How does a book cipher work?

Both parties agree on a shared text — a book, article or page. Each word of the secret message is replaced by a reference pointing to a copy of that same word inside the shared text, such as its line number and position on the line. Anyone without the exact text cannot resolve the references.

What does a reference like 3.5 mean?

In this tool a reference is written as line.word — so 3.5 means the 5th word on line 3 of the reference text. Decoding looks up that exact position and returns the word found there.

What if a word is not in the reference text?

A word can only be encoded if it appears somewhere in the reference text. If a word is missing, the tool marks it so you know to choose a richer passage or reword the message. A larger reference text covers more vocabulary.

Why is the book cipher considered strong?

Its strength comes from the secrecy and size of the key text rather than a fixed algorithm. Without the precise reference text an interceptor sees only numbers with no pattern. Historically it resisted analysis well, though reusing a famous public book weakens it.

What is the Ottendorf cipher?

The Ottendorf cipher is a three-number variant of the book cipher using page, line and word (or letter). This tool uses a two-number line.word scheme, which is simpler and works directly on a pasted passage rather than a whole numbered book.

Book Cipher — Gera Tools

A book cipher is a key-text cipher: instead of a fixed algorithm, its security rests on a shared passage of text known to both the sender and the receiver. Each word of the secret message is replaced with a reference pointing to where that same word appears in the shared text. To anyone without the exact passage, the message is just a meaningless list of numbers.

How it works

This tool uses a simple line.word reference scheme. The reference text is split into lines, and each line into words. To encode, the tool takes each word of your message, searches the reference text for a matching word (case-insensitively), and emits a reference of the form line.word — for example 3.5 meaning the 5th word on the 3rd line.

To decode, each line.word reference is resolved by jumping to that line and counting along to that word, returning the original text found there. Because the lookup is positional, the two sides only need to agree on the exact same reference text — no other key is required.

Worked example

Suppose the reference text is:

the quick brown fox
jumps over the lazy dog

To encode the message the dog, the tool finds the as the 1st word on line 1 (1.1) and dog as the 5th word on line 2 (2.5), giving the ciphertext 1.1 2.5. Decoding 1.1 2.5 with the same passage walks straight back to the dog.

Notes and tips

Choose a reference text that contains every word you need — longer, varied passages cover more vocabulary. Keep the exact passage secret and identical on both ends, including line breaks, since those define the references. Avoid famous public texts if you want real secrecy. Words not present in the reference cannot be encoded and are flagged. Everything runs locally in your browser; nothing is uploaded.