Data

Bitcoin's raw data is a long string of bytes. A transaction, a block header and a network message are each a fixed sequence of fields, and every field has a size measured in bytes. Open a raw transaction and you see hex digits, numbers stored backwards and short length prefixes. This section explains those conventions.

Why does raw data need its own conventions?

Bitcoin software stores and sends data as bytes, not as text. A byte is eight bits, and it has no fixed meaning until a field says what it holds. The same eight bits can be a number, a letter or part of a hash. So each page here explains one convention before you meet it in a transaction or a block.

What does this section cover?

• Bytes: eight bits, the unit that raw data is measured in.
• Hexadecimal: the base-16 digits that write each byte as two characters.
• Byte order: why transaction IDs and block hashes appear reversed.
• Little-endian: the backwards order used for numbers in raw data.
• Compact size: the variable-length counts and lengths that take one to nine bytes.

Which order should you read it in?

Read the pages in this order. Each one uses the words of the page before it.

1. Bytes: what a byte is and how it is read.
2. Hexadecimal: the digits that show bytes in a compact form.
3. Byte order: the natural and reverse orders of hashes.
4. Little-endian: how numbers are stored in raw data.
5. Compact size: the prefix that lets a count take one byte or more.

Once these make sense, the private key and the fields of a transaction will be easier to read.