Catalogue of Nazi Camps in Austria

The Austrian Federal Monuments Authority documents the camps, forced labour sites and other places of persecution across Austria in a catalogue of considerable size. It exists as a PDF — readable by people, but not analysable by research. Searching for a municipality, mapping sites, or comparing holdings all begin with retyping.

The task

Bring the catalogue into a form that can be searched, mapped and connected to other datasets — without sacrificing the precision of the original.

The approach

Apache Tika first converts the PDF into XHTML. A parser then breaks the running text into clearly separated fields: federal state, municipality, camp designation, history and inmates, location with address and cadastral community, land parcel numbers, WGS84 coordinates, and bibliographic references. The output exists as both CSV and JSON.

A web application builds on that: an OpenStreetMap view with markers coloured by federal state and clustered at low zoom levels, a sortable table, full-text search across all descriptive fields, and a detail panel for the complete entry.

The outcome

2,069 catalogued entries across nine federal states, 1,262 of them with coordinates and 1,922 with a cadastral reference. The path from PDF to dataset is documented as a script and can be repeated with every new edition of the catalogue.