0.999... (repeating nines) approaches 1 but never actually equals 1.
0.999... is exactly equal to 1, not approximately equal. They are two representations of the same real number. Since 1/3 = 0.333..., multiplying both sides by 3 gives 0.999... = 1. More formally, the real number system defines a repeating decimal as the limit of its partial sums, and the limit of 0.9, 0.99, 0.999, ... is exactly 1.
What changed?
Most people who encounter 0.999... for the first time form the same intuition: the number is falling perpetually short of 1. Each additional 9 brings it closer, but the distance never quite closes. The intuition is wrong, and the error is understandable, because it conflates two distinct things: the process of approximation and the number the process defines.
In standard mathematics, 0.999... is not a sequence of partial sums that approaches 1. It is the limit of that sequence, which is a completed value, not an ongoing one. The number 0.999... and the number 1 are two different representations of the same real number, in the same way that 1/2 and 0.5 and 2/4 are different representations of the same quantity.
The simplest proof fits on three lines. Since 1/3 = 0.333..., multiplying both sides by 3 gives 3/3 = 0.999.... And 3/3 equals 1. A second approach: let x = 0.999.... Multiply both sides by 10: 10x = 9.999.... Subtract the first equation from the second: 9x = 9, so x = 1. Both proofs are arithmetically valid and arrive at the same conclusion.
The persistence of the misconception in classrooms had a specific mechanism. Teachers introducing limits in pre-calculus courses described decimal expansions as 'approaching' their limiting value, which is technically accurate but obscures the fact that the limit itself is a fixed number. Students heard 'approaching' and visualized motion toward a target that is never quite reached. The language of approximation was applied to a defined value, and the two ideas fused.
What reinforced the confusion was the absence of a formal definition of what a decimal expansion actually means. In the real number system developed by Richard Dedekind and Karl Weierstrass in the 1870s, a decimal expansion is defined as the supremum of its partial sums. The supremum of the set containing 0.9, 0.99, 0.999, and so on is exactly 1. Under this definition, 0.999... = 1 is not a curiosity or a paradox; it is a direct consequence of how 'decimal expansion' is defined. But the 19th-century formalism rarely reached K-12 classrooms, where decimals were presented through long division and intuition rather than through set-theoretic definitions.
Joseph Mazur, in his 2005 book on mathematical proof and intuition, traced classroom mathematics' discomfort with completed infinities to a broader philosophical tradition that treated infinity as an unreachable horizon rather than a mathematical object. Georg Cantor's 19th-century work establishing the arithmetic of infinite sets was not incorporated into standard mathematical education until the mid-20th century, and even then incompletely.
For students who found the algebraic proofs unsatisfying, the issue was usually not with the algebra but with the underlying number system. In the hyperreal numbers, an extension of the reals developed by Abraham Robinson in 1961 for nonstandard analysis, there exist positive numbers smaller than any positive real, called infinitesimals. In that system, the gap between 0.999... and 1 can be assigned a meaning. But the hyperreals are not the number system students are working in when they learn decimal arithmetic. In the standard reals, the two expressions refer to the same number, without qualification.
