Reproducibility and Interoperability in Active Data Publication
Institutions
Partners
Description
The project improves the data lifecycle at the interface between data publishing and computational platforms used for data processing and analysis in the ETH domain. We focus on three primary points of interaction: (i) ingestion of published data for downstream analysis; (ii) publication of data products; and (iii) discovery of data use. The three platforms used are the Renku (SDSC/ETHZ/EPFL) collaborative data science platform and two data repository systems, namely EnviDat (WSL) and SciCat (PSI).
The project provides a simple unified interface for accessing both EnviDat and SciCat DOI-ed data in Renku projects in a streaming fashion, requiring no data downloads or additional storage. The use of datasets is tracked on the project level in Renku, allowing researchers to discover how a given dataset is being used. The information about dataset use also propagates from Renku to the data repository pages. Similarly, users can export their data from a Renku project to either EnviDat or SciCat to ensure that the compute environments, raw data, and code used in creating the published data are preserved and accessible to others for full transparency.
Web