Latour's inscriptions and software development

Following my thoughts about Latour's inscription (see last week's post), I ran across this good paper about distributed software development and the link with 'inscriptions'. Latour's inscriptions are about "social arrangements, debates, divisions of labor, and patterns of work become inscribed into the artifacts and representations in which science trucks". In the context of software development, they want to study the relationship between technological artifacts and the social structures that shape them.

De Souza, C., Froehlich, J., and Dourish, P. 2005. Seeking the Source: Software Source Code as a Social and Technical Artifact. Proc. ACM Conf. Supporting Group Work GROUP 2005 (Sanibel Island, FL.)

Our work has been motivated by the question of whether aspects of informal software process can be found in the structure of the software artifact itself. Using a software visualization tool, Augur, we have been conducting an analysis of the artifacts of a number of software projects, a “software archeology” to explore the relationships between artifacts and activities as they are negotiated in distributed software development through mining software repositories. (...) Each pane displays a different aspect of the system being examined: changes in one view are immediately reflected in the others. The large central pane shows the line-oriented view of the source code. In the figure, the color of each pixel line indicates how recently it was modified; this allows a developer, at a glance, to see how much activity has taken place recently and where that activity has been located.

The conclusions are as follows:

Distributed software development presents two sources of complexity to its participants – the complexity of the software artifacts under development, and the complexity of the process of developing those artifacts. We have presented a study of software artifacts, conducted using a visualization tool, which demonstrates how these twin sources of complexity are intertwined. Software artifacts are not merely the objects of software development processes, but are also the means by which those processes are enacted and regulated. The structure of the artifact both reflects the processes by which it has been created and can be used to control those processes by centralizing points of access, by regulating the relationships between independent activities, and by making visible the relationships between individuals. It is a means, then, by which the articulation work of the project can be carried out.

Why do I blog this? I am fascinated by this: how technological artifacts and social structures might shape a certain phenomenon such as product development.