Addressing Video Game Development Challenges Using Industry Best Practices

2019 – ongoing

Summary

The aim of this project is to investigate both industry and academic data on video game development (VGD) to construct an empirical model of the domain that can be used to analyze the relationship between the contexts, success factors, ways of working, and challenges of game projects. This is to identify the reasons behind the team collaboration and leadership problems, and learn how to address them with more effective processes and practices.

To understand current ways of working and challenges in VGD, we surveyed and interviewed dozens of studios and developers, as well conducted a tertiary literature review on industry Agile/Lean practices. These investigations revealed the core problem in VGD seems to be an incoherent understanding of the domain due to a lack of a shared terminology between game developers, academics, and players to describe it. To help address this fundamental barrier to collaboration and project management, we constructed a descriptive classification domain model of VGD terms and concepts using a novel computer-aided text analytics dictionary building methodology.

The novel methodology employs a hybrid theory-driven and data-driven approach that integrates frameworks and best practices from evidence-based software engineering, data science, topic modeling, and ontology integration, among others. Terms describing VGD concepts, practices, and processes were from a corpus of 240 empirical industry project post-mortems and academic studies, and the resultant prototype dictionary has about 11 subjects, 67 knowledge areas, 302 topics, 1,590 concepts, and 12,705 entries. The dictionary has been validated on seven quality dimensions and its applications include terminological study, trend analysis, characterization, concept co-occurrence analysis, sentiment polarity orientation, practice recommendations, document classification, and as a conceptual framework and vocabulary reference.

Findings indicate game developers tend to weight decision making towards the team, followed by the design, engineering, and production disciplines. However, game developers typically fail to show the rationale behind their team’s decision-making, which appears to have a strong relationship with the collaboration and leadership challenges they face. Also, Agile/Lean frameworks are rarely discussed by game developers, which suggests they are not highly valued. However, studios that do use Agile/Lean principles or practices appear to benefit from them, even if they do not always fully understand them. Analysis also confirms the lack of shared understanding and common terminology between key stakeholders appears to be the root problem from which other issues arise.

Our current work includes refining the VGD dictionary and applying it in industry contexts to help inform team and management decision-making. We are also conducting case studies, workshops, and coaching sessions with industry game studios to further investigate the patterns between their contexts, success goals, ways of working, and project outcomes.

 

Acknowledgements

This study received funding from the University of Canterbury as part of the Department of Computer Science and Software Engineering, School of Product Design, and the Applied Immersive Gaming Initiative (AIGI).

This project received a grant from the Early Career Researcher Development Fund hosted by Research and Innovation, University of Canterbury.

We would also like to thank the NZGDA, Christchurch Game Developers Association, Black Salt Games, and other NZ game studios and developers for their ongoing support of this research.

 

Contact

Timothy McKenzie

 

Publications

McKenzie, T, Morales-Trujillo, M, Lukosch, S, Hoermann, S. (2019). Software engineering practices and methods in the game development industry, in Extended Abstracts of the Annual Symposium on Computer-Human Interaction in Play Companion Extended Abstracts, Association for Computing Machinery, New York, NY, USA. 181–193. https://doi.org/10.1145/3341215.3354647

McKenzie, T, Morales-Trujillo, M, Lukosch, S, Hoermann, S. (2021). Is Agile Not Agile Enough? A Study on How Agile is Applied and Misapplied in the Video Game Development Industry, in IEEE/ACM Joint 15th International Conference on Software and System Processes, ACM/IEEE, Madrid. 94–105. https://doi.org/10.1109/ICSSP-ICGSE52873.2021.00019 

People

Timothy McKenzie, Dept of Computer Science and Software Engineering | HIT Lab NZ

Stephan Lukosch, HIT Lab NZ

Simon Hoermann, School of Product Design

Miguel Morales-Trujillo, Dept of Computer Science and Software Engineering