'Directory Pipeline'—A Tool for Turning Historical Digital Collections into Structured Data
The folks at CNI recently saw a message I wrote on a private email list about the work I'm doing on my "Directory Pipeline," and they very graciously invited me to record a brief presentation on it for the Spring 2026 CNI video briefs series.
It was a great opportunity to lay out some of the underlying ideas I've been thinking about, and show some basic demos of what the Directory Pipeline can output.
The repo is at https://github.com/hadro/directory-pipeline/.
The example I focus on in the slides is the Woods Directory Data Explorer.
The post is now live, with video embedded below, and the slides are also available.