On the Approximation of the Substitution Costs for Free/Libre Open Source Software

In this paper we describe our work in progress towards approximating the substitution costs for FLOSS. Substitution cost is the monetary value of the effort necessary for implementing a FLOSS application from scratch in a software company. We describe our approach for estimating substitution cost, based on building generic estimation models for the modern software industry. For this purpose we use the latest version of ISBSG, and apply our models on a subset of projects from a Debian distribution. This yields an approximation of the total substitution costs for this collection of FLOSS applications. We report on various problems, limitations and issues related to the data, the model precision, and the currently available FLOSS project data. However, the anticipated arrival of more coherent FLOSS databases is expected to alleviate many of these issues, and more precise calculations will become feasible, using the same methodology as outlined in this paper.