The idea of a limit
If, as x approaches a (with x ≠ a), the values f(x) approach a number A, then A is called the limit of f at a, written lim f(x) = A as x → a. The approach must give the same number from the left (x < a) and from the right (x > a); if the two sides give different numbers, the limit does not exist. The limit does not depend on the value f(a): the function may be undefined at a or have a different value there; only the values at points near a matter. For continuous expressions the limit is found by plain substitution. If substitution gives 0/0, use x ≠ a: factorise, cancel, simplify, and then take the limit.
A calculator cannot evaluate 0/0, so sin x / x cannot be found at x = 0. Yet at x = 0.1 rad the value is 0.9983 and at x = 0.01 rad it is 0.99998: the values get closer and closer to 1. A limit is the precise way to talk about this “getting closer”, and it does not depend on the value at the point itself.
Programmers use limits to avoid “division by zero” in numerical calculations, physicists simplify formulas for small angles, and engineers judge what the result looks like as a measurement error tends to zero.
Fold a long paper strip in half again and again, and each time write the remaining length as a fraction: 1/2, 1/4, 1/8, … The length approaches zero without ever being zero; in practice the folding stops after 6 or 7 folds.
“Is there a limit?”: the teacher puts 4–5 different functions on the board (piecewise, fractional, |x|/x). In pairs, students tabulate x-values approaching a from the left and from the right, decide whether a limit exists, and check each other’s tables.