What does it feel like to invent math?

A geometric-looking “nonsense” identity—1 + 2 + 4 + 8 + … = −1—can be made meaningful once mathematicians redefine what “distance” and “infinite sum” mean. The core move is to treat certain infinite series not as literal completed computations, but as limits under a new notion of convergence. In the usual real-number sense, the partial sums of 1 + 2 + 4 + 8 + … blow up, so the expression can’t equal −1. But with a different metric, the same sequence of partial sums can be said to approach −1, turning the equation into a legitimate statement rather than a contradiction.

The path starts with a more familiar paradox: 1/2 + 1/4 + 1/8 + … = 1. Early on, the transcript frames the discovery as an “early mathematician” trying to justify what it means to add infinitely many terms. The key definition is convergence: an infinite sum equals x if the finite truncations form a list that gets arbitrarily close to x—meaning that for any tiny tolerance, the tail of the list eventually stays within that tolerance. This reframes “approach” as a precise limiting behavior, not as a vague feeling.

From there, the discussion generalizes the idea by changing how the interval is split. Instead of halving each time, imagine cutting a segment into proportions p and 1−p, then repeatedly splitting the right piece in the same ratio. The resulting geometric series has the form (1−p) + p(1−p) + p²(1−p) + …, which sums to 1 when 0 < p < 1. The transcript then highlights a crucial tension: the algebraic formula still “works” if p is replaced by values outside that range, producing outcomes like 0.9 repeating = 1, but also producing absurdities such as alternating sums for p = −1 and explosive growth for p = 2.

To rescue the p = 2 case, the transcript argues that the real obstacle isn’t the algebra—it’s the underlying notion of distance. In the usual real metric, powers of 2 don’t get close to 0; they race away. So the discussion introduces a shift-invariant distance function and constructs one by organizing rational numbers into nested “rooms” based on powers of 2 they share. Two numbers are declared close if they fall into the same smallest room in this hierarchy; the distance halves each time the shared room becomes more specific. This produces the 2-adic metric, a member of the broader p-adic family.

Under the 2-adic metric, powers of 2 do approach 0, and therefore the partial sums 1, 3, 7, 15, 31, … genuinely approach −1. The transcript closes by tying the story to a broader philosophy: mathematicians often begin with ill-defined or “fuzzy” discoveries, then invent rigorous concepts—like convergence and p-adic distance—that make those discoveries precise and useful, expanding the toolkit for future questions.