author RANGE, TheSportsGene, and now INSIDE THE BOX: How Constraints Make Us Better. (Pic is w/Frances Hesselbein).
- Tweets 29,846
- Following 1,400
- Followers 58,890
- Likes 18,367
Pinned Tweet
Holy cow, that is crazy praise coming from a thinker I admire 🙏
1/ I just finished one of the best books I've EVER read.
Comes out next week.
My friend @DavidEpstein (Range, The Sports Gene) has written a masterpiece on the virtue of constraints, of not thinking OUTSIDE the box but "INSIDE the BOX"...
As a few quick excerpts that 🤯...
10
9
109
32,921
Fascinating applied case here. I feel lucky to have a reader like @ChrisBorland_ (link to his full article in next post)
2
2
15
2,461
He barred black so thoroughly that when they placed a black shroud over his coffin at the funeral, a friend screamed: "No! No black for Monet!" and went and found a floral table cloth.
Narrow restrictions force the brain into productive experimentation.
By banning black from his palette, Claude Monet was forced to paint light completely differently, launching the Impressionist movement.
@DavidEpstein
8
3,289
David Epstein retweeted
Narrow restrictions force the brain into productive experimentation.
By banning black from his palette, Claude Monet was forced to paint light completely differently, launching the Impressionist movement.
@DavidEpstein
1
4
4,216
I was fortunate to correspond with Brooks before he passed away...brilliant man. This post reminds me a bit of Mokyr's economic history work on practitioners, tinkerers, and innovation.
The linear model of innovation says progress starts at universities, then moves to industry, then reaches customers.
In a recent talk, @cmuratori presented the 1960s software crisis. Dijkstra, a researcher and professor who actually changed practice, was depressed. In 1972, he put it this way:
« As long as there were no machines, programming was no problem at all; when we had a few weak computers, programming became a mild problem; and now we have gigantic computers, programming has become an equally gigantic problem. »
He wanted more planning and more formal verification.
Universities pushed that line for decades, at great expense.
It was the wrong bet, for reasons practitioners such as Fred Brooks could see. A complete formal spec of what the software ought to do sounds good in theory. In practice you often discover the requirement by trying ideas and iterating.
We are much better today at building reliable software (although this knowledge is not as widespread as I would hope). The progress took many years, and was driven by several actors, only few of them being high IQ professors.
We are in a new mess. AI-assisted programming produces what people call slop. Some are depressed. Others push forward. I do not know where we should go.
This innovation (AI programming) was NOT designed on campus. Programmers glued large language models to their tools and got agentic AI.
Now we have to think it through. A professor may invent a framing that changes how we see agents. Or not.
That is why the linear model is false. Innovation does not systematically start on campus. Customers and working programmers often matter more than intellectuals.
You may say software is a special case, not nuclear physics. Is it?
If analysis was the foundation of science in the early 20th century, software is the foundation of science in the early 21st. Software ate the world, including science.
@ericweinstein recently argued on the All-In podcast that we should give selected scientists so much money they cannot be managed, and wait for another Einstein. Fine. But in 1905 Einstein was a patent clerk, not a professor with infinite funding. Princeton came later. It is not obvious that the later work was better.
My home country, Canada, has programs that look like what Eric is advocating. There are hundreds of Canadian professors who have the equivalent of what Eric advocates. Some have their own buildings, dedicated to their work.
Whether that turned them into breakthrough producers is another story.
You cannot engineer great software with a neat up-front design. Trial and error is all you have. The same is true of breakthrough science. Nobody is smart enough to design a government program that reliably produces new Einsteins. Your program may soon be captured.
Around 1905, a patent clerk reinvented physics while two bicycle mechanics built an airplane. Stop looking only up. The next big thing often comes from misfits your program would have excluded.
We have to accept the fact that progress is somewhat unpredictable and that it cannot be engineered up front. Cool down the hubris and let us get serious.
Links
•Dijkstra, The Humble Programmer (1972): cs.utexas.edu/~EWD/ewd03xx/E…
•Brooks, The Mythical Man-Month (1975)
•Casey: piped.video/watch?v=hpj6r6Cj…
•Eric: piped.video/watch?v=y84ypzEL…
6
3,330
This video by @DavidEpstein really hit home. Perfect for the current era of cheap software.
Among the products I have built with different levels of resources/funding: almost every time adding constraints has lead to a better, refined and successful product.
The ones with infinite resources, too many features almost always failed/ never shipped.
piped.video/watch?v=ZtbriAhl…
1
2
2
3,272
David Epstein retweeted
3/3 That’s why David Epstein’s Inside the Box feels especially relevant: constraints don’t just inspire creativity — they test assumptions, force prioritization, and develop judgment.
Too much capital too early can sometimes do the opposite.
davidepstein.com/inside-the-… @DavidEpstein
2
2
1,102
David Epstein retweeted
‘Travel sports,’ a term that has come to encompass all sorts of private teams, leagues, and clubs, used to be for the truly exceptional young athlete. Now, they’re for anyone willing to pay — an expensive, exhausting commitment that has become popular not necessarily because millions of parents want them for their children but because it can feel like there’s no other choice. Rec programs are underfunded and struggle to compete with for-profit leagues for prime field, rink, and court times. And the level of play is diluted, often very early, as travel teams are incentivized to enroll children at young ages.
Around 5 million children in the U.S. now compete on travel teams. Parents find themselves trapped. They know the benefits of sports and want that for their children, yet giving it to them often means buying into a sports ecosystem that’s inherently flawed. As the pay-to-play model increasingly takes hold, so does the idea that kids should start competing at younger and younger ages, risking their sense of self along with their shoulders and ACLs.
“I never want my kids to look back and say I didn’t do everything I could to help them live their dreams,” a mother of three soccer players said. She’s hired professional photographers to take action shots for her 12- and 13-year-olds’ Instagram accounts — the platform has become a recruiting tool — and they’ve both asked her to pay for personal trainers.
“I think the word isn’t ‘specialize’; you are ‘professionalizing’ your kid early. You are exposing them to everything professionals have to deal with — I’m talking injury, time, schedule, everything,” said John Van Benschoten, a former MLB player and travel-baseball coach.
It makes one ask: All this for what? And for whom? Read Caitlin Moscatello’s cover story on how travel leagues have made youth sports more expensive, hypercompetitive, and all-consuming — and the investors who are making a killing: nymag.visitlink.me/1-gbZQ
297
1,249
5,549
4,162,268
David Epstein retweeted
Excited to read this next as I thought his book Range was full of exceptional stories and its central thesis about breadth felt right.
1/ I just finished one of the best books I've EVER read.
Comes out next week.
My friend @DavidEpstein (Range, The Sports Gene) has written a masterpiece on the virtue of constraints, of not thinking OUTSIDE the box but "INSIDE the BOX"...
As a few quick excerpts that 🤯...
10
13
169
56,172
"Adults struggling to make art often have the problem that they have too much adult in them and not enough child." @DavidEpstein interviews bestselling author and artist @austinkleon on his new book, Don’t Call It Art: 10 Ways to Create Like a Kid Again davidepstein.substack.com/p/…
4
13
6,075
David Epstein retweeted
What if innovation thrives not despite constraints, but because of them? We’re pleased to welcome @DavidEpstein, No.1 New York Times bestselling author of Range, to our 33rd Investors’ Forum in Grand Hyatt Hong Kong shorturl.at/jYKMX #CITICSecuritiesInternational
1
1
1,487
David Epstein retweeted
Constraints breed creativity.
6 ideas from "Inside the Box" by @DavidEpstein
1/ I just finished one of the best books I've EVER read.
Comes out next week.
My friend @DavidEpstein (Range, The Sports Gene) has written a masterpiece on the virtue of constraints, of not thinking OUTSIDE the box but "INSIDE the BOX"...
As a few quick excerpts that 🤯...
1
3
2,785
