4TU.ResearchData is an international data repository service for the science, engineering and design domains, led by the four technical universities in the Netherlands.

Since 2023, the repository service runs on in-house developed free and open source software called Djehuty. Through Djehuty, we are facilitating contributions from the 4TU.ResearchData community and are collaborating on its use as a multi-organisational repository system.

At 4TU.ResearchData we strongly believe that a community-driven development approach allows for rapid iteration and improvement of our service, as bugs are identified and fixed, features are suggested and implemented, and documentation is refined by the very users who engage with the software daily. For users, the ability to influence the development of the data repository is empowering and leads to a service that more closely aligns with their needs.

Additionally, by using open standards and protocols we are aiming for a service that is flexible and enables smooth integrations with other tools and services.

In this presentation we will address and showcase the value of community contributions and collaborations to meet domain-specific needs and share future plans to support these further.

The University of California, Berkeley historically lagged behind peer institutions in research data storage, offering 200 GB per faculty member while peers provided 5 - 30 TB. Faculty turned to ad hoc, insecure storage practices, and appealed to campus to take action. In response, campus IT, Research IT, and the Library launched a Faculty Storage Allocation Program, offering 5 TB per faculty across five storage options, improving security, compliance, and resilience to data loss and ransomware. The Library’s local Dataverse instance was initially deployed to accommodate licensed collections with fine-grained access controls, and now faculty will use it to publish and manage their larger research datasets (1 TB+). A campus-funded data curator will onboard researchers and teach FAIR principles, promoting discoverability and reuse. The service integrates emerging technologies and standards: machine-actionable metadata, persistent identifiers (DOIs, ORCID), AI-assisted metadata enrichment, automated fixity checks, and ransomware-resistant backups, all of which advance preservation, interoperability, and open knowledge. Combined storage, curation, and community outreach create a sustainable, future-ready ecosystem that protects research, amplifies scholarly impact, supports compliance, and fosters campus-wide stewardship of research data. Planned governance, training, and disciplinary partnerships will ensure equitable access, responsible stewardship, and long-term sustainability into the future.

UX testing mitigates confusion and complaints. This presentation outlines the user testing processes performed during and after a repository platform migration. Methods include: timed exercises, testing external sites, developing user requirements, observation sessions, and environmental scans. Limitations are discussed, along with practical advice about implementing a user testing project.

DSpace-CRIS is becoming an increasingly popular open source alternative to implement an institutional research information management system (RIMS/CRIS): with 54 instances in 21 different countries on the euroCRIS Directory of Research Information Systems (DRIS) at the time of writing, DSpace-CRIS is the 6th most popular CRIS software solution in the directory (as well as by far the most popular open source one). This is because its expanded data model allows the main research entities (such as researchers, organisations, projects, publications, datasets, etc) to be effectively collected and cross-linked.

In January 2026 the Vietsch Foundation awarded a project to euroCRIS, 4Science and the Technical University of Hamburg to expand the DSpace-CRIS data model so that it becomes possible to add instruments and facilities to the snapshot for the institutional research activity. The project plan includes the modelling of this new CERIF research entity by euroCRIS on the basis of the work done by the RDA PIDINST WG, its implementation in the DSpace-CRIS version under development by 4Science and its real-life testing by the Technical University of Hamburg via their DSpace-CRIS-based institutional CRIS TORE.

This talk will summarise the project achievements and the lessons learned at the time the presentation gets delivered.

This proposal addresses archival and digital preservation strategies in contexts where financial, technological, and human resources normally taken for granted are lacking or unstable.

Based on his experience with archival and digital asset management systems for large Western institutions, last year the author began contributing to a non-profit project to help salvaging cultural heritage at high risk of damage in conflict regions. After partnering with volunteers local to the areas of interest and abroad, it became quickly apparent that most of the underpinnings of his initial approach to the problem were based on assumptions that are far from the reality of the new context, which requires an entirely different mindset to face the challenges at hand.

This presentation describes the main challenges of safeguarding critically endangered cultural heritage in areas affected by conflict, censorship, and poverty, and how standard "best practices" can serve both as models and anti-models for an effective solution.

Newly developed tools and related methodologies will be described in this presentation, as well as a status report on the related ongoing projects.

This presentation demonstrates practical implementation of AI-powered search in academic repositories, moving beyond the "put an AI in it" hype to examine real capabilities and limitations. Using public data from Project Gutenberg, I'll showcase how Large Language Models (LLMs) enable semantic search that surpasses traditional keyword matching and manual tagging by understanding thematic relationships between texts.

For over fifteen years, County Health Rankings & Roadmaps (CHR&R) has provided data, evidence, and tools to help communities understand health and advance health equity. CHR&R data are used by researchers, policymakers, journalists, and state and local public health departments across US contexts. In response to funding loss and long-standing concerns about transparency and sustainability, CHR&R has transitioned its measure calculation system from a closed, proprietary SAS environment to an open-source R repository on GitHub. This Repository Showdown demonstrates how the CHR&R Measure Calculations repository supports open knowledge exchange while remaining usable for a diverse applied public health audience, and how code, documentation, versioning, and archival practices advance FAIR principles, preserve methods over time, and support trust and reuse.

Institutional repositories (IRs) are key infrastructures for publishing, referencing, and monitoring of scholarly publications. As networked systems, they play a crucial role in advancing the transition to open access (OA) and are increasingly used as instruments for OA monitoring. Drawing on a systematic review of international literature, expert interviews, and seven networking forums with open access professionals in Germany, this presentation examines the technological, legal, and organizational challenges currently faced by IRs and outlines strategic recommendations for their future development. The challenges addressed include the management of preprints, integration with current research information systems (CRIS), secondary publication workflows, publication and cost monitoring, research data management, and compliance with OA funding requirements.

The next major release of EPrints boasts a host of improvements and new features. EPrints 3v5 beta was demonstrated last year at OR, since then further work to refine it has been completed. This includes reworking of how various key data fields are modelled, specifically those that relate to people such as creators and editors, and organizations such as funders and publishers. A new powerful ‘contributions’ field allows for these entities to be modelled in a more consistent, rich, and more as first class actors in the repository.

Background: National Health Service (NHS) Repositories have a role in capturing grey literature. 60% of conference posters in paramedicine are unlikely to be further published and become Lost Research. Capturing conference posters is an area where a repository can add value. amber, the UK ambulance service repository, has been working with conference partners to do this and now has a critical mass of posters. Capturing conference posters is expensive and we hope to demonstrate that the investment is worth it.

Methodology: An analysis of the item views and download data from amber to measure usage.

Results: We will answer the questions: Are posters being used as evidenced by downloads and views on amber? What are the patterns of usage? What subjects/topics drive high usage?

Conclusion: Repository managers must make choices where to deploy often scarce resources to best effect. We hope to demonstrate and provide evidence (or not) for capturing conference posters.

Institutional repositories increasingly depend on external scholarly data sources to improve coverage, timeliness, and metadata quality. However, tight coupling between harvesting logic and repository platforms often introduces operational risk, complicates upgrades, and limits sustainability.

This presentation describes the design and implementation of an independent crawler microservice developed for the Imec institutional repository. The service harvests publication metadata from Web of Science and Crossref, performs deduplication and controlled metadata merging, and communicates with DSpace exclusively through its REST API. By fully decoupling crawling and enrichment logic from the repository core, the solution enables independent scaling, configuration, and failure isolation, while remaining upgrade-safe across DSpace versions.

We will present the crawler’s architecture, deployment as a Linux service, incremental harvesting strategy, DOI-based deduplication, and a transparent metadata precedence model balancing licensed and open sources. The approach directly supports FAIR principles by improving findability, interoperability, and machine-actionability, while reducing long-term maintenance and preservation risk.

The session concludes with lessons learned, design trade-offs, and recommendations for repository developers seeking resilient, future-proof integrations with emerging scholarly infrastructure.

Generative AI is reshaping scholarly communications at a pace that surpasses existing policy frameworks. Open repositories face a double-edged sword: leveraging AI to enhance discovery, accessibility, and metadata creation or preserving scholarly works that may lack attribution, validity, and originality or have uncertain intellectual property status. Based on a case study at Appalachian State University that involved the implementation of new AI-related policy elements to the university’s institutional repository General Deposit Agreement, this presentation proposes a practical framework for adapting open repository agreements to address the legal and ethical risks introduced by generative AI. This presentation will also discuss model language, implementation strategy, and outreach considerations to the university community when updating agreements to reflect policy changes that address the changing nature of generative AI in scholarly publishing and scholarly communications.

The Federated Research Data Repository (FRDR) is a bilingual national platform for publishing and preserving research data. In FRDR, a Collection is a way to organize related datasets under a shared identity or research context. Collections provide a framework to showcase groups of datasets produced by a lab, research project, institution, or collaboration, helping researchers manage multiple outputs and increase their visibility.

Behind the scenes, FRDR distinguishes between two types of storage groups that underpin how Collections work. Default Storage Groups are available to all registered users and support individual dataset submissions that are fully searchable, but not branded or grouped into a Collection. Special Storage Groups are created by request and act as branded, curated collections for entities such as labs, large-scale research projects, multi-institutional collaborations, or departments that want to group datasets from different laboratories.

This Repository Showdown presentation will demonstrate FRDR’s new collection-specific metrics, which give Collection owners a self-service view of engagement with their datasets (for example, views, downloads, and basic trends), and will outline future directions for Collections, including richer dashboards, stronger collection identity, and closer connections with governance and storage models.

Research organisations registries, such as ROR.org, do not provide authoritative persistent identifiers for sub-organisational units such as faculties and departments, hindering reliable linking to research outputs. The OpenAIRE Graph addresses this gap by leveraging Current Research Information Systems (CRIS), which natively model organisation hierarchies using Universally Unique Identifiers (UUIDs), and deduplicating these with ROR.org’s root organizations. This presentation demonstrates how the OpenAIRE Graph can perform this integration via OpenOrgs, a service for the deduplication and curation of research organisation metadata. OpenOrgs collects as input organization metadata from institutional CRIS systems and known organization registries, such as ROR.org, PIC IDs from the European Commission, and other funder databases. Institutions can entitle data curators to analyze and adjust the hierarchy of potential duplicates automatically identified by OpenOrgs and therefore curate their organisational structures linking root organizations to major known registries and compensating with local sub-units originating from CRIS systems. As a result, the OpenAIRE Graph can bear unambiguous linking between sub-organisational units and research outputs, powering organisation-unit–level filters in research monitoring tools. With this solution, institutions can analyze Open Access uptake, FAIRness indicators, and other metrics at the faculty or department level, supporting internal assessment and strategic planning.

Generative AI models and products have become very popular over the past few years. Many large institutions, including universities, are licensing these products and are strongly encouraging their employees to use them in their work. However, these advocates often skip over problems inherent in generative AI platforms, such as inaccuracies, slop, climate and power concerns, aggressive crawling behavior, and ethical and copyright concerns. To help mitigate these issues in my own work, I have developed key points and questions to consider when evaluating generative AI projects which are proposed for the systems under my purview. In this presentation, I will very briefly describe my thought process about adopting AI in repositories as a background for attendees. I will share my key points and questions with which I evaluate potential AI projects and speculate on future generative AI plans for my managed repositories.

For Small Island States (SIS), digital preservation is constrained by limited financial resources, technological infrastructure, and human capacity. This lightning talk discusses the case of the Open Knowledge Repository maintained by the University of Aruba, Aruba. Taking back control of digital preservation of research outputs from University of Aruba staff is a privilege as it allows for distribution and maintenance of our research outputs, but it is also a costly endeavor. Compared to countries from the Global North that are well-resourced, such as the Netherlands, the University of Aruba has access to limited staffing and financial resources to maintain the Open Knowledge Repository - despite Aruba falling into the category of a high income country. While the content of the repository is maintained by the University of Aruba, the University still relies on external experts and technical maintenance to keep the repository running. This contribution illustrates the tension between the desire for local agency over digital preservation, and the realities of sustaining these efforts under precarious conditions even in high income countries - highlighting the uneven costs of digital preservation and the challenges of limited human capacity.

Institutional repositories are undergoing a paradigm shift, evolving from passive digital archives into dynamic environments central to the Open Science ecosystem. Despite their growing importance, many platforms still lack integrated tools to visualize the multi-faceted impact of scientific production, often relying on limited or proprietary data sources. This paper addresses the technical and strategic challenges of incorporating diverse and verifiable metrics into repository interfaces. We introduce the HERA project (Herramienta para el Enriquecimiento de Recursos Académicos ), a centralized platform designed to automate the retrieval, normalization, and aggregation of indicators from heterogeneous sources such as OpenAlex, Scopus, and DOAJ. Specifically, we present a new DSpace module that operates as a client-side component, utilizing persistent identifiers (PIDs) to asynchronously fetch and render metrics via a standardized, non-intrusive widget. This architecture ensures system robustness by decoupling complex API interactions from the repository's main performance. By providing transparency in data provenance and highlighting discrepancies between open and commercial sources, HERA facilitates a more nuanced research assessment and enriches the user experience within institutional repositories.

University repositories often live at the margins: a compliance box, a dusty shelf, a link nobody clicks. This session asks an important question: what would it take to make the repository feel as ordinary as email every day? Using workflow ethnography across research offices, libraries, IT, and teaching units, we trace the tiny frictions that shove deposit to “later”, then rewire them into default moves: capture from CRIS and LMS feeds, one-screen rights choices, and nudge-like feedback that shows where outputs travel. Sustainability here isn’t a budget line; it’s habit, trust, and shared ownership. We’ll examine governance and incentives that stitch repository actions into daily routines while protecting openness, attribution, and metadata integrity. Expect candid counterpoints: when automation backfires, when “easy” harms quality, and when culture beats code.

The FAIR Guiding Principles (Wilkinson et al., 2016) aim to drive the transformation towards Open Science by ensuring that data is findable, accessible, interoperable, and reusable (FAIR). While typically associated with research data, these principles are also essential for sustainably and inclusively managing publication cost data, which plays a key role in assuring transparency in scholarly publishing. By making cost data accessible and comparable, institutions can enhance financial planning, budgeting, and negotiations.

Therefore the "openCost" project (www.opencost.de) provides a standardized format for the automated exchange of publication cost data. This Lightning Talk explores whether the openCost format can be considered FAIR enough by analyzing the openCost metadata schema in the context of each of the four FAIR principles. Persistent identifiers, the OAI protocol or a community-approved vocabulary are just examples of creating the metadata's compliance with these FAIR principles. The repository serves as an ideal platform for organizing FAIR publication cost data, and future development of openCost will extend to internal data recording within repositories, supporting institutions in their cost data management. This presentation emphasizes the importance of FAIRness also in in the perspective of publication cost data to finally facilitate an efficient and transparent Open Access landscape.

This lightning talk shares a case study from the University at Buffalo Libraries about using generative AI to create alternative text and long descriptions for digital collections. To assess AI-generated responses, librarians developed a scoring rubric with three criteria: factual accuracy and correctness, relevance and task completion, and clarity and communication quality. This approach allowed an objective review of three AI tools which were tested on 45 images. The rubric showed problems with the responses, including hallucinated details, omissions of key visuals from the photographs, and cultural insensitivity. The rubric also showed the importance of incorporating iterative changes to prompts and workflows. Lessons learned include the importance of metrics, human review, and collaboration. This talk offers suggestions for libraries and repositories seeking scalable approaches to accessibility compliance and a rubric for evaluating AI tools.

Academic repositories are uniquely positioned to collect and preserve creative and practice-led research outputs (otherwise referred to as non-traditional research outputs), yet manual processes for capturing these works remain time-intensive and sometimes prohibitive. Unlike research outputs published through commercial and academic publishers, creative and practice-led research outputs are not automatically harvested, necessitating close collaboration between the library, research office, and researchers to curate and make these outputs available.

This presentation summarizes a comprehensive review at an Australian university which re-examined their repository’s purpose, eligibility criteria for submission, workflows, and support mechanisms for creative and practice-led research outputs. Established in 2022, the Griffith Creative Works repository aimed to enhance discoverability, equitable access, staff engagement, and accessibility of creative and practice-led research outputs; however, several years on, these goals were not fully realized. In response, a 2025 change proposal was released to address submission rates, eligibility, workflow improvements, and support resources. The approved changes will be implemented in phases, beginning in 2026 with a focus on operational and cultural practice enhancements. Subsequent stages will emphasize embedding FAIR principles and expanding open access, ultimately aiming to encourage sustained participation between the library, researchers and the research office, and promote open research practices.

This presentation examines how the Universidad Nacional del Sur (UNS, Argentina) has sustained open knowledge exchange through the development of an institutional repository for Open Educational Resources (OER), grounded in a strategic partnership between the Central Library, the Distance Education Directorate, and the Distance Education Advisory Committee. The experience is explicitly aligned with the UNESCO Recommendation on Open Educational Resources, particularly regarding capacity building, supportive policy frameworks, sustainability models, and inter-institutional cooperation.

The presentation highlights how governance structures, quality management practices, and national academic networks—namely the Red Interuniversitaria Argentina de Bibliotecas (RedIAB) and the Red Universitaria de Educación a Distancia de Argentina (RUEDA)—contribute to resilient, sustainable, interoperable, and community-driven OER infrastructures. It also discusses how emerging technologies are being leveraged to enhance discoverability and reuse while mitigating risks related to sustainability, ethics, and digital preservation. The case provides transferable insights for institutions seeking to balance openness with long-term resilience.

This proposal outlines key areas for collaboration where 4Science’s advanced repository solutions, CRIS systems, and data management capabilities can integrate with CORE ’s vast aggregation of open access content to maximise research visibility, interoperability, and impact.

The proposal addresses the ‘invisible problem’ faced by university libraries and open science infrastructures: the discrepancy between the technical excellence of repositories and low adoption by researchers. The central argument is that the success of a repository does not depend solely on technology (servers, metadata, FAIR compliance), but fundamentally on marketing, communication, and cultural change strategies.

Digital repositories are among the most significant mechanisms for supporting open access to scholarly output and represent one of the core applications of the self-archiving concept within the academic environment. Over the past two decades, these repositories have received growing global attention due to their role in enhancing knowledge accessibility, increasing the scholarly impact of researchers, and improving the international visibility of academic institutions in global rankings. In the Arab context, digital repositories—particularly institutional repositories—serve as a key reflection of the scholarly and documentary output of universities and research institutions; however, their representation in global directories remains relatively limited.

This study aims to analyze the current status of Arab digital repositories listed in the OpenDOAR directory in terms of their number, geographical distribution, typology, and relative position compared to the total number of repositories registered worldwide. The study adopts a descriptive–analytical approach, drawing on data from the OpenDOAR directory, which is regarded as one of the most reliable global directories due to its reliance on human review and verification of listed repositories.

The Research Data Service at the University of Edinburgh supports researchers through two complementary repository platforms: DataShare, an open-access research data repository using DSpace, and DataVault, a home-grown, restricted-access, long-term retention solution for large or sensitive datasets. While both are built on open-source software, maintaining two separate systems has prompted questions about sustainability, cost, resilience, and development capacity.

We have completed a project which examined existing infrastructure, risks such as single points of failure, and alternative systems, drawing on institutional documentation, community frameworks (including COAR and FAIR-aligned criteria), and a structured MoSCoW requirements analysis. This led to the question of whether the two platforms could be combined into a single, modern, community-supported research data repository platform.

We will outline the rationale for building prototypes of DataVerse and InvenioRDM, focusing on their ability to meet our requirements and fit within our institutional infrastructure.

This lightning talk presents the results of a project to preserve the original appearance and functionality of the West Chester University Institutional Repository during a site redesign. Beginning in late 2025 and finishing in early 2026, West Chester University Digital Commons underwent its first redesign since it was first launched in 2014. The redesign was intended to meet current accessibility requirements, improve site navigability and browsability, and refresh the site’s overall appearance for 2026.

Throughout the project, repository administrators, wishing to honor the work of previous administrators and document the growth of the repository over time, sought to preserve and archive the site’s history. This goal was prompted, in part, by the presence of several non-traditionally archived websites already held within repository collections. Because West Chester University does not maintain a formal web-archiving program, multiple approaches were explored to ensure ongoing and functional access to the legacy site.

The presentation highlights the methods used to capture and preserve the original site’s appearance and functionality, describes how the archived version was made accessible to patrons, and situates the project within broader repository goals related to institutional memory, preservation, and transparency for the campus community.