Conway's Constant and the Power of Slow Attention

Watch talented people get passed over long enough and you notice the reason is almost never what they think it is. It wasn't performance. It wasn't potential. It was usually this: they'd been trained — by culture, by management, by the whole optimize-and-ship machinery of modern work — to move fast, produce visibly, and never appear to be just sitting with something. Sitting with something looks like idling. Idling doesn't get you promoted.

The pattern, watched closely enough, is unmistakable since watching a Quantia video this week about a children's counting game that hides, inside its absurdly simple rule, an irreducible polynomial of degree 71. The math is genuinely beautiful. But what can't stop being turned over is the story of how it was found — because it is a direct rebuke to everything modern work culture told workers to be.

Here's the game. Write the digit 1. Now read it aloud: "one one." Write that down: 11. Read that aloud: "two ones." Write that: 21. Keep going. You're never doing arithmetic. You're just describing what you see in the previous line, and the description becomes the new line. A four-year-old can do this. The resulting sequence — called the look-and-say sequence — starts:

1, 11, 21, 1211, 111221, 312211, 13112221...

It looks like noise. Strings of ones, twos, and threes that seem to have no pattern. But if you stop caring about what the lines say and start measuring how long they are, something clicks into focus. The lengths grow: 1, 2, 2, 4, 6, 6, 8, 10, 14, 20, 26, 34, 46, 62, 79, 106. Take the ratio of consecutive lengths. It bounces around at first — 2.0, 1.0, 2.0, 1.5 — but it starts settling. By the time you're deep into the sequence, consecutive length ratios are converging on a number just above 1.3.

Plot the line lengths against the line number on a log scale, and you get something close to a straight line. That means exponential growth, and the growth rate — the slope, once you take the antilog — is approximately 1.303577269034496. This number has a name: Conway's constant. And Quantia's video spends considerable time on what that name really means.

Conway's constant is not an approximation. It is the unique positive real root of a specific irreducible polynomial of degree 71. Irreducible meaning it cannot be broken down into smaller polynomials with integer coefficients. The polynomial's coefficients — integers, most of them small, some reaching into the thousands — are not arbitrary. They encode the structure of the sequence itself. They fell out of a counting game.

"The polynomial's coefficients are not arbitrary," the Quantia video explains. "They encode the structure of the sequence itself."

How does a children's game contain something like this? The answer is where John Conway enters.

Conway did not invent the look-and-say sequence. By the time he turned his attention to it in the mid-1980s, it had been kicking around mathematical recreation circles for years — a curiosity, a parlor trick, something you'd show a kid to make them feel clever. Nobody had looked at it seriously because it didn't look serious. It looked like a game.

What Conway did was refuse to accept that framing. He sat with it.

What he eventually noticed — and this is the part that makes the hair on your arms stand up a little — is that long lines in the sequence aren't random at all. They're built from recurring chunks. Specific strings of digits that show up over and over, in the same forms, behaving in predictable ways. He called these chunks atoms, deliberately borrowing the chemistry metaphor. There are 92 common atoms, mirroring the 92 naturally occurring chemical elements, plus additional infinite families. He named them after those elements: hydrogen, helium, lithium, all the way to uranium.

Each atom, when you apply the look-and-say rule to it, breaks apart into a specific list of other atoms — exactly like nuclear decay. The whole system is closed. Hydrogen, which is just the string "22," reads as "two twos" and produces "22" again — it's stable, the only stable atom in the system. Everything else decays. Uranium decays into a chain of other atoms. Conway worked out the full decay table by hand.

Once you have that table, you can write it as a matrix. A 92×92 grid where each entry tells you how many copies of atom i appear when atom j decays one step. The long-term growth rate of any sequence under a matrix like this is governed by the matrix's largest eigenvalue. That eigenvalue is a root of the matrix's characteristic polynomial. When you factor Conway's characteristic polynomial, most factors correspond to stable atoms with eigenvalues less than one — they fade out and don't contribute to growth. But one irreducible factor remains. Its degree is 71. Its largest positive root is 1.303577269034496.

"The interesting part is concentrated in a single irreducible block of size 71. Inside that block, all the long-term growth lives," Quantia explains.

Conway called his result the cosmological theorem: every starting string, no matter what it is, decomposes into atoms within at most 24 steps. After that, atoms only decay into other atoms, forever. The whole universe of look-and-say behavior is governed by 94 building blocks and one constant.

A note on the sequence's scale, since the video's claim here deserves scrutiny: the video states that line 100 has "around 100 million digits." But working from Conway's constant — 1.3036^100 is in the neighborhood of 5 × 10^11 — the 100th line would more plausibly run to hundreds of billions of digits, not 100 million. The 100-million figure may apply to a much earlier line, around line 70 or so. A small thing, but precision matters when the whole story is about exact algebra emerging from apparent chaos.

Similarly, Conway's paper "The Weird and Wonderful Chemistry of Audioactive Decay" is most commonly cited as 1986, but some sources list 1987 — it was published in a festschrift, which can complicate dating. The mathematical content is not in dispute; only the citation year warrants a caveat.

Now here's where the turn comes, because there's something real here and not just a hook.

Read enough job descriptions and the thing they say they want is "self-starter," "results-driven," "thrives in fast-paced environments." What actually gets rewarded — and you can watch this cycle repeat until it becomes depressing — is visible busyness. Action. Output you could point to in a meeting. The career advice industrial complex has spent decades reinforcing this: optimize your time, ship faster, fail fast, show results.

John Conway looked at a thing the entire mathematical community had categorized as a parlor trick and spent serious time on it. Not because someone told him to. Not because it was on his OKRs. Because he noticed something that looked like structure and he didn't dismiss the noticing.

What came out of that patience was not merely a clever observation. It was the cosmological theorem. It was an exact closed-form description of exponential growth hiding inside a game a child could play. It was a degree-71 polynomial with integer coefficients that no one had any reason to expect existed.

"None of this was intended," the Quantia video says. "The sequence was a curiosity. The mathematics emerged from staring at it long enough."

The optimize-and-ship culture eats people alive — specifically the careful, deep-thinking ones who do their best work when they're allowed to sit with a problem. They get labeled as slow, or not strategic, or lacking urgency. Meanwhile the people who move fast and produce visible output get promoted, even when the output is shallow, even when the underlying problem hasn't actually been solved.

The look-and-say sequence doesn't care about any of that. It only gives up its structure to someone willing to stare at it long enough to see the atoms. Conway's constant isn't the reward for hustle. It's the reward for the kind of attention that our current work culture has systematically devalued.

I'm not saying your career will produce a cosmological theorem if you just slow down. That's not the argument. The argument is narrower and, I think, more honest: the things that are actually hard to see — in math, in organizations, in the work right in front of you — don't reveal themselves on a deadline. They reveal themselves to sustained, non-transactional attention. That capacity is rarer than it used to be, because the workplace has spent two decades training it out of people.

Conway sat with a parlor trick until it told him something true. That's not a method everyone is in a position to use. But it's worth knowing it's a method at all.

Vanessa Torres is a career and workplace writer for Buzzrag. She covers the gap between how workplaces present themselves and how they actually function.