“Works on My Machine”: A Case Study of Replicability Challenges in Computational Humanities Research

The replicability of computational research is often stated as a key concern in digital humanities scholarship, yet its practical realisation frequently encounters limitations. This contribution analyses some technical conditions for the replicability of articles in the 2024 Computational Humanities Research conference proceedings. A survey was conducted to determine the stated availability of source code, the programming languages employed, and whether and in which way dependencies were declared. The results show that a majority of contributions were not able to make their source code available. Among those that did provide code, many supplied insufficient information on software dependencies to reproduce their computational environments. This circumstance sheds light on infrastructural challenges that complicate the replication of results and, by extension, the review and reuse of these works.