Management Accounting for Inner Source

我们建议开发和验证模型和工具原型，以计算内部源代码劳动力并评估内部源代码项目和计划。内部源代码(IS)是在组织内使用开放源代码软件开发实践。采用IS的组织在内部开放其部分软件代码和其他构件，并邀请组织内的所有开发人员参与代码更改。因此，IS打破了开发孤岛，在以前没有协作的组织内部边界之间建立协作。在此项目的第一阶段，我们开发和评估(使用案例研究研究)一种通过测量补丁流来测量IS协作的方法和工具。补丁流是代码贡献跨越组织边界的流动，开发这一基本度量方法的目的是建立模型和工具，用于核算IS劳动力，并在此基础上评估IS项目和计划。这些模型和工具仍然不存在。在这个提议的后续项目中，我们将开发这些模型并典型地实现工具来支持它们。我们将使用多案例研究来验证我们的工作。组织可以使用这些模型和工具来更有效地管理IS项目和计划。我们预计我们的工作将简化信息系统的采用和管理，并将使软件开发更加有效和高效。结合IS经验报告、经济学文献中的模型和我们自己的项目经验，我们将开发一个IS劳动力的会计模型，并开发一个实现该模型的工具原型。模型和工具将能够提供组织内IS劳动力的聚合视图(工作包α)。以该工具为载体，我们将在三个大型软件开发组织(Work Packageβ)的多案例研究中验证会计模型的可行性和实用性。随后，我们将为IS项目和计划开发一个评估模型。从文献和专家访谈中，我们将确定IS项目的质量属性。我们将根据已确定的质量属性(工作包γ)设计和实施指标来评估IS计划。我们将通过对三个大型组织的第二个多案例研究(Work Packageδ)来验证评估模型的可行性和实用性。所提出的模型将是第一个考虑组织内的IS劳动力和评估IS的模型。

We propose to develop and validate models and tool prototypes for accounting for inner source labor and evaluating inner source projects and programs.Inner source (IS) is the use of open source software development practices within an organization. An organization adopting IS opens up parts of its software code and other artifacts internally and invites all developers within the organization to contribute code changes. Thus, IS breaks down development silos and establishes collaboration across intra-organizational boundaries where no collaboration existed before.In the first phase of this project we developed and evaluated (using case-study research) a method and tool for measuring IS collaboration by measuring the patch-flow. Patch-flow is the flow of code contributions across organizational boundaries.The purpose of developing this basic measurement method was to base models and tools for accounting for IS labor and for evaluating IS projects and programs on it. These models and tools still do not exist. In this proposed follow-up project, we will develop these models and prototypically implement tools to support them. We will validate our work using multiple-case case study research. Organizations can use these models and tools to more effectively manage IS projects and programs. We expect that our work will simplify IS adoption and management and will make software development more effective and efficient.The proposed work is structured into four work packages. Combining IS experience reports with models from economics literature and our own project experiences, we will develop an accounting model of IS labor and develop a tool prototype implementing the model. Model and tool will be capable to provide an aggregated view of IS labor within an organization (work package α). Using the tool as a vehicle, we will validate the viability and utility of the accounting model in a multiple-case case study with three large software developing organizations (work package β). Subsequently, we will develop an evaluation model for IS projects and programs. From literature and expert interviews, we will identify quality attributes of IS programs. We will design and implement metrics to evaluate an IS program in regards to the identified quality attributes (work package γ). We will validate the viability and utility of the evaluation model with a second multiple-case case study with three large organizations (work package δ).The proposed models will be the first of their kind to account for IS labor and evaluate IS within an organization.