Starting FDO in the Cradle -- ROcrating Live Data

This talk discusses the use of Fair Digital Objects (FDOs for short) a democratised approach to FAIRness, that is, adherence Findable/Accessible/Interoperable/Reusable set requirements, collectively called FAIR. capability is being built CS3MESH4EOSC project. 3-year EU-funded project in EOSC context (we started Feb 2020) addresses challenges fragmentation file and application services, digital sovereignty FAIR principles everyday practice researchers. Initially, 7 major data services will be combined into ScienceMesh - federated service mesh providing frictionless collaboration platform hundreds thousands users (researchers, engineers, students staff). The offer easy access across institutional geographical boundaries. infrastructure gradually expanded offered entire education research community Europe beyond. initial connect NL (SURFdrive), PL (PSNCBox), AU (CloudStor), DE (Sciebo), CZ (owncloud@CESNET), CH (SWITCHdrive) DK (ScienceData), as well domain stores at CERN (CERNBox) EU's own Joint Research Centre's Copernicus (earth observation) datastore. busy designing, building deploying necessary technology achieve this. grew out grassroots "CS3" community, which self-help forum builders providers from academic sector who look after rapidly growing datastores "synch-'n'-share" paradigm (dropbox commercial equivalent); this type store more popular basic block live storage research. mission improve scientific (essentially an interoperating federation stores), ensure sharing resulting done according principles. puts unique position: we need bring tooling front broad audience (not just "FAIR literate" ones), convice them FAIRness relevant point where collected, not when has congealed collections. We recognise two main obstacles: FAIR-aware needs simply available, right every user's face, so usable it gets uptake. By rule thumb additional step required sheds half userbase you start with. communities motivated, trained assisted infrastructure. It make their lives easier. Without place, there no mounting awareness campaigns. Therefore, CS3MESH4EOSC's uptake Science Mesh described first paragraph, already widespread by add Description Service these stores, available any researcher system (an instance "Describo" tool, https://arkisto-platform.github.io/tools/description/describo-online/). Thus they can create Object packages (using RO-Crate standard) also manage deposition process, initially targeting open Zenodo Dataverse repository Open Framework (OSF) science workflow portal. metadata annotation user guidance wizards facilitate process "ScieboRDS" (https://www.research-data-services.org/page/about/). thus adding capabilities several tens Petabytes on it, end up with democratised, low-barrier-of-entry Allowing researchers generate FDOs (no onerous upload / collections steps) help supply. far are deployed starting Q3 2022. Further development underway allows better negotiation between ("the Mesh") backend repository, such rely schema autoprovisioned ontologies agreed upon before FDO packaged, improving quality.