The distributed and fragmented nature of training materials across research infrastructures, institutions, and within project silos often leads to duplication of materials, wasted resources and storage, the inefficient use of those materials in upskilling personnel, and contributes to the lack of sustainability of the materials themselves. This situation is further exacerbated when considering cross-disciplinary materials, such as those for Research Data Management, which could be used across multiple domains. Several metadata standards are used by training catalogues, including schema.org and schemas.science.
One way to address these challenges would be to offer a federated solution, connecting across those project/institutional communities, and promoting a cohesive strategy towards FAIR and open training. This would additionally facilitate the identification of Learning Paths or trajectories to enable individuals to leverage content across multiple ‘siloed’ materials, to achieve knowledge goals. It could also offer ‘alternative’ paths across the problem space, which would be exposed, recognised and attributed, thereby identifying the original contributors. Many Learning Paths are linear and sequential, lacking legitimate and viable alternative trajectories.
In this project, we will work in parallel, interrelated streams:
- We will develop a strategy to identify similar or related training materials from distributed resources.
- We will interlink related materials as Learning Paths where appropriate, and explore means to facilitate future automation.
- We will improve the schemas.science profiles concerning ‘Course’ and ‘TrainingMaterial’ to propose how those could be updated to incorporate properties that facilitate exposure in Learning Paths.
For the Learning Paths component, we bring together Training content resources and platforms, as well as expertise in federation. For the schema and automation aspects, we bring together expertise in metadata standards and Machine Actionability.
Our use case involves DALIA, a project which makes open educational resources (OER) on data literacy accessible and interoperable. While this material is exposed through a knowledge graph, Learning Paths have not been implemented. To that end, DALIA have developed the moDALIA ontology, which could be leveraged further (includes LearningPath class). Expressing linked materials (such as Learning Paths) could also be extended to other purposes, such as Jupyter notebooks, Workflows and SOP (Standard Operating Procedure) orchestration, giving broader utility beyond just Training Materials. This Use Case also leverages an open source platform (TeSS) for sharing Training Materials, and which implements simple Learning Paths. This work could serve as an exemplar for broader ELIXIR, NFDI, deNBI, ELIXIR-DE and other communities (Projects, Institutes, etc).
de.NBI & ELIXIR-DE goals addressed:
- Contribute to an open source infrastructure: Building on existing FAIR identifiers, metadata standards, ontologies and metadata catalogues.
- Community engagement: Identified fragmentation and repetition of training resources, collaborate on federated solutions.
- Strengthen collaborations: Working directly with NDFI, DALIA, ELIXIR-DE and ELIXIR-UK.
Outcomes:
- A documented strategy and pilot study for identifying and cross-linking related materials, using training resources as an exemplar
- Development of a prototype schema for expressing document or workflow sequence
- Identification and extension of an appropriate ontology to express required relationships for the above
Project co-leads:
Nick Juty - UNIMAN (schemas.science, ELIXIR-UK) -
Petra Steiner - University of Darmstadt (DALIA) -