About Plangs

Plangs is a database of programming languages and the things that grow around them — compilers, editors, libraries, frameworks, platforms, licenses. It exists because the useful facts about a language are usually scattered across a Wikipedia infobox, a README, a release page and somebody’s blog post, and nobody keeps them in one place you can query.

What gets included

The bar is usable for real work, or close to it. A language does not need a foundation behind it or a conference, but it should have documentation someone could learn from and an implementation someone could run. Experimental and young languages are welcome when they clear that line — plenty of them do.

What Plangs deliberately does not do is rank languages or tell you which one to pick. The trend charts show attention over time because that is a fact about the world; they are not a leaderboard, and a language being small is not a defect.

How the data works

Every node — each language, tool, library, license, paradigm, tag and platform — is one YAML file. That folder is the canonical source of truth. There is no database server, no admin panel, and no hosted table that could disagree with the files. The site is generated from them.

The practical upshot:

Everything is diffable. A correction is a one-line change to a text file, with a history attached.
Every page has a machine-readable twin. Add .md to any URL to get the markdown behind it.
Agents get a description. See /llms.txt.
Relationships are first-class. Influence, implementation, tooling and licensing are edges in a graph, readable from both ends — which is what the influence diagrams on language pages are computed from.

The whole site is static. Pages are built ahead of time and served as files, which is why they load quickly and work with JavaScript switched off. The approach owes a lot to Simon Willison’s Baked Data pattern.

Corrections and contributions

The data started as Wikipedia extracts, some AI prompting, and Languish trend data, all of it then corrected by hand. That origin story guarantees errors: dates that are off, influences that are backwards, a paradigm nobody who uses the language would recognise.

If you know a language well, you will spot its wrong facts immediately. Those corrections are the most valuable contribution to this project. The source repository is not public yet — it will be shortly, and this page will say so when it is.

Credits and license

Trend data comes from Languish and PYPL. Language logos and descriptions come from their respective projects and from Wikipedia.

The data lives as one YAML file per node in the repository. Every page has a machine-readable markdown twin (add.md to any URL) and the site is described for agents at /llms.txt.