IG Research Software

The IG Research Software provides support for and exchange of experiences on research software, i.e., sustainability, attribution, archival, discovery, review, collaboration and computational reproducibility, i.e., computing environments, transparency, validation, packaging, research compendia, reuse) as the counterpieces of research data. Scientific knowledge generated for the benefit of society does not only rely on the community-wide application of open access and FAIR data publication practices, but also strongly on how research software is used in a community. Not only research data, but also software tools are fundamentally connected to scientific analyses and thus insight, therefore both also must be considered for transparent and reproducible research so that findings are understandable, reviewable, and reusable.
The IG provides a forum for discourse about research software in and for NFDI4Earth and the IG members use the cross-cutting nature of software to connect and collaborate across different disciplines, task areas, and institutions within the consortium.
The IG offers a specialised mutual exchange between peers on complex and challenging technical topics and explicitly targets people who would usually not directly participate in NFDI4Earth research or have leadership roles, but might  primarily support scientists in conducting their research. We welcome everyone who works with/on research software at any level, from experts in research software engineering (RSEng) to individuals interested in  reproducible research. The IG's scope includes software used to build the NFDI, software created as part of NFDI4Earth activities, as well as software used for research by any researchers in the national Earth System Science context. The IG may reach out and collaborate with similar groups in other NFDI consortia. Our understanding of research software starts with the smallest units possible, such as a personal program, script or macro, written in general purpose languages such as Bash, Python, R, Mathematica/Matlab etc., over the application of, e.g., Jupyter notebooks for data analysis tasks, all the way up to simulation codes running on, e.g., HPC systems. These programs may be used to convert or analyse a small, individual dataset or solve Big Data, data streams, or integrated time series data analysis tasks, or may produce data in various simulations of Earth System Models.

The perspective of research software and people building research software is important to support the cultural change towards more openness, transparency, and reusability of research outcomes. Furthermore, reproducibility is an especially important topic in research that is closely connected with research software running in virtual research environments (from laptops to unique HPC clusters). We aim to represent all kinds of RS aspects in this interest group.

At the core of our IG is the work with concrete applications, exchange of experience, knowledge, problem solving, etc. The collaborations can be built around application examples, applications themselves, or best practices. Hence, the concrete application used for an Earth science task takes the center of our efforts. In addition, the IG can initiate collaborations on common tools, e.g., filters, in particular across the various ESS domains – or even across NFDI domains. RS has similar requirements as data regarding meta-data, standards, or FAIR criteria. RS must be transparent, well-documented, and ideally open. Such generic aspects and much needed infrastructure, e.g., repositories or databases are likely to be jointly discussed and developed with other NFDI consortia or even internationally. For these connections, the IG aims to become both an internal and public contact point for all questions surrounding NFDI4Earth RS.