Incorporating Software Curation into Research Data Management Services: Lessons Learned

Abstract

Many large research universities provide research data management (RDM) support services for researchers. These may include support for data management planning, best practices (e.g., organization, support, and storage), archiving, sharing, and publication. However, these data-focused services may under-emphasize the importance of the software that is created to analyse said data. This is problematic for several reasons. First, because software is an integral part of research across all disciplines, it undermines the ability of said research to be understood, verified, and reused by others (and perhaps even the researcher themselves). Second, it may result in less visibility and credit for those involved in creating the software. A third reason is related to stewardship: if there is no clear process for how, when, and where the software associated with research can be accessed and who will be responsible for maintaining such access, important details of the research may be lost over time.

This article presents the process by which the RDM services unit of a large research university addressed the lack of emphasis on software and source code in their existing service offerings. The greatest challenges were related to the need to incorporate software into existing data-oriented service workflows while minimizing additional resources required, and the nascent state of software curation and archiving in a data management context. The problem was addressed from four directions: building an understanding of software curation and preservation from various viewpoints (e.g., video games, software engineering), building a conceptual model of software preservation to guide service decisions, implementing software-related services, and documenting and evaluating the work to build expertise and establish a standard service level.