Category: 2.5 Inner Source

Abstract: Inner source software development is the use of open source development’s best practices inside a company. In inner source, developers collaborate on reusable software components across company-internal organizational silo boundaries for mutual benefit. As such, inner source goes against the grain of traditional management techniques. In this article, we present two conceptual models of management accounting for inner source. We derived these prototypes by performing a literature review and triangulating the results with interviews of industry practitioners. We demonstrate how the conceptual models can be used for monitoring and controlling inner source projects and to determine their future viability.

Abstract: Inner source is the use of open-source practices within companies. It enables more efficient software development, shortens time-to-market, and lowers costs through increased company-internal collaboration. While existing studies examine social and organizational impact factors on inner source adoption, only a few have looked at measuring and economically assessing inner source. This article presents an overview of current research regarding inner source, its measurement, economic assessment, and impact on businesses and their processes. Based on a systematic literature review we build a research model for economic inner source assessment. This research model shows thematic dependencies between the economic impact of inner source and its measurement. Additionally, it proposes research questions and hypotheses on measuring, economically assessing, and subsequently adopting inner source.

Abstract: A key part of taxation, controlling, and management of international collaborative programming workflows is determining the costs of a supplied software artifact. The OECD suggests the use of the Cost Plus method for calculating these costs. However, in the past, this method has been implemented using only coarse-grain data from the costs of whole organizational units. Due to the move to inner source software development, we need a much more fine-grain solution for computing the detailed time spent on programming specific components. This is necessary, because a more accurate work time distribution is required to fulfill the fiscal and administrative challenges posed by collaborating across organizational boundaries. In this article, we present a novel method to determine the time spent on an individual code contribution (commit) to a software component for use within cost calculation, especially for taxation purposes. We demonstrate the usefulness of our approach by application to a real-world data set gathered at a large multi-national corporation. We evaluate our work through feedback received from this corporation and from the German Ministry of Finance.