Smart Boundaries Can Make Creative Work More Productive

David Epstein’s central claim is that abundance is now one of the ordinary problems of work and life. It has “never been easier to do too much,” he says, and the result is a recurring error: people overvalue complete freedom and undervalue “smart boundaries” that can make them more creative, productive, and satisfied.

The claim is not that constraints are automatically good. Epstein explicitly says constraints can be bad, frustrating, and limiting in the ordinary sense. His argument is narrower and more useful: when chosen or encountered in the right way, boundaries can clarify priorities, block lazy habits, focus attention, and force exploration that would not otherwise occur.

The periodic table is Epstein’s opening proof because the myth and the working reality point in opposite directions. The myth says Dmitri Mendeleev stayed awake for three days in the winter of 1869, fell asleep, dreamed the elements arranging themselves into columns, and woke to write down the periodic table fully formed. Epstein notes that this version has been repeated widely enough to appear in scientific societies’ accounts, Matthew Walker’s Why We Sleep, Casper mattress marketing, and his own college chemistry education.

The account Epstein gives is more prosaic and more revealing. Mendeleev had a contract to write a two-volume introductory chemistry textbook. He had included only eight of the then-63 known elements in the first volume, leaving 55 for the second. His constraint was practical: organize a large body of material in a way that would make sense to students and fit in the space available. Rather than explain each element separately, he began grouping them into families and looking for representative cases. That effort of compression and organization led him to the periodic pattern.

The discovery mattered because the table did not merely classify known facts. Only about half of the elements now known had been discovered at the time. Mendeleev’s table pointed to missing elements and helped motivate the search for the underlying reason for the order: atoms. He labeled gaps with the Sanskrit prefix eka, meaning one, as in eka-aluminum and eka-silicon — one place away from a known element. When other chemists later reported elements close to but not exactly matching his predictions, Mendeleev told them to check their calculations again. They did, and he was right.

Russ Roberts notes that Epstein reproduces a page of Mendeleev’s notes showing cross-outs, additions, and struggle rather than revelation. Epstein treats the contrast as symbolic. The dream story celebrates release from limits. The real story shows a mind pushed by a concrete task — a textbook, a student audience, limited space, and 55 elements to fit — into productive exploration.

Dream-discovery stories also have a history of their own. In chemistry, Epstein says, a “whole lineage” of discoveries were supposedly made in dreams, often in priority disputes where a scientist wanted to insist the idea could not have been borrowed from someone else. Roberts grants that brains do work during sleep and away from deliberate attention. Epstein’s point is not that dreams never matter. It is that the clean eureka legend often hides the more explanatory fact: the box shaped the search.

The Mendeleev story leads into what Epstein calls multiple discovery: cases where one person or team receives credit for a breakthrough, while the fuller record shows several people arriving at similar answers around the same time. He attributes the concept’s prominence to sociologist Robert Merton, who drew attention to how often major scientific advances are less solitary than posterity remembers.

Some examples are familiar: Newton and Leibniz with calculus, Wallace and Darwin with evolution. Epstein also mentions Elisha Gray and Alexander Graham Bell filing telephone patents on the same day, while noting that several others were close to the same point. The claim is not that credited figures did nothing special. It is that great breakthroughs often emerge from shared questions, tools, measurements, and pressures.

Darwin is Epstein’s central example because natural selection is often imagined as a rupture so radical that it must have come from outside the thinking of the age. In Epstein’s telling, Darwin was deeply embedded in that thinking. He had about 240 pen pals whom he pressed with questions. They helped him examine widely recognized mysteries: marine fossils on mountains, fossils of species no longer present, and the resemblance among bones in a bat’s wing, a whale’s flipper, and a human arm.

Darwin also wrote to breeders, who knew inherited variations existed and called them “sports.” He read Adam Smith, which Epstein says attuned him to competition and to the emergence of order from competitive pressures. He read Malthus on population, as did Wallace. Malthus’s argument, tied to the British Poor Laws, was that charity would be overwhelmed because population would grow geometrically while food supply would not. Roberts notes that Malthus missed developments that were coming, and Epstein agrees. But a question-framer does not need to be right to be useful. Malthus crystallized a problem that helped Darwin and Wallace think differently.

Mendeleev’s own priority story fits the same pattern. There were no periodic tables before 1860 and six in the 1860s. Mendeleev’s version had important advantages: it was complete, and he made bold predictions others did not. But the main idea was in the air. One competing arrangement, the “telluric screw,” placed the elements around a barber-pole-like cylinder; viewed from above, Epstein says, the periodic pattern appeared. Its diagram was so difficult that a publisher left it out, and it received no notice.

A less glamorous boundary helped make the periodic work possible: standardization. Epstein points to an Italian figure who told chemists they were measuring atomic weights differently and distributed a pamphlet establishing a common method. That allowed work to communicate across distance. Researchers could compare one another’s measurements, and the field became more searchable.

The same logic appears in Epstein’s account of David Hilbert. Hilbert surveyed mathematics, identified two dozen important problems, defined them precisely, and distributed them to colleagues. Epstein says this helped set the agenda for 20th-century mathematics. Many of the problems were solved because Hilbert focused attention on well-defined questions. In Epstein’s formulation, a good problem definition can make other people look like geniuses.

Even Einstein is treated with this kind of humility. Roberts says he initially resisted the thought that Einstein might not be singular in the usual way. Epstein points to a footnote early in Einstein’s famous relativity paper noting that he had not read a paper by Lorentz. He does not treat this as evidence against Einstein’s originality; Einstein seems to have had unique physical interpretations of some discoveries. But he was not the only person “alighting on these equations at the same time.”

Roberts connects this to a broader pattern of memory. Posterity often keeps one name because one name is easier to narrate than a field. Adam Smith did not appear in an economic vacuum. Clausewitz was not the only 19th-century theorist of war. The remembered singleton may deserve the honor, but the surrounding field usually did more work than the legend admits.

Epstein’s personal examples show how a constraint can work by precluding the default tactic. In eighth grade, he was a strong athlete playing quarterback in gym-class touch football when his upper arm bone snapped in a spiral fracture on a throw. The injury immobilized his writing hand and ended contact sports for a year. At the time, sports were the center of his identity, so the injury “kind of ruined” his life.

The first unexpected change appeared in French class. The tests required students to listen to a French speaker and fill in blanks quickly enough to keep up. With his writing hand immobilized, Epstein could not write fast enough. He began memorizing the words as the recording played and then writing them down later with his left hand. To do that, he attached the words to sports images as mnemonics. His scores improved sharply.

Years later, Epstein read a famous memory study involving a Carnegie Mellon undergraduate who went from memorizing seven digits to 80 by using sports-related mnemonics. Memory palaces and related techniques have long histories; his point is not priority. It is that losing his normal tactic forced him to discover a better one. He still uses mnemonics to memorize hour-long keynote talks, though he rejects the idea that he has a photographic memory: if he puts down his keys and spins around, he says, he loses them.

He calls this a “preclude constraint.” When the typical method is blocked, a person must search. The substitute will not always be better, but sometimes it reveals capacities the old habit concealed. The same injury led Epstein to running because contact sports were unavailable. He eventually became a college runner and university record holder, a path he says he never would have explored otherwise.

Roberts adds the necessary caution: stories like this suffer from selection bias. We hear from people who turned lemons into lemonade, not from everyone who was simply harmed. Epstein’s argument does not require recommending injury or hardship. It requires only the narrower point that blocked options sometimes expose better routes.

The London Underground strike supplies the same pattern without the biographical drama. Certain lines were down for a few days, forcing commuters to find new paths. These were daily commuters who might have been assumed to have optimized their routes. Yet a significant portion found new routes and kept using them, saving about 1,500 commuting hours per day. Epstein’s inference is that people often do not experiment enough because they follow what cognitive scientists call the path of least resistance.

Here he quotes Daniel Willingham: “You may think your brain is made for thinking, but it’s actually made to prevent you from having to think whenever possible,” because thinking is energetically costly. Unless a familiar route is blocked, people often continue doing what they have always done.

A second personal example came after Epstein needed stitches in his head and was told to keep his blood pressure down, avoid turning his head independently of his shoulders, and sleep sitting up. After several days, he found himself oddly happy. Journaling revealed why: he had to do one thing at a time. If he tried to toggle between tasks, he felt his blood pressure rise and the scar tingle. His body gave him immediate feedback against multitasking. The experience became part of his chapter on attention and focus because it made the stress of toggling visible.

Epstein changed both the structure of Inside the Box and the structure of his workday while writing it. The book constraint responded to a weakness he saw in his first two books: he wrote about 150 percent of the needed length and then cut back. For the first book, he took a reporting trip to Arctic Sweden that did not make the manuscript. After becoming a parent, he says, he did not want to take trips to Arctic Sweden that were not going into the book.

For Inside the Box, he spent the first year reading papers, interviewing, and mapping the terrain rather than writing. Then he forced himself to make a one-page outline. The idea came partly from Tony Fadell, the lead designer of the iPod, who tells teams to write the press release before starting a project so they have a bounding box.

The recording shows Epstein holding up the outline: a spiral notebook page densely covered with hand-drawn diagrams, boxes, arrows, circles, and handwritten notes. Roberts calls it “a big mess.” Epstein agrees, but emphasizes the rule: if it is not on the page, it is not in the book.

That boundary forced priorities. Epstein says he had previously suffered from what designers call “featureitis”: trying to include anything interesting. The one-page outline also pushed him to use a structure he had not used before, returning to the periodic table story in slices throughout the book, each time adding a layer and introducing linked chapters. The constraint made him arrange related material rather than simply proceed one chapter at a time.

He connects the process to Bent Flyvbjerg’s project-management distinction between “think slow, act fast” and “think fast, act slow.” In Epstein’s summary, successful projects spend an early period defining the problem and setting boundaries, which allows execution to move faster. Rushing into execution on a broad idea makes later lessons more painful because the project already has momentum. Epstein’s first book required an extension, his second was turned in at 5 p.m. on the contract deadline, and this one was finished a month early.

His daily constraints came from research on attention, especially psychologist Gloria Mark’s work. Epstein describes Mark as having spent roughly 25 years shadowing people at work, first with a stopwatch and later with computer-activity logs, heart-rate monitors, and other tools. Her findings, as he recounts them, show a steep compression of attention: about 25 years ago, people switched tasks every three minutes; by 2012, about every minute and a half; by 2022, about every 45 seconds, where it seems to have flattened in recent years.

The number of switches predicts lower end-of-day productivity, Epstein says, and higher stress as measured by heart-rate variability. But he finds the self-interruption result even more troubling. A person accustomed to interruption may turn off notifications and still interrupt themselves with intrusive thoughts at the rhythm to which they have been trained. Epstein describes it as an internal distraction barometer.

During deep writing blocks, he did not turn on his phone until late in the day and did not check email until his work for the day was done. He acknowledges that this is not feasible for everyone. His more general recommendation is batching: answer email in one, two, or three blocks rather than throughout the day. Mark found office workers check email, on average, 77 times a day.

Email is costly not only because of the time it consumes but because of the Zeigarnik effect: open tasks occupy mental space. Epstein’s inbox always contains more than he can answer, so opening it early loads his mind with unfinished obligations before the central work is done.

The difficulty of sustaining good attention habits is part of the problem, not an excuse from it. Epstein gives two reasons those habits are hard. First, attention takes work; anyone who has tried to meditate early on knows that mindfulness is uncomfortable. Second, people have been trained into distraction, and apps are built by “an army of psychologists” trying to capture attention. His practical conclusion is sharp: if people are not engineering their own attention, it will be engineered for them.

Isabel Allende’s writing life gives Epstein a case where constraint takes the form of ritual. Allende began writing in midlife, while in exile after the coup that overthrew Salvador Allende, her father’s cousin. When she learned that her grandfather was dying, she began writing him a letter on January 8, 1981, to preserve the family stories he had told her. As she wrote, the letter turned into The House of the Spirits, which became an international bestseller.

The date became a rule. Every January 8 thereafter, if she has finished the previous book, Allende begins a new one. Epstein describes an elaborate structure around the date: she needs silence, clears out remnants of the previous project, cleanses her office, and places a Pablo Neruda book under her typewriter or computer in case creativity by osmosis exists. Before she was internationally famous, she sometimes wrote in a clothing closet with a typewriter because it was quiet.

The ritual gives her life a season. People around her know that if they need something, they should ask by January 7. Once the January 8 writing period begins, Epstein says, she turns down even large speaking fees — he mentions $150,000 — because the season is sacred. He says the pattern has produced roughly a bestseller every 18 months for 40 years and about 80 million books sold. He also notes that she has given away a large amount of money through a foundation.

The one major interruption came after Allende’s daughter, Paula, died in her 20s of a rare disease discovered only after she fell into a coma. Allende wrote the memoir Paula and said she thought she was done writing. She skipped one January 8. But when the next January 8 approached, she decided to sit down and see. The ritual pulled her back.

Epstein contrasts this working discipline with Allende’s public legend. Because some of her books use magical realism, profiles can make her sound like a mystical medium through whom characters speak. His account emphasizes boundaries: date, silence, season, office, cleansing, beginning, ending. Even the workday is bounded: Allende lights a candle to begin and blows it out to finish. Epstein adopted his own version with electric candles.

Roberts connects this to Stephen King’s routine of writing a fixed number of words before leaving the chair. Epstein says this is why ritual matters most on days when desire is absent. As a former competitive 800-meter runner, he remembers that necessary training was often unpleasant. Habit, ritual, and sometimes other people brought him back when motivation did not.

That pattern appears across creative work. Haruki Murakami describes endurance training as preparation for writing a book because writing is also an act of endurance. Music producer Rick Rubin writes about highly creative people whose days can look rigidly structured. The structure is not opposed to creativity. It reduces the number of decisions around the work so the work itself has room.

The philosophical version of Epstein’s argument comes from Bernard Suits, especially Suits’s account of games. Epstein presents Suits as answering Wittgenstein’s claim that language is fuzzy and that there is no core essence common to all games. Games differ too much: some have strict rules, some do not; some are solitary, some team-based; some are competitive, some make-believe.

Suits’s answer was that games share an attitude, which he called the “lusory attitude”: the voluntary attempt to overcome unnecessary obstacles.

For Epstein, that phrase captures a large part of the book. A race imposes an inefficient way to get from one point to another. Basketball would be easier if the basket were lower or players did not have to dribble. Golf would be easier if the hole were larger. But the unnecessary restriction is what makes the activity a game.

Epstein discusses Suits’s The Grasshopper, a parable built from Aesop’s fable. In Aesop, the grasshopper plays all summer while the ants store food and therefore faces winter unprepared. In Suits’s version, the grasshopper defends play as meaningful because it involves pushing against chosen obstacles. Epstein treats this as an analogy for life: add lines to a field, and people can enter a shared pursuit with collective meaning.

Roberts links the point to sports and to difficult but possible challenges. Moving a ball down a football field has no inherent importance. The meaning comes from the rules, the shared goal, and the refusal to cheat. His example is the scene in Raiders of the Lost Ark where a swordsman displays his weapon and Harrison Ford’s character simply shoots him. The humor depends on a violation of the expected game.

This is where Epstein’s argument becomes more than a productivity claim. Rules do not merely prevent chaos. In games, and sometimes in work and life, the obstacle is the medium through which effort becomes vivid and shared.

General Magic is Epstein’s organizational cautionary tale. Founded by three former Apple employees in the late 1980s and early 1990s, the company was effectively trying to build an iPhone in 1990, when only 15 percent of American households had computers. Two founders had helped design the original Mac. The third, Marc Porat, is described by Epstein as a technology visionary; Epstein says Porat’s 1976 Stanford dissertation coined the term “information economy” and reads eerily in retrospect.

General Magic spun out of Apple to build a “personal communicator.” The idea and talent were strong enough that, in Epstein’s telling, Goldman Sachs took it public in what he calls the first “concept IPO”: a public offering based on an idea rather than a product. Six months after the product debuted, the company had sold 3,000 units, mostly to people it knew.

The diagnosis is that the company had too much freedom and not enough definition. Porat later said he raised so much money so quickly because he wanted to create “heaven for engineers,” where they were limited only by imagination. The company also had a 17-partner alliance spanning so much of the communications technology world that meetings had to begin with an antitrust lawyer listing topics participants could not discuss. That breadth made decisions hard. The refrain from former employees, in Epstein’s interviews, was: “We could not figure out what not to do.”

The customer was equally vague. General Magic called its imagined user “Joe Sixpack,” but after missed deadlines it became clear that no one knew who Joe Sixpack was or what he wanted. So the team built for itself.

The calendar feature makes the problem concrete. Engineer Steve Perlman first built the calendar to run from 1904 to 2096. Then he was asked to support historical apps and future-oriented apps, so he extended it back to year one. Another team objected that this tied the product to an arbitrary religious context, so the calendar was pushed back to the beginning of astronomical time. Perlman told Epstein that the 1904-to-2096 version would have been four lines of code. Instead, the calendar dragged on for months.

The shipped product had many features, but Epstein says the user experience was choppy, battery life was poor, and the device was confusing. It came with a 200-page manual, including eight pages on the battery. Venture capitalist Bill Gurley gave Epstein the line that summarizes the lesson: “More startups die of indigestion than starvation.”

General Magic nevertheless produced an extraordinary alumni network, according to Epstein’s account. People associated with it later co-founded or helped create products and companies including Android, the iPod, the iPhone, Nest, LinkedIn, and eBay, and others went on to lead work on Google Maps and Safari. eBay began as AuctionWeb, incubated by a low-level service engineer at General Magic who was obligated to offer his intellectual property to the company. General Magic declined because it had a bigger thing underway.

The PalmPilot comparison is especially pointed. Its creator had originally built a General Magic third-party app called Graffiti, which converted stylus strokes into writing. When General Magic’s failure became clear, he defined a narrower customer problem: busy professionals wanted to sync contacts and calendars and take them on the go. The product would do contacts, calendar, and memo pad — period. General Magic laughed because those were only three of the many things it was trying to do. PalmPilot became a hit.

Tony Fadell, later known as the “pod father” for leading iPod design, took the opposite lesson from General Magic. Epstein describes him as a zealot for constraints. When Fadell co-founded Nest, the smart thermostat company, he made the team prototype the packaging before the product. The box forced priorities: if a feature or message belonged there, it mattered to the customer; if it did not, it could wait. Fadell told Epstein that constraint-based methods slow teams down but make them think hard.

The simple economic instinct says more is better than less, and constraints lower well-being. Income is a constraint; more income should create more satisfaction. Roberts says the older he gets, the less adequate that picture seems. People care not only about how much they have, but where it came from, whether they earned it, and whether they were respected in earning it.

The startup version makes the tension clear. More money can keep a company alive long enough to discover a product. But too much money can also destroy focus. Constraints matter because most people cannot hold unlimited possibility in mind and still choose well.

Epstein’s response relies on Herbert Simon, whom he calls the most important thinker in the book. Simon was trained as a political scientist, won the Turing Award in computer science for co-producing the first AI demonstration, won what Epstein describes as the highest award in psychology, and won the Nobel Prize in economics for bounded rationality and related work. In Epstein’s account, Simon’s life’s work was motivated by the observation that people do not behave according to classical models of optimal decision-making.

Simon coined “satisficing,” a blend of satisfy and suffice. Because people have finite brains and finite capacity to evaluate options, they cannot truly optimize. They use shortcuts and heuristics. Simon’s argument, as Epstein presents it, was that people should set good-enough rules deliberately.

Simon practiced this himself. Epstein says Simon believed a person needed only three pairs of clothes: one being worn, one in the wash, one in the closet. He ate the same breakfast every day and wore the same beret. He wrote that “the best is the enemy of the good.” The point was not laziness. It was that the attempt to maximize carries costs in time, money, and energy. If those costs are counted, satisficing may be the maximizing strategy.

Epstein applies this to ordinary decisions: decide in advance the few things a purchase, project, or choice must accomplish. Once those conditions are met, decide and move on. Otherwise, people fall into Fredkin’s paradox: they spend the most time on the least important decisions because the options are hard to distinguish. The difficulty of distinguishing them may itself be evidence that the difference is small or that further analysis will not help.