Base65536

Base65536 is a character encoding designed to represent binary data as text, using 65,536 Unicode code points (16 bits per character).

Similar to Base64, which uses 64 ASCII characters, Base65536 utilizes Unicode characters (except spaces, control characters, and other unassigned characters), allowing twice as much data to be stored per character (counting Unicode characters, not bytes after UTF-8 encoding).

Base65536 encoding relies on reading pairs of bytes (16 bits) that are mapped to a Unicode code point from U+3400 to U+285FF.

Base65536 decoding reverses the process:

— Read each Unicode codepoint in the string (ignore ASCII and other non-Base65536 characters)

— Convert each codepoint to its 16-bit value

— Reconstruct the original 2 bytes

Base65536 encoded text is distinguished by:

— Exotic Unicode characters, primarily Chinese, with code points ranging from U+03400 to U+04CFF, U+04E00 to U+09EFF, U+0A100 to U+0A3FF, U+0A500 to U+0A5FF, U+10600 to U+106FF, U+12000 to U+122FF, U+13000 to U+133FF, U+14400 to U+145FF, U+16800 to U+169FF, U+20000 to U+285FF, and finally U+1500 to U+15FF (used mainly at the end of a string as padding)

— The absence of ASCII characters (unless added manually)

— A character count reduced by half for the same content

Base65536 inspired other encodings such as Base2048 or Base32768.

The first implementation would be qntm in 2015 here