An extended and validated dataset of game development problems based on postmortem analysis

This article presents an extended dataset, along with validation records and quality-control procedures, of software engineering problems reported in video game development postmortems published between 1998 and 2025. The dataset combines the Politowski et al. reference corpus with a contemporary extension collected from Game Developer, Reddit, and the GDC Vault. The extension preserves the original taxonomy and coding structure used in the reference corpus. Problem instances were coded using the original taxonomy and validated through a human-in-the-loop review process supported by AI-assisted extraction. The resulting resource contains 267 postmortems and 1227 coded problem instances, together with source information, temporal metadata, and validation records. The dataset includes source metadata, taxonomy labels, supporting quotations, and validation records that may be reused for software engineering and game development research.

All data records, documentation, and coding procedures are provided to facilitate reuse, reproducibility, and future dataset extensions. The complete dataset, source documents, coding records, validation materials, and associated documentation are available through Zenodo (DOI: 10.5281/zenodo.19683430).