What is research data?

At first glance, the term "research data" seems easy to define. It is usually described as the outcome of the research process. However, at a university with a broad range of disciplines, this definition quickly reaches its limits. Different departments generate, use, and interpret data in fundamentally different ways, making a uniform definition problematic. This challenge is particularly evident in the humanities. Humanities research often relies on existing sources— such as archival documents, manuscripts, images, or audiovisual materials—rather than creating entirely new datasets. Often, these sources are not inherently digital but must be digitized before they can be analyzed, enriched, or reused in research contexts. Are such digitized materials research data, cultural heritage data, or both? Clarifying this question is central to modern strategies for research data management and longterm digital archiving. At the University of Vienna, the management of heterogeneous data from a wide range of research projects and its sustainable availability has become increasingly important. Research data sources range from experimental measurements and simulations to digitized archival holdings and curated digital editions. These collections are extremely diverse in terms of format, structure, size, and disciplinary context. As a provider of institutional infrastructure for long-term digital archiving, the central computing center works closely with the university library, the archive, and researchers from various disciplines. Together, we aim to provide a flexible, sustainable repository infrastructure that accommodates this diversity while supporting long-term archiving, accessibility, and reuse. PHAIDRA, a custom-developed repository system based on Fedora, plays a key role in this endeavor. Designed as a generic yet adaptable platform, PHAIDRA supports a wide variety of content types and offers integrated viewers tailored to different types of digital objects. Its comprehensive, well-documented API enables seamless integration into diverse research workflows and supports automation across various stages of the long-term digital archiving cycle. By positioning PHAIDRA not only as a storage solution but also as an integrated tool within institutional research and archiving processes, we are accounting for the evolving understanding of what constitutes research data. This approach allows us to bridge disciplinary boundaries, support the FAIR principles, and flexibly respond to the needs of both data-producing and dataconsuming communities within the university.