Oxford University creates and modify many hundreds of digital objects every day through its education, research and administration activities. The variety of materials in digital form is very broad, ranging from email messages, office documents, reading lists, journal articles, images, web pages, through to huge and complex datasets. Most of these digital objects reside on the hard disks of personal computers, networked filestores or buried within software systems. There is, however, an increasing need to ensure digital assets for which the University is (or should be) responsible are properly stored, curated, disseminated and preserved. The role of digital repositories is to provide a set of services which address these requirements.
The development of digital repository services in Oxford is not confined to any one part of the University. Whilst the Library Services is a primary provider of digital repository infrastructure (and the Oxford Research Archive, in particular), other relevant activities are located within academic divisions, research groups, and other academic services.
Many universities are developing centralised institutional repositories. For the most part these focus on providing a digital repository of research outputs (e.g. pre-prints, peer-reviewed articles, monographs, theses) and are centralised within the institution to the extent that a single unit (often the library) provides both the technical infrastructure and the data curation services. Increasingly, however, the institutional repository is expected to support several purposes (e.g. a preservation archive of digital records, Open Access scholarly publication, store of learning materials accessible from within the VLE).
Oxford has a similarly broad view of the institutional repository. However, like other universities with a devolved ICT structure, Oxford supports a number of digital repository activities rather than one centralised service. Therefore, it is more appropriate to speak of a federated institutional repository, the services and data of which might be both organisationally and technically distributed within the institution. Any given repository 'service' combines technical infrastructure and domain expertise, either or both of which might be located within, for example, research groups or other divisional units. Many of the existing repository-related activities have operated from the ground-up and, without a more coherent and co-ordinated approach, there is the risk that some existing activities will become unsustainable silos.
In 2004/2005 Oxford established a Digital Archiving Group (DAG) to explore the broader issues of relating to institutional archiving policies and practice. A stakeholder workshop was held under the aegis of this group in June 2005, an outcome from which was the report, “Towards a Research Repository for Oxford University” (2005). It is clear that a digital repository, driven by the research needs of the University, has to interoperate with data creation, analysis and dissemination workflows at the point at which the research occurs as well as the outputs of scholarly publication. For research, many of these issues are being addressed by the Oxford Research Archive and its management structure (the Steering Group for which includes the heads of the divisions). This principle, that repository services should be embedded within the working practices of the domains they are intended to serve, applies equally to repository services for other areas of the University's working life (e.g. educational, library or administration).
In addition, there may be clear relationships between different types of digital object within different domains (e.g. within the knowledge lifecycle including resource discovery, data analysis, scholarly communication, teaching and learning, administrative reporting, preservation and access). It is also increasingly recognised that it is desirable and possible to define common services required by repositories in different domains. Examples include: access management, file storage systems, ingest or deposit, cataloguing or metadata standards, identity and version control, IPR and preservation policies.
The Oxford Digital Repositories Steering Group has been established to both coordinate digital repository activities across domains and to help ensure that stakeholder requirements and priorities are represented.
It is anticipated that the technical infrastructure for the federated digital repository will be largely built on the Fedora repository architecture, managed by OULS and underpinning the Oxford Research Archive (ORA). The ORA will form a key component within the federated repository, and there will be many more locations where digital output will be stored. In many cases digital output will be curated and managed locally by academics.