We test claims about people and the past against something outside our own models: inscriptions we held back, machine-checked logic, analyses registered before the data is opened, and the outcomes of real replications. Only what passes is published, with everything needed to check it again.
Now working on
- Classic arguments checked in Lean, with the premises each one cannot do without
- Ancient inscriptions restored and dated, scored on texts held back
- Published studies reproduced from their own data and code
- Open, sourced data on human history