What is Research Software?

  • For hundreds of years the basis for science are theory and experiments; since a about 50 years simulations allow to investigate questions where experiments are too expensive, too complex, too long-lasting or can not be done. Since about 10 years the extraction of knowledge from data is seen as the 4th paradigm in science.
  • Hence, software in research is a key component of scientific work and scientists have today the same rigorous requirements towards research software as they have towards data, devices, and technologies that they are using. Research software have become valuable assets in many research disciplines and can be seen as research infrastructure itself, which need to be financed and further developed in a long-lasting way like other research infrastructures, e.g. research vessels and stations, particle accelerators, laboratories and the like.

  • The importance of research software has e.g. been addressed in a strategy paper by the Netherlands eScience Center in 2019 entitled “Raising the Profile of Research Software: Recommendations for Funding Agencies and Research Institutions”, which stats “If open science is to truly lead towards better, more transparent, and reproducible research, then research software needs to be treated in equal footing to research data and publications at the policy level and in practice.”

  • The Software Sustainability Institute has put it simply in short: “Better Software, Better Research”

What is Research Software Engineering?

  • Many answers exist to this question, so let’s look at what is written on Wikipedia: “Research software engineering is the use of software engineering practices in research applications . The term was proposed in a research paper in 2010 in response to an empirical survey on tools used for software development in research projects. projects. It started to be used in United Kingdom in 2012, when it was needed to define the type of software development needed in research. This focuses on reproducibility reusability, and accuracy of data analysis and applications created for research. research.”
  • This encompasses a sustainable/long-lasting paradigm shift in developing software for research through using Continuous Integration/Testing/Delivery (CI/CT/CD), through educating, training and promoting (young) Research Software Engineers, through adapting common key performance metrics and indicators to reflect the importance of research software, as well as through a co-design approach with an early integration of computers scientists, mathematicians and – possible  also hardware operators in the development of community software in science.

  • At KIT the importance of research software and research software engineering is reflected in “Dachstrategie KIT 2025”, which states in chapter 9 about “Digitalisierung” under aim 9.1.5 (page 33, in German only): “KIT understands Research Software Engineering (RSE) and the development of research software as an essential task in the digitized research process and strives for new research projects, which are increasingly data and computation intensive.

  • To address this aim a KIT-internal small project without funding, called “Leitprojekt 9.1.4”, was initiated. More information about this RSE-related project as well as the other projects in the domain of “Digitalisierung” is here, again only in German.