NADA visual atlas
Interactive, recomputable views of the vocabulary of all code.
the atlas of code's vocabulary
98,522,628 names → 2,552,494 words → 190 locales
NADA reads the world's programming languages and libraries, distills them to their essential shared vocabulary, and localizes that vocabulary into every human language. These pages show what that looks like.
Scale
The whole project's scale at a glance — what the brain is, and how big.
Morphemes
The 100 most-used word-parts of code.
Identifiers
The 100 most-used whole identifiers across 22,336 library packages.
Convergence
Why adding languages stops adding words — how alike code is.
Universal words
The words nearly every language shares — 'name' is in 532 of them.
Language kinship
A kinship grid: how much any two languages' vocabularies overlap.
Coverage
How much of each programming language is localized into each human language — and the frontier still to fill.
Every number here is computed from the live NADA pipeline — nothing is hand-typed.