With Jim Clifford, Jessica Jack, Mark Humphries and Lianne C. Leddy, I benchmarked optical character recognition on historical documents from 1612 to 1921. The results are published as a working paper in Working Papers in Critical Search, no. 4 (2026).
Reading the Archive by Machine: An OCR Benchmark for Historians, 1612–1921
A benchmark of how well current OCR reads historical documents from 1612 to 1921, and what its errors mean for historical research.