You recently published data with TUDOdata, TU Dortmund University's data repository. What is the added value of publishing data in an institutional repository?

PD Dr. Kerstin Wohlgemuth: First of all, we want to secure and store our data according to the FAIR principles, i.e. it should be findable, accessible, interoperable and reusable. In our working group, we have decided that we also want to make the data for each publication available as a data publication, enabling all researchers and interested parties to reuse it. We use TUDOdata for this because the data is stored here in accordance with the FAIR principles. For example, this TU-owned repository offers an appropriate storage period and findability via DOI (Digital Object Identifier). This allows us to reference the data in the publication itself. There are certainly many other repositories out there, but thanks to the support of the Research Data Service team, we can be confident that we are doing everything correctly.

How did you manage to make the data understandable for other researchers?

This is actually the crucial point that presented my employees with major challenges: presenting the research in such a way that it can be understood. I am the first person to check the comprehensibility at a scientific level. And Bernd Zey from the Research Data Service also supports us and, as a non-specialist, checks at a technical level whether he can understand what we are writing and whether all the necessary documents and fields have been prepared and filled in comprehensibly. This outside perspective is a great help.

There is already a vocabulary in the community for comprehensible documentation at content level, which we adopt from existing publications. But at the same time, we are also involved with innovative devices and methods for which we design the vocabulary ourselves and establish it in the literature. We think in advance about what is comprehensible and cannot be confused with other, already existing vocabularies, and we strive for clear definitions in the publications because we recognise that the same vocabulary is used for different things in different communities.

The pioneer in my working group is Ronja Heming, who was the first to publish data and has now published three data sets. She is also currently developing instructions for TUDOdata for subsequent doctoral students in my working group. This will certainly make things easier for them because they can start with completely different standards. We specify standards for storing data in Bachelor's and Master's theses, and we have written readme files: What is to be designed and how, how to name it and what may and may not be done. Everyone in our working group now benefits from this preparatory work.

What would you like to pass on to your colleagues who are thinking about publishing data?

Ideally, you should think about how to store data and how to structure and name it before the project starts. A policy helps you to consider this from the outset. To really get the best out of a publication, you should always publish it together with a data publication. This ensures the reusability of data. If this becomes standard practicec at some point, everyone will benefit because it will be much easier to compare your own data with the data of others. We currently lack this, for example when comparing devices. I hope that in the future, we will be able to conduct research in a more transparent way within the research community.