Character sets: ASCII, extended ASCII and Unicode

A computer can't store the letter "A" as a shape. Instead, every character is given a unique number, its character code, and text is stored as a sequence of those codes in binary.

Type a character, and its code is what gets stored. Display text, and the computer looks up which symbol each code stands for. Like two spies sharing a codebook, both sides must use the same table: the character set.

• ASCII's first 32 codes (and code 127) are control characters that don't print, such as line feed and carriage return.
• 7-bit ASCII is usually stored in a byte with the MSB set to 0.
• Extended ASCII and Unicode both keep the original 128 ASCII codes unchanged.

• ASCII is too small. 128 codes cover English, but not é, ñ, Ω, 中 or 😀.
• Extended ASCII is inconsistent. Its extra 128 codes vary between versions (called code pages). The same code can be é on one computer and a completely different symbol on another, so text arrives garbled.
• Unicode fixes both. It gives every character in every language one unique code point, written in hex as U+ and the number: U+0041 is A, U+00E9 is é.

The cost: Unicode characters can take more storage than ASCII ones (up to 4 bytes instead of 1).