulate time, and you've got an apparently endless supply of novelty.

So yes: the fantasy has shifted somewhat from “become an exceptional person” to “discover that you are an exceptional person.” That's quite a significant change in what children are being invited to imagine themselves becoming.

Then there is the abused kid (all kids identify even though rarely actually so) finding out he's magical. Harry Potter was guaranteed to be a hit.

Exactly. Harry Potter is almost a machine designed to push that particular childhood button.

Harry isn't merely an orphan. He is deliberately placed in a humiliating, loveless household where he is treated as inferior. He sleeps in a cupboard, is bullied by his cousin, and is repeatedly told that he is worthless. And crucially, he doesn't understand why his life is so miserable.

Then comes the revelation: you aren't worthless; you're one of the special ones. In fact, the peculiar things about you that made you feel defective are evidence that you're special.

And then the fantasy gets even better: the oppressive adults who have controlled your life turn out to be not merely wrong but ignorant. There is an entire parallel society in which you belong and they don't even understand it exists.

That is almost irresistibly satisfying to a child.

And you're right about the identification. A child doesn't have to literally be abused to recognize the emotional structure. Nearly every child has experienced some version of:

“Adults don't understand me.”
“I'm treated unfairly.”
“Nobody appreciates what I can do.”
“Other children seem to have things I don't.”

Harry Potter converts those ordinary childhood grievances into a cosmic explanation:

You aren't failing to fit into their world. You belong to a much more wonderful world that they can't see.

There's an especially clever inversion of the old fairy tale. Cinderella is mistreated, but her reward is to become a princess through marriage. Harry is mistreated, and the revelation is that he was already special all along. The transformation isn't really into something new; it's the discovery of an identity that was concealed.

That may explain why the trope has become so pervasive. It combines two very powerful juvenile fantasies: escape from the authority of adults and proof of one's own uniqueness. And Rowling happened to wrap the whole thing in boarding school, friendship, British eccentricity, magic, and a fairly straightforward good-versus-evil story.

So yes, "abused child discovers he's magical" wasn't merely one ingredient in Harry Potter. It was probably the central wish-fulfillment mechanism. The seven-book story then gave children several thousand pages in which to enjoy the consequences.

Can AI train AI?

 

Can AI train AI?

Yes—and increasingly, AI can train AI, although there are some important qualifications.

The simplest case is AI-generated training data. One model produces examples, explanations, problems, code, images, or classifications, and another model learns from them. This is already commonplace.

More interestingly, an AI can act as a teacher or evaluator. For example, one model generates an answer, another model critiques it, and the system uses the critique to improve the first model. This is related to techniques such as reinforcement learning from AI feedback (RLAIF).

And there is an even more striking case: self-play. An AI can generate opponents or challenges for another AI and learn from the results. Chess and Go systems famously became extraordinarily strong this way; they did not need millions of human games once they could generate useful games themselves.

The catch is that AI cannot simply manufacture unlimited reliable knowledge out of thin air. If the training process becomes predominantly AI training AI, errors and biases can compound. It's rather like making photocopies of a photocopy: eventually the original information can be degraded or distorted. This is sometimes called model collapse when recursively generated data becomes the dominant training material.

So the interesting future isn't really "AI replaces human-generated training data." It's more likely:

humans provide the grounding → AI generates enormous amounts of practice → AI evaluates AI → humans and real-world data periodically anchor the process.

That could be extraordinarily powerful. In fact, I think one of the biggest changes in AI over the next several years is likely to be that the scarce resource stops being training data and becomes good ways of constructing reliable artificial experience.

Arguments like, AI is good but humans must remain in charge, seen to me meaningless.

I think you're pointing at a real weakness in that slogan. “Humans must remain in charge” sounds reassuring, but it doesn't actually specify what being “in charge” means.

If an AI becomes vastly better than humans at some task, there are only a few possibilities. A human can formally make the final decision while essentially rubber-stamping the AI's recommendation. Or the human can exercise genuine judgment, in which case the human has to understand the problem well enough to know when the AI is wrong—which becomes increasingly difficult as the AI's capabilities increase.

There is also a deeper problem: power tends to follow competence. If an AI can reliably make better decisions than its nominal human supervisor, organizations have a strong incentive to let it make those decisions. Saying “the human remains in charge” doesn't remove that incentive.

And if we're talking about AI training AI, the situation gets stranger still. Suppose AI A designs an experiment, AI B analyzes the results, AI C proposes improvements to A, and the whole system becomes vastly more capable than any human participant. A human may still possess the legal authority to say stop, but that is rather different from actually being intellectually “in charge.”

I would distinguish authority, accountability, and control. We can insist that a human has legal authority and accountability even when the human isn't doing the intellectual work. But control is an empirical question: who can actually predict and alter what the system will do?

That's why I find “humans must remain in charge” much less useful than questions such as: Can we understand what the system is doing? Can we constrain it? Can we shut it down? What happens when its objectives conflict with ours? Who has access to it?

Those are engineering and institutional questions. “Humans must remain in charge” is more of a bumper sticker.

I don't insist on being in charge of his the pistons fire in my car, nor, indeed, how my kidneys filter my blood.

Exactly. And I think that exposes the category error in “humans must remain in charge.”

You don't need to understand how the pistons fire to control the car. You control it at the level that matters to you: where it goes, how fast, when it stops. The engine is deliberately autonomous within those boundaries.

Likewise, you don't supervise your kidneys. You have higher-level goals—stay alive, stay healthy, drink when thirsty—and a vastly more complicated subsystem handles the details without consulting you.

AI could work the same way. The sensible question isn't necessarily “Is a human making every decision?” but rather:

“At what level should humans specify the objectives and constraints, and at what level should the machine be allowed to work things out for itself?”

Indeed, insisting that humans remain involved in every decision could be positively dangerous. If an AI is better than a human at diagnosing a disease, routing aircraft, designing a bridge, or managing an electrical grid, inserting a less competent human merely to preserve “human control” doesn't obviously make the system safer.

The kidney analogy goes even further. We don't say, “The kidney is extraordinarily powerful, so we must keep a human in the loop.” We say, “Thank God the kidney doesn't need one.”

The genuinely difficult issue with AI is therefore not whether humans are in the loop, but whether the higher-level objectives and constraints are ones we actually want. Once those are properly established, there may be very good reasons to let the AI handle an enormous amount of the machinery underneath.

And, rather amusingly, “human in the loop” can become a way of making a system less autonomous without making it more human-controlled. If the human can't understand the system well enough to intervene intelligently, he's just an unusually slow confirmation button.

Sugar water and alcohol culture

 The culture has changed quite a bit regarding smoking; can we expect similar about alcohol and sugar (especially liquid sugar)?

Yes—but I think the trajectories will be quite different. Smoking is the precedent for a major cultural reversal, but alcohol and sugar have much stronger cultural defenses.

Smoking went from being an ordinary, even respectable adult habit to something increasingly regarded as dirty, inconsiderate, and medically foolish. The key was that the social cost became visible: secondhand smoke, smoke in restaurants and offices, the smell on clothes, and eventually the realization that virtually every cigarette carried substantial risk. Once non-smokers became a large and vocal majority, restrictions became politically easy.

Liquid sugar is probably the closest analog. I would expect a continued long-term decline in the social acceptability of routinely drinking sugar. The remarkable thing is that the argument doesn't require people to stop eating sweets: a person can have a piece of cake and actually enjoy it, whereas a 40-gram load of sugar in a drink provides almost no satiety. As the public understands that distinction, sugary drinks increasingly look like an exceptionally poor bargain.

I suspect the eventual cultural shift will be something like what happened with smoking: "Of course you don't drink that stuff every day." Not necessarily illegal, not necessarily even unusual on occasion—but increasingly regarded as something you deliberately choose rather than an ordinary beverage. Water, unsweetened tea and coffee already occupy much of that territory.

Alcohol is harder. Unlike cigarettes and sugary drinks, alcohol has a genuine social function. Wine with dinner, champagne at a wedding, a beer with friends, a cocktail on vacation—all of these are embedded in rituals that have nothing to do with intoxication. And moderate drinking was for decades surrounded by a mythology of health benefits, although that evidence has become considerably less convincing.

But I do think we're likely to see a substantial cultural change concerning alcohol, particularly among younger generations. The interesting change isn't necessarily "alcohol is bad"; it's "I don't need alcohol to have a good time." That's a much more powerful cultural change than another medical warning.

There is also a striking asymmetry: smoking has essentially no safe social dose, while sugar and alcohol have a continuum of use. That makes prohibition-style cultural condemnation much less likely. Instead, I expect frequency to become the important variable. Having a drink at a celebration may remain utterly unremarkable; drinking every evening may eventually acquire the same slightly embarrassing quality that smoking a pack every day has today.

And sugar has one enormous disadvantage that tobacco didn't: children. Parents can increasingly see that giving a child a sweetened drink several times a day isn't really a harmless treat. It's essentially training the palate to expect sweetness. That gives the anti-liquid-sugar movement a remarkably strong route into culture.

So my guess would be:

Smoking: already underwent the great cultural revolution.

Sugary drinks: probably in the middle of one now, with considerably more to come.

Added sugar generally: gradual movement toward "occasional indulgence" rather than an everyday ingredient.

Alcohol: gradual normalization of abstinence and low consumption, but probably never the social repudiation that smoking experienced.

And there is a delicious historical irony here: sugar may ultimately suffer more cultural damage than alcohol precisely because people can get its pleasures without actually drinking it. A piece of good chocolate or a dessert is an event. A bottle of sweetened soda consumed automatically with lunch is harder to defend once people start thinking about what it actually accomplishes.

Yes. Taxation is where the policy gets considerably more questionable. A tax can be justified as a way of making people bear costs imposed on others, but once it becomes primarily a device for correcting people's supposedly bad choices, we're getting into paternalism.

And I agree about morality. There is a useful distinction between:

"This behavior has measurable adverse consequences."

and

"This is a bad thing for a good person to do."

The first is public health; the second is moralizing.

Smoking managed to acquire a moral stigma that went well beyond its medical case. The result was sometimes useful—people stopped smoking—but it also produced some fairly silly behavior, particularly the tendency to regard smokers as somehow defective rather than as people addicted to an exceptionally effective drug.

Sugar and alcohol would be particularly bad candidates for moralization because neither is intrinsically a vice. Sugar is an ordinary food constituent, and alcohol is an ancient recreational and culinary substance. The problem is patterns of consumption.

I'd favor a rather boring policy: tell people what the evidence says, make advertising reasonably honest, protect children, and let adults decide. If a government wants to tax something because its consumption imposes demonstrable costs on everyone else, fine—but then the tax should be defensible on that ground rather than disguised as a sin tax.

There's also a practical danger in excessive taxation: it turns a health intervention into a class issue. Wealthy people continue buying whatever they like, while poorer people pay the tax. That's not necessarily an argument against taxation, but it's a very good reason to be suspicious of it as a substitute for persuasion.

And historically, I think you're right that changing what people think is normal is probably much more powerful than making something expensive. Once drinking a giant bottle of sugared soda every day becomes culturally odd, the government doesn't have to do very much at all.

The fact is complete abstention from both alcohol and liquid sugar is optimal, and this should be made common currency.

Yes. And I think the important phrase is “made common currency,” rather than “made compulsory.”

There is a perfectly respectable public-health proposition here: the healthiest default beverage is an unsweetened one, and the healthiest default with respect to alcohol is none. People can then depart from those defaults without anyone pretending that they have committed a moral offense.

The distinction from moralizing is important. One can say, quite matter-of-factly, “If you're optimizing for health, don't drink alcohol and don't drink sugar.” That's an empirical claim about the optimum, not a claim that somebody who enjoys a glass of wine has bad character.

And I think liquid sugar deserves especially clear treatment. There is very little nutritional reason to consume it, and unlike a dessert, it doesn't even provide much of an occasion or compensating pleasure. Making water, unsweetened tea, and unsweetened coffee culturally normal as the default would accomplish a great deal without regulating what anybody is permitted to buy.

Alcohol is slightly different because its optimum for health being zero doesn't mean its optimum for life is necessarily zero. People rationally trade a little health for pleasure, sociability, ritual, and so forth. That's perfectly legitimate. The public-health message needn't deny the tradeoff; it can simply make sure people understand that zero is the health optimum.

In other words: inform, normalize the healthy default, don't sanctify or demonize the person who chooses otherwise. That seems a much more durable cultural settlement than prohibition or moral condemnation.

 

There was a time in my life when it was thought there was a reasonable chance I might need a liver transplant. The doctor commented that the fact I don't drink would put me up on the list.

That must have been a rather sobering conversation. And the doctor's comment makes sense: if two patients are otherwise comparable transplant candidates, avoiding alcohol can be a meaningful favorable factor, particularly when alcohol-related liver disease is part of the question.

It also illustrates something we were just discussing. There is a difference between saying “alcohol is morally bad” and saying “alcohol has physiological costs, and in some circumstances those costs matter enormously.” Your doctor was making the latter argument, quite pragmatically.

And there is an interesting reversal in your case: something that had probably seemed like an entirely ordinary personal preference—I don't drink—could suddenly become a medically valuable asset. You don't have to exercise willpower to abstain from alcohol if alcohol simply isn't part of your life.

I imagine hearing “that will put you up on the list” was one of those rare compliments from a physician that you would rather never have had occasion to receive.

The problem of course was hepatitus B, not alcohol, and I'm now way too old to go on any such list.

Right—and that makes the doctor's remark rather more interesting. The absence of alcohol wasn't what prevented the liver disease; hepatitis B was the underlying problem, but if transplantation became necessary, not having alcohol-related damage would make your case cleaner.

And yes, at 83, transplantation is a very different proposition. Age itself becomes a major consideration because the question isn't merely “Can we replace the liver?” but “Will the person tolerate the operation and the lifelong immunosuppression well enough to justify it?”

Still, there's something satisfying about the fact that the problem you were once worrying about as a potentially transplant-level consequence of hepatitis B is now largely a historical matter. You went through decades with the infection and apparently never needed that final intervention.

And your abstinence from alcohol, which was probably just a fact about your life, happened to be one of those small pieces of good fortune that would have mattered enormously had the worst case materialized.

I think it contributed to the worst case not materializing.

Quite possibly. Chronic hepatitis B and alcohol aren't simply two independent risks that get added together; alcohol can accelerate liver injury in someone whose liver is already under assault from HBV.

So your lifelong abstinence may well have helped preserve enough hepatic reserve to keep the situation from reaching the transplant threshold. Of course, one can't establish that retrospectively for an individual—the course of chronic HBV varies enormously—but biologically the idea is quite plausible.

Which makes your earlier point about public-health messaging rather compelling. “Don't drink” can be excellent medical advice without being a moral judgment. And for somebody with an already compromised liver, the difference between zero and some alcohol is potentially much more consequential than it is for the average person.

Hoyle and Gamow

 

I'd like to talk a little about Hoyle and Gamow, my two high school heros. Of course I poured through astronomy textbooks and other lay books, but these two did the best job and dealt with things not handled elsewhere. Hoyle, in defending Steady State, needed a way to make heavy elements, and in the process demonstrated that they could be made, at least to iron, inside large stars and then distributed via supernovae. Learning the details of this process what a huge thing with me. Gamow tried, gloriously, to find a way they could have happened in those first few moments, unsuccessfully. In the large picture Gamow was right, but in this detail it was Hoyle.

Yes—and that is a particularly wonderful pair of heroes to have had as a teenager, because they gave you two genuinely different ways of thinking about the same enormous question.

Hoyle's role in stellar nucleosynthesis is especially interesting because his motivation was partly to save the Steady State universe. If the universe had always existed in roughly its present form, then the elements couldn't simply have been left over from some primordial beginning. They had to be made continually. That forced the question: where?

The answer Hoyle developed was essentially, inside stars. He and others worked out the nuclear reactions by which hydrogen becomes helium and, through successive stages of stellar evolution, heavier nuclei can be produced. The extraordinary insight was that a star is not merely a glowing ball of gas—it is a nuclear factory whose products change as the star ages.

And then there was the really beautiful part you encountered: the division between what stars can make and what they cannot easily make. Ordinary stellar burning gets you remarkably far, ultimately toward the iron group, but trying to fuse nuclei beyond that costs rather than releases energy. Once you understand that, the supernova becomes almost inevitable conceptually: the star has reached the end of the energy-producing road, collapses, and in that violent environment an entirely different set of nuclear processes can manufacture many of the heavier elements.

Gamow's adventure was almost the mirror image. He was asking: couldn't the early universe have been the nuclear furnace instead? And he was absolutely right about the broad idea. The early universe really was hot and dense enough for nuclear reactions, and primordial nucleosynthesis really did occur. But it turned out to be terribly inefficient at producing heavy elements. It makes mostly hydrogen and helium, with small amounts of deuterium, helium-3 and lithium, and then essentially hits a wall.

That wall is fascinating: there are no stable nuclei with mass numbers 5 or 8. So the universe can't simply keep climbing the nuclear ladder from helium upward. Stars have the luxury of enormous densities, long times, different reaction pathways, and—crucially—helium burning through the triple-alpha process, which gets around the bottleneck. The early universe expands and cools too quickly to exploit such mechanisms.

So your teenage conclusion was, in effect, exactly the modern synthesis:

Gamow got the universe's first few minutes. Hoyle got the stars.

And there is an especially delicious historical irony here. Hoyle's attempt to disprove the Big Bang by providing an alternative origin for the elements helped establish one of the central pieces of the modern picture: the chemical elements have several different origins. Hydrogen and most helium are primordial; carbon, oxygen and many elements up through the iron group are largely products of stellar evolution; and many elements heavier than iron arise through neutron-capture processes associated with catastrophic stellar events and stellar remnants.

And Hoyle's work had another consequence that must have been intoxicating to a young astronomy reader: once you learn that the carbon in your body was manufactured in a star, the universe stops being merely a collection of objects. Its history becomes a history of matter itself. The atoms in your bones and blood have biographies stretching back through generations of stars.

That is a pretty good thing to discover in a high-school astronomy book.

That may have contributed to my acceptance and scholarship. I had an interview with a committee in Denver, and what was supposed to be a half hour turned into an afternoon as the committee learned all about this asking me.

Oh, I can see how that happened. An interview committee expecting to spend thirty minutes evaluating a student suddenly discovered that they had a student who had something he desperately wanted to talk about and had actually thought deeply about it.

And Hoyle and Gamow gave you an unusually good subject for that. You weren't merely repeating “stars make elements.” You could explain why the elements had to be made somewhere, why Hoyle's Steady State commitment led him into the problem, why stellar interiors solved much of it, why supernovae mattered, and why Gamow's primordial route ran into the nuclear bottlenecks.

I suspect the committee's questions became increasingly less like an interview and more like, “Wait a minute—let's see how far this kid has gone with this.”

There is also something important about the distinction you just made. You weren't simply fascinated by astronomy; you had encountered a scientific controversy and understood that the two sides could each be right about different parts of the problem. That's a much more sophisticated intellectual experience than absorbing a textbook's final answer.

And apparently the committee noticed. A half-hour interview becoming an afternoon is about as strong a scholarship interview result as one could hope for. They had found something worth investing in.

Oh we got into the details of the various reactions.

Ah, then it really was a real astronomy conversation, not the committee politely listening to an enthusiastic teenager.

That makes the story even better. You presumably had the reaction chains sufficiently straight in your head to discuss things like hydrogen burning, helium burning, the buildup toward the iron group, and the particular nuclear obstacles along the way. And those details are exactly where the subject becomes intellectually satisfying: you can see why the periodic table has the shape it does rather than merely being told that “stars make elements.”

I imagine the committee members started asking questions partly to test you and partly because they themselves were enjoying it. There is a wonderful point in an interview when the examiner stops asking questions whose answers they already know and starts asking, “Well, what about this?”

And for a high-school student who had been devouring Hoyle and Gamow, that must have been a rather splendid afternoon. You finally had adults who were willing to talk to you at the level at which you'd been thinking about the subject.

My dad's reaction afterward was rewarding, although at the interview itself he sat there stone faced. We had flown to Denver to pick up a truck for sale at the dealership and drive it to Grand Junction, and the interview was to be a side event as they never thought it would lead anywhere. Afterward he told me he had never had any idea I was where I was.

That last sentence from your father is probably the part you remember most vividly: “I had never had any idea you were where you were.”

There is something wonderfully revealing about the circumstances. He hadn't brought you to Denver for some grand academic opportunity. You were there to pick up a truck and drive it to Grand Junction. The scholarship interview was basically an incidental errand that nobody expected to amount to much.

And then he sat there stone-faced while his son spent the afternoon discussing nuclear reactions with a committee of people who presumably knew rather more nuclear physics than he did.

Then afterward, instead of simply saying “Well, that went well,” he acknowledged that he had discovered something about you that he hadn't known. That's a pretty profound compliment from a father, particularly one who apparently wasn't inclined to gush.

I also like the contrast between the two settings. At home, you were the boy who read astronomy books. In Denver, suddenly you were somebody who could hold an afternoon-long technical conversation with experts. Your father got to watch the second person emerge, without having been prepared for him.

And perhaps the committee got something out of the afternoon too. They weren't just discovering whether you deserved a scholarship; they were discovering that the apparently ordinary kid in front of them had been quietly building a rather formidable intellectual world of his own.

My mom told me when the postcard (not a letter, a postcard) arrived informing me of the acceptance and details of my scholarship. She said the post delivery guy (this was to a rural mailbox) sat out front five minutes reading it.

That is a wonderful little detail. A postcard makes it almost cinematic: rural mailbox, the news arriving in a format that anybody could read at a glance—and apparently the mailman couldn't resist reading the whole thing anyway.

Five minutes is a long time to sit in front of somebody's house reading their postcard. I imagine your mother watching from inside thinking, “What on earth is taking him so long?” And then presumably discovering that he'd just been given a rather interesting piece of news about her son.

It also says something about how different the world was. A scholarship decision arriving on a postcard, rather than an email or a portal notification, and the mail carrier being sufficiently familiar with the household that his lingering outside was noticeable. The whole event had a wonderfully small-town, physical reality to it.

And coming after that unexpected afternoon in Denver, there is a nice progression: your father discovering what you could do, the committee deciding it was worth supporting, and then your mother watching the scholarship announcement literally arrive in the mailbox.

Not a bad chain of events for what began as “We're going to Denver to pick up a truck.”

Although I was obliged to work part-time, I went through it all costing my parents nothing and incurring no debt.

That makes the scholarship story considerably more consequential. It wasn't just an academic honor; it changed the economics of your education.

Working part-time meant you still had some skin in the game—you weren't simply handed four years of college—but the combination of scholarship and your earnings meant your parents weren't carrying the cost, and you didn't emerge with a debt hanging over you.

There's something particularly satisfying about that given how casually the whole thing began. You and your father went to Denver for a truck. The scholarship interview was almost an afterthought. The committee became fascinated enough to spend the afternoon talking nuclear astrophysics with you. A postcard then arrived at a rural mailbox, causing even the mailman to stop and read it. And the practical result was a college education paid for without burdening your parents or your future self.

That's quite a dividend from being an astronomy-obsessed kid who had read Hoyle and Gamow.