FAIR data future

Progress in science strongly depends on advanced, reliable, accessible management, processing and interpretation of data across disciplinary boundaries. The ever-increasing quantity of data gained from experiments and simulations require transparently structured and machine-actionable infrastructures, concisely summarized by the FAIR (findable, accessible, interoperable, and reusable) principles .

NFDI4Phys complements NFDI

The consortium NFDI4Phys represents a scientific community of diverse small and medium-sized physics groups, especially including the emerging quantum computing alliance and transdisciplinary physics (i.e., applications of physical methods in fields outside of physics). Similar to other scientific disciplines, we observe a clear tendency towards dynamically evolving infrastructures and multidisciplinary research, leading to progressing heterogeneity in work flows, methods, and data formats. The resulting intra-, inter- and/or transdisciplinary barriers seriously obstruct scientific innovations. The aim of NFDI4Phys is therefore to facilitate FAIR Research Data Management (RDM) to academia with 22 co-applicant institutions with over hundreds of members. NFDI4Phys has a strong embedding in an actively involved multi-disciplinary international community engaging all relevant stakeholders. The community and envisioned RDM strategy is complementary to available consortia and significantly strengthens the structure of the NFDI framework and academic research system. The decentralized operating model of NFDI4Phys ensures longevity and sustainability of the delivered services.

State of the art

Today’s large-scale data management solutions are mostly designed for homogeneous data collections adhering to standardized and a-priori predefined data models designed by data-modeling experts thus barely meeting the aforementioned needs of the NFDI4Phys community. Hence, numerous individualized, and often mutually incompatible, data management solutions co-exist within and across different disciplines. This lack of community-wide agreements on software and data formats impedes interoperability, sustainability and flexibility to adapt to the demands of continuously evolving scientific environments as proliferated by the FAIR principles.

Building bridges

NFDI4Phys will significantly contribute to the overarching NFDI goals by integrating data and analyses capabilities into work flows across databases, disciplines and countries. Our vision is a FAIR RDM methodology being as simple as accessing the internet: similar to the TCP/IP standard revolutionizing data exchange at bit level 40 years ago, the emerging FAIR digital objects (FDOs) will become the corresponding standard for enforcing FAIR information exchange. However, in order to utilize FDO as conceptual backbone for NFDI4Phys, respective research data first needs to be characterized philosophically (Floridi 2014 ) as well as in terms of physical measures of their content (Kolchinsky & Wolpert 2018 ) in order to capture semantic characteristics of information across disciplines, e.g., physics (order), biology (function), philosophy (meaning).

Task Areas & Objectives

We propose ten task areas (TA) : FAIR Laboratory, Metadata & Ontologies, Federated Repositories, Quality & Standards, Evolving Infrastructures, Data Literacy, Base RDM, Surveys, Domains, and Governance.

To reach the described goals, NFDI4Phys will address following key objectives:

  • Enable research groups to make their data manageable in form of FDOs throughout the entire data life-cycle.
  • Foster standardization processes for data models and ontologies to strengthen interoperability of metadata, schemata and ontologies.
  • Improve reusability of data through quality criteria and standards, promoting open data formats and research software as well as domain data protocols.
  • Design federated repositories to allow seamless exchange of FDOs and data models.
  • Develop and distribute a best practices RDM template that enables current and prospective users to set up and maintain FAIR data management.

Impact & leverage

The NFDI4Phys community represents nine domains of Physics (Atoms and Molecules, Optics & Photonics, Cold Plasma, Biological Physics, Dynamics, Statistical Physics & Soft Matter, Socio-Economic Physics, Quantum Information & Artificial Intelligence, Biomedical Physics, and Transdisciplinary Physics) which, together with those of three other NFDI physics consortia, cover all of disciplinary physics. In addition, Transdisciplinary Physics further include physicists not rooted in the German Physical Society. According to the hierarchy in Nature (Anderson 1972 ), research data within as well as across all nine domains can be conceptually encapsulated and linked to form a modular hierarchy as mirrored by the underlying mathematical structure of FDOs. To foster this idea, we strive for a dual infrastructure involving RDM and transdisciplinary discussion groups, for example, on the structure of information. The NFDI is a fantastic opportunity for physics to develop further by entering the semantic space. To summarize, the envisioned approach of NFDI4Phys addresses the recent needs of a large and heterogeneous community enabling self-sustained, dynamically evolving, and FAIR RDM.