As its bottom-line, the EOSC aims to support three objectives: (1) to increase value of scientific data assets by making them easily available to a greater number of researchers, across disciplines (interdisciplinarity) and borders (EU added value) and (2) to reduce the costs of scientific data management, while (3) ensuring adequate protection of information/personal data according to applicable EU rules (e.g. REGULATION (EU) 2016/679). Indeed, cost reduction is a key element to ensure that resources remain available to carry out the first objective without requiring reductions in the resources necessary to carry out cutting-edge research that will generate the next generation of scientific data. Moreover, an EOSC needs to be developed with the fundamental principles of data protection by design and information security in mind. Further ‘ancillary’ objectives of the EOSC are inclusiveness, bearing a clear value proposition that is simple to adopt, among other attributes.
The idea of creating an EOSC PPP for the future could consider tackling the issue of inclusiveness and assessing the role of commercial entities offering paid services which may run the risk of excluding researchers/ universities/institutes that may face economic/funding issues.
A key part of the governance framework is a set of rules of participation that support these objectives by setting the rights and responsibilities of participants to the EOSC. Collectively, they will help ensure that the objectives of the EOSC, described above, are met.
In relation to the rules of participation, the FAIR principles can be mapped into different parts of the EOSC MVE.
Findability may be a function that is best dealt with centrally, by the ‘core’ of the EOSC. Data producers are responsible for the adoption of a globally unique and persistent identifier assigned to their (meta)data. For harvesting machine-readable metadata and for building up a central index in the MVE, data providers can facilitate findability via the adoption of standardised interfaces to their data repositories.
If (meta)data are not open and free, accessibility may be best embedded in the EOSC MVE via the integration of an existing, generic (i.e. independent from any discipline) authentication and authorisation infrastructure. If for data protection issues, e.g. for highly sensitive data, access to data cannot be fully automated, this infrastructure might offer services to connect the data scientists with the data repository to negotiate the access to the data.
The EOSC MVE may best support interoperability via services which operate with standardised vocabularies (e.g. thesauri) or standardised knowledge representation models (e.g. for an ontology) used to formally describe the (meta)data in a machine-understandable way.
Re-Usability of (meta)data, is a function of the EOSC MVE which is of utmost importance to unfold the full potential of EOSC, particularly on reproducibility. Services of the EOSC MVE may support data producers to provide best descriptions of their (meta)data so that they can be replicated and/or combined in different settings.