Creating the EOSC will depend strongly on the ability to connect infrastructures and data and to make them accessible to the community at large in a seamless fashion. Although this goal seems to be commonly accepted, there are lots of questions remaining how infrastructure and data interoperability can be achieved on a large scale.
The aim of the session was to:
- Give a short update on the status of the EOSC Pilot activities on interoperability
- Get the view from high-profile and high-impact stakeholders that are not necessarily already participating in the EOSC pilot on difficulties to achieve an interoperable EOSC
- Derive opinions on what has to be done with highest priority in order to make progress
- Provide information on how the EOSC vision is accompanied by national initiatives
Two panel discussions were organised:
- “E-infrastructure interoperability” with Cristina Duma (INFN / CNAF), Licia Florio (GEANT), and Laurent Crouzet (French ministry MESRI).
- What are the major challenges for e-infrastructures and research infrastructures to be part of an EOSC architecture?
- How does the connection of e-infrastructures in the EOSC context correlate with similar efforts on the national level?
- What advantages do you expect from the EOSC that will be worth doing the effort of the interoperability?
- What should be the next step and/or what is the most urgent action needed to progress towards infrastructure interoperability?
- “Data interoperability” with Françoise Genova (Univ. Strasbourg, RDA), Donatella Castelli (CNR), Carole Goble (Univ. Manchester, ELIXIR), Andrew Treloar (Monash University), Luiz Bonino (GO FAIR).
- What are the synergies between EOSC and RDA (for Françoise) and GO FAIR (for Luiz) ?
- What actions are taken in the context of FAIR data in your countries (for Carole, Andrew, and Donatella) ?
- What should be the next step and/or what is the most urgent action needed to progress towards data interoperability in ESOC ?
The two panels had lively discussions, and we list here some key statements that were made:
e-Infrastructure and research infrastructure interoperability:
- There are technical challenges, but political and social challenges seem to be more important in the task to achieve interoperability
- Particular challenges are governance and long-term sustainability
- Some concern was raised that some ESFRIs might want to maintain their own environment and way of working, which might make their integration in EOSC more difficult
- Next steps should be to define a catalogue of services and to demonstrate their quality and purpose for the different communities
- The governance structure of the EOSC might have to be decided before member states and large infrastructures are able to fully engage into the EOSC
- Building the EOSC should take into account recommendations of the European Interoperability Framework (EIF)
FAIR awareness has to be raised — at researcher level there is little knowledge about FAIR and about the goals of the EOSC.
Many initiatives to become more FAIR and to align with the EOSC are already ongoing on the national level.
This is an opportunity but also an issue in view of the fact that the EOSC is not yet fully defined.
For a successful and FAIR EOSC it is necessary to work closely together and to integrate activities from RDA, GO-FAIR, OpenAire, and other data interoperability efforts.
When thinking about the EOSC, it is important to think globally but to act locally with researchers.
Agreements should be bottom up: first within research domains, then across domains, with real use-cases for interdisciplinarity.
Standardisation and guidelines are key for a successful EOSC.
The end user lacks a “Facebook experience”, i.e. easy to use portals, added value, etc.
CALL FOR ACTIONS
- The EOSC (pilot) seems to have a communication issue and it is important to pass the message to the researchers and other stakeholders:
- What the general goal of the EOSC is
- That the EOSC has not yet been fully defined and not been implemented
- That the EOSC is an evolving program and thus there is still time to influence how it is going to be put into place
- Provide a catalogue of services foreseen to be part of the EOSC to demonstrate its added value
- The EOSC pilot should provide as soon as possible working examples that give users a ‘FaceBook’ experience, i.e. something with added value, easy to use, visually inviting.
Panel discussion on data interoperability with (from left) Volker Beckmann, Françoise Genova, Andrew Treloar, Donatella Castelli, Carole Goble and Luiz Bonino.