Why should I publish data?
Publishing data should always be considered for several reasons. These include improved traceability of results and reusability of data, as well as increased visibility of research and the associated reputational gains.
Furthermore, many stakeholders in the research community, especially funding agencies and publishers, now recommend or mandate the publication of data. On the other hand, there are justified exceptions, such as data protection or commercial exploitation, for which data publication can or must be avoided. In principle, “as open as possible, as closed as necessary” should apply.
How can I publish data?
Data is usually published in data repositories. Here, files are stored together with structured, descriptive metadata. There are repositories that are specifically for the datasets of a particular scientific community. Such subject-specific repositories are often the first choice for data publication for members of the respective community. The re3data.org database is a good place to search for subject-specific repositories. Selected repositories should ensure long-term availability of data, assign persistent identifiers to datasets, and provide the ability to license appropriate data for their own.
Alternatively, there are repositories that are available across disciplines. A special case here are the institutional repositories, which are supported by a research institution in order to allow its own researchers to publish data. The TU Darmstadt provides TUdatalib, an institutional repository for the publication of research data.