Research software on wings: Automating software publication with rich metadata

Kurzfassung

Publishing your research software in a publication repository is the first step on the path to making your software FAIR! But the publication of just the software itself is not quite enough: To truly enable findability, accessibility and reproducibility, as well as making your software correctly citable and unlock credit for your work, your software publication must come with the rich metadata that support these goals. But where will these metadata come from? And who should compile and publish them? Will RSEs have to become metadata experts as well now? In this talk, we argue that source code repositories and connected platforms often already provide many useful metadata, even if they are distributed over heterogeneous sources. We present an open source software toolchain that will help harvest these metadata, process and collate them, preparing them for submission to publication repositories. This toolchain can be automated via continuous integration platforms, and publish the prepared metadata with or without the respective software artifacts for open and closed source software alike. It can also feed the collated metadata back to source code repositories, or provide them in different formats for further reuse. The talk will outline the concept for the automated publication of research software with rich metadata and describe the current state of the software toolchain that is being developed. It will also detail the CI and publication platforms it will initially be available for, additional resources such as documentation and training materials, and give an outlook on sustainability and future development.