What dynamic network metrics can tell us about developer roles

Abstract

Software development is heavily dependent on the participants of the process and their roles within the process. Each developer has his specific skills and interests and hence contributes to the project in a different way. While some programmers work on separate modules, others developers integrate these modules towards the final product. To identify such different groups of people one approach is to work with methods taken from social network analysis. To this end, a social network has to be defined in a suitable way, and appropriate analysis strategies have to be chosen. This paper shows how a network of software developers could be defined based on information in a software repository, and what it can possibly tell about roles of developers (and what not) in the process of the application server Tomcat.

Detecting Communities of Methods Using Dynamic Analysis Data

“…A further take on the network aspect of large-scale software is to consider the collaboration network of software developers who contributed to the development of the software [13], [14], [17], [18]. This approach represents the human development environment of the software, assigning parts of the software to parts of this network, depending on the involved developers.…”

“…More recently through the consideration of network representation of large-scale software a number of other network-based metrics and measures have been considered [3], [11]- [14], [17]- [19]. These metrics and measures in general are based on structural analysis of the network and are associated with the functional parts of the software system which is represented by the analysed structural component.…”

“…A similar approach can provide a classification of developers indicating likely developer roles which may be useful for setting up developer teams in the future [17]. Another approach uses network analysis to assess the level of vulnerability of methods in the context of delivery of functional features of the software [11].…”

“…So network analysis applied to software systems is expected to reduce the complex search for functionally meaningful parts in a large-scale software in some functional context to the search for the structurally important parts of network representation of the software. Network analysis has been applied to develop novel metrics for large-scale software [3], [11]- [14], [17]- [19] and these metrics may be used to assess software quality, error proneness, or for functionality localisation [11]- [14], which are all useful for the understanding and maintenance of large-scale software systems.…”

Visual Analysis of Dynamic Networks Using Change Centrality

“…Other approaches consider the time series of static centrality metrics measured over subsequent time windows (e.g. daily or monthly), and compute basic statistics [14], [15]. In general, new metrics have been proposed for multidimensional networks, in which the attributes of nodes and links chosen as dimensions are not necessarily referring to time, but for example to the type of relation in multi-relational networks.…”