Digital Source Criticism and the Versioned Apparatus: Why Historical Research Needs Git-Based Infrastructures

Arbeitsbereiches Digitalität (DH Labs IEG), Leibniz Institute of European History Mainz

Invited Talk at Arbeitsbereiches Digitalität (DH Labs IEG), Leibniz Institute of European History Mainz

This talk approaches Git-based platforms such as GitHub, GitLab, and Codeberg not merely as tools for project management, but as infrastructures with methodological significance for historical research. It argues that Git-based workflows can support and extend central practices of historical scholarship: version histories make the development of data classifications, textual changes, modelling decisions, and arguments traceable; branching and forking allow competing interpretations, data models, or visualisations to be developed as experimental variants; issues and pull requests create transparent spaces for collaborative criticism directly at the level of data, scripts, documentation, and text. In this sense, Git can be understood not simply as a technical aid, but as a versioned apparatus for digital source criticism.

The practical thread running through the talk will be the project Stadt.Geschichte.Basel, a workflow for producing reproducible figures, structured metadata, and publishable documentation from historical data (project documentation: https://dokumentation.stadtgeschichtebasel.ch/sgb-figures/; overview of plots and visualisations: https://dokumentation.stadtgeschichtebasel.ch/sgb-figures/docs/plots.html).

As a contrast, the talk will briefly discuss a project on sexual abuse in the context of the Catholic Church in Switzerland, which combines a technical replication framework with a reflection on the limits of openness when working with sensitive historical data (article: https://zfdg.de/2026_006; technical framework and project documentation: https://maehr.github.io/Sexueller-Missbrauch-im-Umfeld-der-katholischen-Kirche-in-der-Schweiz/).

This case shows that methodological transparency does not always mean full disclosure: when working with sensitive research data, data protection, source protection, and research ethics have to be reconciled with the demand for scholarly traceability — for instance through technical frameworks, documentation, and synthetic sample data. The talk does not require prior knowledge of Git; it focuses not on terminal commands, but on the question of how digital historical research can become more traceable, collaborative, and publishable.