Publications

Master Thesis:

2015

Quantitative and qualitative analysis of the processes of development of video games vs. traditional software in open source.

Abstract

The video games development plays an important role in software industry, yet the software engineering community rarely studies its processes. This limitation can decrease the video games quality and productivity such as some games development expert has suggested. This Master’s Thesis studies the differences that can be found in the video games vs. traditional software development.
The work was based on two approaches: quantitative and qualitative study. The first one explores and categorizes the distribution of some objects in GIT repositories, like: files, commits, authors, bugs, unit tests, etc. The second one, analyzes the qualitative results of ad hoc surveys sent to 600 participants, investigating over: software design, software maintenance, software testing, required skills, problem solving, etc.
Resources were extracted from 30 video games and 30 traditional software, both from open source world, including a wide classes of them. The analysis of these entities in the time (fixed periods or releases based) shows different behavior in software development, proving implications for research, industry and practice. For instance, as a starting point for impacting game development, researchers could create testing tools that enable game developers to create tests that assert flexible behavior, with little up-front investment.
Future work would explore the possibility of an alternative bug classification, in order to create new test cases that would improve productivity and fault rate detection.

Back to Top