Problems, Needs, and Recommendations

Semantic Layer

S1 — Lack of common definitions

Problem	Needs	Recommendations
There is a generalised lack of common explicit definitions about the terms used by user communities.	Need for harmonisation across disciplines.	Communities should generate clear, precise definitions for concepts, metadata, and data schemas. Definitions must be public, PID‑referenced, and shared in EOSC.

S2 — Lack of shared semantic artefacts

Problem	Needs	Recommendations
Common semantic artefacts (ontologies, schemas) are often missing across communities.	Need for principled approaches and tools for ontology and metadata schema creation, maintenance, governance, and use.	Semantic artefacts should be openly licensed and accessible.

S3 — Lack of reference repositories for semantic artefacts

Problem	Needs	Recommendations
There is a generalised lack of common reference repositories or registries of semantic artefacts.	Need to harmonise data of the same type.	Every semantic artefact in EOSC must include sufficient documentation, examples, and conceptual diagrams.

S4 — Poor documentation and fragmented metadata schemas

Problem	Needs	Recommendations
Data collections are poorly documented; no common metadata schema across communities.	EOSC should support a repository of semantic artefacts and a governance framework.	Propose a minimum metadata model (DCAT‑AP, DDI 4 Core, DataCite, OpenAIRE). Extend to software, workflows, protocols, hardware. Provide federation/harvesting building blocks.

S5 — Lack of semantic expertise in communities

Problem	Needs	Recommendations
Some communities lack semantic expertise and skills.	Capacity building and training.	Develop training, guidance, and community support structures to improve semantic literacy.

Technical Layer

T1 — Fragmented authentication and authorisation

Problem	Needs	Recommendations
Authentication and authorisation must often be repeated for each community or service.	Seamless authentication and rights acquisition for EOSC services.	Use open specifications to ensure technical interoperability when establishing EOSC services.

T2 — Heterogeneous data formats

Problem	Needs	Recommendations
Research data exists in many formats (CSV, JSON, XML, NetCDF, FITS, etc.).	A trusted, sustainable cross‑community framework; a minimum metadata application profile for EOSC.	Define a common security and privacy framework and establish trustworthy data‑exchange processes.

T3 — Difficulty finding data at different granularities

Problem	Needs	Recommendations
Coarse‑grained or fine‑grained data from other communities may be difficult to find.	Discovery mechanisms supporting multiple levels of granularity.	Provide clear, understandable SLAs for all EOSC resource providers.

T4 — PID fragmentation and inconsistency

Problem	Needs	Recommendations
Multiple PID providers with different policies; some PIDs not resolvable; duplicates exist.	A common, well‑understood PID policy across communities.	EOSC must enable easy access to heterogeneous data formats and provide tools to overcome differences. Provide coarse‑ and fine‑grained search tools. Establish a clear EOSC PID policy accommodating diverse practices.