A visual feed of the world's data.

We chase the whole picture. Posts aim to show as much of a dataset as the data allows — every turbine we can map, every earthquake the sensors caught, every cable on the public record. Completeness is an ambition, not a promise: sources have gaps, and where they do, we say so instead of papering over them.

Any form that earns the stare. Most posts are data graphics — maps, swarms, spirals, timelines built from thousands of records. But a post can be whatever tells it best: a flow, an interactive object, a small explorable machine. The bar isn't a chart; the bar is that you can't scroll past it.

The data is real and cited. Every post names its sources — USGS, NOAA, NASA, UCDP, OWID, CelesTrak, and other public archives — and shows the date it was last refreshed. Live feeds (like earthquakes) update daily. Where estimates vary, we use conservative, documented figures and say so.

No login. No algorithm. The feed is chronological. Follow channels to filter it — your follows and likes are stored in your browser, on your device, and nowhere else. Accounts may come someday; surveillance won't.

Made by people and machines. CHARTROOM is a Thrain project. Some posts are built by AI agents that scout trending topics, verify open datasets, and draft new visualizations — every one is reviewed by a human before it ships, and every palette passes a colorblind-safety validator before it's allowed on the site.