JSON
The lingua franca of the web: data as plain text.
Where it came from
JSON was formalized by Douglas Crockford around 2001, building on JavaScript object literal syntax from the mid-90s. The joke in the name is real: 'JavaScript Object Notation' was never meant to be a standard — Crockford famously said he 'discovered' it rather than invented it.
The spec (RFC 8259) fits on a few pages, which is exactly why it won. By 2005–2010 JSON displaced XML in web APIs almost everywhere: it was lighter to parse, easier to read, and every language could map its types natively.
Today it's the default interchange format for APIs, config files (package.json, tsconfig.json), logs, and databases (PostgreSQL, MongoDB).
How it works
A JSON document has exactly six value types: object («…»), array («[…]»), string, number, boolean, and null. Strings are double-quoted; numbers are decimal (no hex, no NaN); trailing commas are forbidden.
Everything else you see in 'JSON-like' files — comments, single quotes, unquoted keys — is a dialect (JSON5, HJSON), not JSON itself. That strictness is a feature: a parser can be tiny and fast.
Whitespace between tokens is insignificant, so pretty-printing and minifying are lossless operations — that's what our JSON pretty / minify / sort-keys converters do.
Convert JSON right here
UWC parses and serializes JSON entirely in your browser — your data never leaves the machine. Round-trips through YAML, TOML, CSV and XML preserve values and typing.