Present position: PostDoc Researcher
|Thesis title:||An empirical analysis of the impact of social network structure on the economics of software projects|
|Research area:||Information systems|
As proved by literature results, it is difficult to highlight a direct relationship among software development effort and the internal structure of a software product. In order to explain this phenomenon, two different motivations can be argued: a) investing in quality actually leads to lower development and maintenance effort, but the two terms are a zero-sum, and b) the relationship is not direct, but influenced by other drivers: as a consequence, a 'mediation' effect should be posited.
It has been recognized that the management model of a software development project influences the relationship between development effort and the internal quality (i.e., the software product structure) since it impacts on the coordination mechanisms. However, the governance practices of a software development project cannot be assumed to be discrete: conversely, they are positioned in a continuum between the open and the closed extremes.
In practice, different governance models affect the structure of the social network of developers and team members in different projects: the analysis of the structural properties of the developers' social network can thus provide interesting insights in order to help understanding the nature of the relationship between software development effort and internal structure and quality. For example, greater centralization of developers leads to a higher success of community Open Source projects; however, it causes also a lower quality level of internal structure.
This thesis describes the analysis and empirical test of the following hypotheses in order to better understand the nature of the relationship between the quality of the software internal structure and development effort under different project management models.
On the completely open side (community Open Source projects), the structure of the social network influences the structure of the software product, which in turn drives the productivity of developers (software product acts as a coordination mechanism, since no formal mechanisms are typically set in communities).
On the completely closed side (proprietary software projects), the structures of the network and of the software product co-evolve with a mutual influence. In this case, the driver of productivity is the network structure, as a consequence of the use of formal coordination mechanisms.