How to tell them apart
If you have a message and don't know how it was made, a few quick checks narrow it down a long way.
| What you see | What it probably is |
|---|---|
| Only dots, dashes and slashes | Morse code, or tap code if there are no dashes |
| Only 0 and 1 | Binary; in groups of five, perhaps Baconian |
| Numbers from 1 to 26 | A1Z26 |
| Pairs of digits, each 1 to 5 | Polybius square |
| Mixed capitals, small letters and digits, maybe ending in = | Base64 |
| Angular symbols, some with dots | Pigpen |
| Letters, with E, T and A still the most common | A transposition: rail fence, columnar, scytale |
| Letters, with a few unusual ones very common | A simple substitution: Caesar, Atbash, affine, or a scrambled alphabet |
| Letters, all roughly equally common | Polyalphabetic: Vigenère, Beaufort, autokey |
| Letters in pairs, never a double, no J | Playfair |
Or skip the detective work: the cipher identifier runs these checks and tries the likely ciphers for you.
Cipher or code?
The words are used loosely, but there is a useful distinction. A cipher works on individual letters according to a rule and a secret key; change the key and the same message comes out differently. A code, strictly, replaces whole words or phrases using a codebook. In everyday use, "code" also covers public encodings like Morse and binary, which have no secret at all. Those are grouped here under codes and encodings.
Are any of these safe to use?
For a real secret, no. Every cipher on this site can be broken, most of them in under a second, and the pages explain how. They are here because they are the foundation of the subject, because they make excellent puzzles, and because working out why each one fails is the best way to understand what modern encryption has to get right. For actual privacy, use well-reviewed modern software.