Since 2008, the University of Vienna has been using a repository that is open to all disciplines and to all academic and administrative staff. This opens up many possibilities in the area of research data management, as data can be blocked or opened for certain groups of people at any time, if this is necessary for legal or ethical reasons. The content of the repository can also be linked in different ways, for example, publications and research data can be linked together. A number of metadata standards are provided to allow a very rich description of the objects. At the same time, there are a number of challenges to overcome in order to serve all interested parties and disciplines equally. In addition to technical, structural and financial challenges, the service has to be constantly adapted to different needs. The presentation will take a closer look at these challenges, focusing on solution strategies and positive results in terms of successful collaboration with researchers.

“Why do we have three repositories?” This is an over-simplified version of a question that a working group at Duke University Libraries (DUL) found themselves asking (and in some cases returning to) last year. A supremely helpful member of this working group produced an impeccable 24-page timeline detailing the history of repositories at Duke, beginning with the very first collection launch in 1995. Pulling the greatest hits from our colleague’s impressive timeline, this presentation will take attendees on a journey through repositories at DUL, with a bit of bragging about wins, invitation to commiserate about shared challenges, and sincere reflection with an eye towards the future.

The aim of this communication is to provide an overview of the twenty-year history of UFPR’s repository, offering a focused perspective on its creation and maintenance in Latin America. The Universidade Federal do Paraná (UFPR) launched its Institutional Digital Repository (RDI) to provide access to academic outputs, such as theses, dissertations, and other scholarly production, in 2004. Over two decades, the repository evolved to include Open Educational Resources (OER), institutional research projects, and Brazil's first public scientific data repository. UFPR has progressed, yet its siloed infrastructure disrupts integration and accessibility across repositories. This fragmentation prevents the effective management, sharing, and discovery of academic and research resources, challenging Open Science ideals. UFPR aims to build a consolidated institutional repository using an integrated digital platform to solve this. Its centralized portal will simplify submission and access by applying FAIR principles. Key technological innovations include developing machine learning tools to automate metadata extraction, enable dataset interconnectivity, and improve resource discovery. The unified platform will promote proactive knowledge management, capacity building through training, community engagement, and open data policy development. UFPR intends to lead Open Science in Brazil and Latin America by fostering international partnerships, citizen science, and collaborative research.

As US institutions look to how institutional data repositories can help meet federal funder data access requirements, we are also considering the purpose of repositories beyond compliance. If we are doing more than ticking boxes, does a generalist repository really meet our needs, or most importantly, the needs of those we want to find, and use our data?

With the receipt of NEH funding for the UChicago Node project, the Library is hosting three repositories at UChicago—Node and Unbound as specialist repositories and data infrastructure and the institutional repository supporting public access to research. Many institutions find themselves in a similar situations. This paper explains why this approach currently makes sense, while considering the future and what might trigger a rethink of approach.

UChicago Node builds on the long-standing OCHRE data service, developed to support the humanities division. With Node expanding and scaling this to the management of Library collections as data, and for initiatives across campus, we're exploring three questions, the current answers for which we believe are of interest to any institution that finds itself managing multiple repositories:

- Can a single repository meet all of an institution’s needs?

- If not, why not?

- If not now, then when?

Institutional repositories are digital archives designed to collect, preserve, and disseminate the intellectual output of an institution, particularly in academic settings. Therefore, ensuring the longevity of the IRs is critical for preserving and disseminating academic scholarship. This study explores innovative practices that can enhance the sustainability and accessibility of University of Johannesburg's Institutional Repository (UJ IR), which serves as a vital resource for archiving a diverse range of intellectual outputs, including journal articles, theses, books and book chapters. By implementing advanced digital preservation technologies and adhering to best practices in archival standards, the repository can effectively safeguard its collections against technological obsolescence and unauthorized access. Furthermore, fostering collaboration among academic staff, library personnel, and researchers will promote a culture of open access and encourage the submission of high-quality research outputs. This study underscores the importance of continuous investment in infrastructure and training to ensure that UJ IR remains a reliable platform for future generations of scholars, thereby contributing to the global body of knowledge and enhancing the university's academic reputation.