- Dr. Alexander Schlemmer, Max-Planck-Institute für Dynamik und Selbstorganisation, CaosDB
- Frank Tristram, Karlsruhe Institute of Technology, 3DMM2O
The key objective of the task area FAIR Lab is to FAIRify local workflows of data generation and through that, ease administration and lookup during the daily work with data. The guiding idea is to design scientific data management workflows on the local scale such that best practices for handling research data are easy to introduce and beneficial for the daily research work. At the interface to repositories, the optimization of local workflows highly facilitates findability and reuse on a larger scale. The task area consists of two core parts, focusing on the digital (file) object on the one hand and the local handling system on the other.
Information extraction and processing focuses on the data object itself, typically starting as raw (instrument) data that is ingested into a digital research workflow (analysis, transformations etc.). The data has to meet several requirements to be become a FAIR object:
- transformable into other (file) representations without writing new code to ensure interoperability
- findable via the used workflow and steps (metadata extraction)
- publishable with as few addional manual steps as possible
In physics a lot of data is generated by sources that differ from group to group or insturment to instrument. Even if a standard for this obects would exist, one would have to transform to that standard for each device and software version as long as this standard is not supported by the manufacturer. With NFDI4Phys, we want to solve this problem by providing low-code data readers that can retrieve information and transform files, such that the user only needs to confirm the structure of the data that the tools automatically analyzed.
Typically input from different file handlers will be passed to the local Research Data Management Software (RDMS). This shall automate a first structured data collection at the beginning of data generation.
The Local RDMS that needs to be filled with daily unconsolidated information, but has FAIR interfaces to directly connect with standardized repositories.
We want to suggest interfaces and modular tools for the following goals:
- Classify and index new raw data, results of analyses and algorithms upon first use (automatically)
- Make all local digital research items findable and queryable in a structured way for further processing within a collaborative environment
- Link information in a way that starting from a publication the full chain of analysis steps can be traced back to the raw data in order to guarantee reproducibility.
To allow work groups and laboratories to achieve the Task Areas goals, FAIR Laboratory will establish a platform for guiding the users to setup and maintain a local RDMS. A variety of different use cases from different domains of physics will be mentored in order to develop best practices that will result in a set of manuals, documentation and software packages which can be used to assist the users in the adoption of a local RDMS. Furthermore, this task area will provide consulting on design of workflows, setup of data models and devlopment of data integration tools.
In close relation with Task Area 3, the interchange between local descriptions and global standardized metadata and ontologies will be enabled. Task Area 2 will provide another important linkage by making registries for the local objects available that can refer to open or closed data of the local RDMS.