abstract
- Finding an appropriate repository for publishing research data remains a challenge for many scholars across disciplines. While general-purpose repositories are widely available, subject-specific repositories offer advantages such as improved discoverability, enhanced visibility, and domain-specific support. However, such infrastructures are often lacking in many research areas, including mobility and transport research. As highlighted by Lin et al., this gap presents a barrier to achieving FAIR data principles and limits the reuse and impact of research outputs. The Specialized Information Service for Mobility and Transport Research (FID move) addresses this challenge by supporting researchers with sustainable research data management tailored to the traffic and transport community. Through consultation, training, guidelines, and assistance in publishing subject-specific datasets, FID move promotes good scientific practice. Its research data repository enables the archiving and publication of transport-related data, such as survey results, measurement campaigns, and observational studies. Each dataset is assigned a persistent Digital Object Identifier (DOI), ensuring long-term findability and citability. To enhance discoverability and interoperability, the repository employs standardized metadata aligned with the latest DataCite schema. A dedicated plugin enables automated DOI assignment and metadata exchange via the DataCite API, ensuring seamless integration with external catalogs and portals. To promote FAIR and open data, FID move recommends using open licenses to clarify reuse conditions. The repository represents institutions as "organizations" and associates users accordingly. It introduces "Groups", cross-institutional collections of datasets created for collaborative projects. Groups facilitate shared management and improve the discoverability of related datasets. Users can subscribe to updates from organizations, groups, users, and datasets. Integrated change tracking and visualization document modifications to metadata and data. An API offers programmatic access, supporting visualization and map-based exploration. Based on feedback from domain experts, further technical requirements have been identified. These include handling very large data files exceeding several gigabytes, supporting versioning and multi-part datasets with persistent identifiers, enabling workflows for peer-reviewed data, and offering flexible mechanisms to control access when open publication is not feasible. Meeting these requirements is essential to address practical needs, even when they partially diverge from open science ideals. Currently, the CKAN-based infrastructure imposes certain limitations. Files larger than a few gigabytes must be manually integrated. Only one DOI per dataset is assigned, without integrated support for versioning or complex dependencies. Updates do not automatically generate new DOIs or restrict access as required. Moreover, CKAN's rights and role management system does not yet fully support workflows for peer-reviewed data releases or access controls, such as releasing datasets upon request. The ongoing development of the repository is embedded within the Leibniz Data Manager (LDM) infrastructure. This integration enables access to additional functionalities, including Jupyter Notebooks, data comparison, entity linking, and knowledge graph exploration through SPARQL queries and Large Language Models (LLMs). Through this strategic alignment, the sustainable advancement of the FID move repository is secured, ensuring its continued role as a robust, evolving platform for researchers in mobility and transport.