[Beginner] — The base you already use, examined properly: why ten, why it is not mathematically special, and which fractions it can and cannot write down.
Before you start
- You can convert between bases — number bases.
What you will be able to do after this lesson:
- Explain why humans use base ten, and why that reason is not a mathematical one.
- Explain which fractions terminate in a given base, using prime factors.
- Explain why exactly.
1. Why ten
Because we have ten fingers. That is the whole reason, and it is worth saying plainly, because it makes the arbitrariness obvious.
Other bases have been used for exactly the same kind of reason:
- Base 60 (Babylonian) — survives in 60 seconds, 60 minutes, 360 degrees. It has many divisors (2, 3, 4, 5, 6, 10, 12, 15, 20, 30), which makes fractions convenient.
- Base 20 (Mayan, and traces in French quatre-vingts) — fingers and toes.
- Base 12 — twelve has more divisors than ten, which is why dozens, inches and clock faces persist.
Base ten is not better. It is habitual. Twelve would arguably be more useful for hand arithmetic, since it divides evenly by 3 and 4.
2. Which fractions terminate
This is the useful mathematical content of the lesson.
A fraction terminates in base exactly when its denominator, in lowest terms, has no prime factors other than those of .
Base ten factors as , so:
| Fraction | Denominator factors | Terminates in base 10? |
|---|---|---|
| 2 | yes — 0.5 | |
| 2×2 | yes — 0.25 | |
| 5 | yes — 0.2 | |
| 2³ | yes — 0.125 | |
| 3 | no — 0.333… | |
| 7 | no — 0.142857… |
And the same rule explains floating point. Base two factors as just , so only denominators that are powers of two terminate. has a factor of 5, so 0.1 cannot be written exactly in binary — which is why 0.1 + 0.2 != 0.3 in nearly every language.
That is not imprecision. Binary floating point is exact; 0.1 simply is not a binary-representable number, in the same way is not a finite decimal.
3. Why 0.999… = 1
Not “nearly 1”. Exactly 1. Three arguments, and the third is the real one:
The rigorous version: is defined as the limit of the sequence , and that limit is exactly 1. The two numerals denote the same real number.
This is a fact about notation, not about numbers. Decimal expansion is not a unique naming scheme — some numbers have two names. Nothing is wrong.
4. Practice — independent task
- (a) Without dividing, predict which of , , , terminate in decimal. Then check.
- (b) Which of them terminate in binary? State the rule you used.
- (c) In base 12, which of , , , terminate? What does that suggest about base 12 for everyday arithmetic?
- (d) Write a function
terminates(numerator, denominator, base)that answers this without doing any division — reduce the fraction, then check the denominator’s prime factors against the base’s. - (e) Use it to find every denominator under 30 whose reciprocal terminates in base 10 but not base 2.
Done when: your function agrees with actual long division for all denominators up to 50 in bases 2, 10 and 12.
Hint for (d) — open only after an attempt
Reduce the fraction first usinggcd. Then repeatedly divide the denominator by each prime factor of the base, as many times as it goes. If you reach 1, it terminates; if anything is left over, it does not.You do not need to factorise the base properly — dividing out each of its prime factors greedily is enough, and for small bases you can hardcode them.
Before moving on
- Explain why base ten was chosen, and why that is not a mathematical reason.
- State the terminating-fraction rule and apply it in bases 2, 10 and 12.
- Explain why 0.1 is not exactly representable in binary.
- Explain why .
Recap: Base ten is a biological accident, not a mathematical one — other cultures used 12, 20 and 60 for equally practical reasons. A fraction terminates in base exactly when its reduced denominator has no prime factors outside those of , which explains both why never terminates in decimal and why 0.1 is not exactly representable in binary. And equals 1 exactly, because decimal notation does not name every real number uniquely.
Next: Hexadecimal — the base chosen deliberately, for a reason that is entirely mathematical.
Related
- Number bases · Binary
- computer-architecture/data representation — floating point in detail
Software engineering
Infrastructure & security
Computer science
Hardware & engineering