Open Science and Research Data Management
Guidelines for Open Science and Research Data Management of the Research Cluster Knowledge Discovery
We, the members of the Knowledge Discovery research focus, are committed to the Open Science idea with all its facets (including Open Access, Open Data, Open Source, Open Methodology) and the FAIR principles of Research Data Management (RDM):
- Findable
- Accessible
- Interoperable
- Reusable
These principles are the cornerstone of transparent scientific work, aiming at open collaboration, verifiability and reusability of scientific content.
We consider research data to be any data that is generated in the course of a research project. In addition, we intend to apply the following guidelines to all other research artifacts, especially code, research data and publications.
With the help of these guidelines, we want to ensure that all members of the research focus have access to all internal research artifacts and all external researchers and interested parties have access to all published research data, code and publications. In doing so, we are committed to the ideas of open access and open science, i.e., we will publish all publications, their associated research data (excluding input data) and other research artifacts open access and, in the case of code, open source.
To ensure compliance with FAIR principles, we will also adhere to the following guidelines:
Findable:
- Use of standard repositories, e.g. Zenodo or subject-specific established repositories
- Creation of meaningful metadata
- Generation of persistent identifiers for research data and publications (e.g. DOI)
- Linking of data and code in publications
- Generation of an overview of papers, research data, code links, and other research artifacts on the research focus website
Accessible:
- Archiving of research data, code and publications according to DFG recommendations for at least 10 years (cf. DFG GWP)
- Accessibility of metadata also via the same persistent identifiers (e.g. DOI) as the data itself (even if the data have been deleted again)
- Ensuring availability of metadata when datasets are not yet available (e.g., embargo periods, etc.)
Interoperable:
- Use of standard file types to the greatest possible extent and avoidance of proprietary formats
- Use of established and appropriate metadata standards (e.g., Dublin Core) with controlled vocabularies
Reusable:
- (Short) documentation of code as well as self-generated data
- Use of clear Open Access licenses (e.g. Creative Commons or OSI-compatible licenses)
Further read:
DFG GWP - Good Scientific Practice (German)