Repositories will be established as a collection and delivery channel for content. In five years time we will have a respectable collection of content at each institution. Open access will be available to the full text of a variety of research outputs including peer reviewed research papers, work in progress, grey literature, and theses. Whilst some of the content of digital library collections will be closed due to licence agreements, much will be open access. Images and multimedia objects will be included. Content of institutional repositories and institutional digital libraries will be integrated using joint software platforms and a single interface for deposit.
Deposit of research outputs in institutional open access repositories will be mandated by the institution, some funders will also establish subject based repositories. Institutions will showcase (disseminate and share) their research outputs.
Depositing research for open access will be embedded in the research workflow and will be seen by researchers as a positive step towards enabling the wider use of research and achieving high impact. Deposit will be integrated into the research process with minimal effort from the researcher. Deposit of research outputs (including data) will be integrated with the top scholarly research output tools in each subject domain (ranging from word processing to experimental tools).
Institutional repositories will interact with each other to form a network of UK repositories following similar policies and offering interoperable interfaces. UK institutional repositories will further interact with funders repositories and subject repositories by means of a two way flow of content (digital objects) and metadata. There will be an integrated workflow for deposit of journal articles that automatically checks copyright permissions and manages embargoes.
Repositories will support Web based collaboration between institutional and cross-institutional research teams. Collaboration between research groups will be supported by ‘transitory’ Web based repositories intended to last for the lifetime of the collaboration. These repositories will be used to manage research outputs and data. They will reference content elsewhere, as well as forming the basis for re-use of data and work in progress for the research group. Some of the content of such repositories may well be closed for access by the group until the time comes for wider dissemination. When the group is ready, content will be transferred to institutional or subject repositories for longer term curation as appropriate.
Repositories will facilitate a variety of processes within an institution (publishing, preservation, information management, assessment, teaching, portfolios, marketing). In five years time we will have more effective processes (e.g. REF) because of the involvement of repositories. Institutions will no longer duplicate organisational systems to manage research process and outputs. Institutions will establish institutional strategies for open access, re-use, and preservation of digital resources as part of their management of the lifecycle of digital content. Preservation services will be a hybrid of in-house and third party services (national or ‘in the cloud’).
Access to content will be integrated with discovery through global and domain search tools. Institutions will ensure search engine optimisation by disclosure of their resources to search engines whether this is done locally or by a third party added value service. Linked data (links between data, articles, learning materials) will be embedded in portals, search engines and social networking tools.
The research publishing industry will be influenced by the open access agenda. A new business relationship will evolve between the Research Communication Support Industry (i.e. publishers) and the Research Industry upon which they depend.
An emerging e-infrastructure will see publishers and other third parties (commercial and not-for-profits) finding new roles to add value to the process of discovery, access and re-use of research outputs. For example: metadata creation and enhancement, providing XML mark-up versions of texts, text mining, SEO, linking data and journal articles.
Repository access will be integrated with HEI Authorisation and Security provision. This will allow more fine grained access restrictions to content. Whilst this would facilitate restricted group access to work in progress, this is seen as particularly important with regards to access to data.
Automated statistical information will be available on usage and citation of repository content, metrics will be available on the impact of specific items
There is genuine growth in the use of institutional repositories, although deposit of full texts is trailing behind this growth curve. There is much concern that institutional repository content is slow to grow. A recent study1 by the White Rose Increase project has shown that researchers express willingness to comply with open access mandates, however despite the RCUK and Wellcome mandates, in the absence of awareness raising or enforcement, there is an unwillingness amongst researchers to spend the additional effort required to deposit in open access repositories. In addition there is a lack of awareness in the research community about the benefits of open access and the repository deposit process. Only a few institutional mandates have been introduced (e.g. at Southampton and Glasgow Universities). There is a lack of strong institutional leadership. The latter is starting to appear, with 21 institutional mandates (worldwide), 24 funder mandates (worldwide) and especially Harvards statements.
Within the UK, the RAE/REF have driven social and technological engagement with repositories. Other agendas (e.g. open access and preservation) are trying to take advantage of this government mandated activity. Typically management of digital resources and open access are not embedded in institutional strategy. Overall there is a lack of policy in this area. A change of culture is needed to realise the wider benefits, this depends on development of services layered on repository content and an increase in content types being managed. The idea of a common managed source of content for different organisational systems does crop up but is not widely implemented. Organisational system management is often too geared around each system, building barriers between them and preventing common solutions from being sought/found.
Uncertainty about copyright issues is widespread amongst the research community because of lack of clarity regarding publisher and institutional policies. Conflicting statements of policy from publishers and even delayed and embargoed rights to use are blocking the effective deployment of repositories. As interest grows in curating and preserving data, publishing companies are seeking to increase their control over the primary data and secondary tools that the research industry depends on. The UK has a very informed network of repository managers who have effective mechanisms of influence within their institutions, but they are largely being frustrated by the lack of leadership from the research industry over the demands of its own service industry i.e. the publishing industry.
More weight needs to be given to scholars’ requirements. Some studies are underway as to current research practice in different disciplines e.g. JISC’s SCARP project,2 RIN disciplinary case studies.3 There has been too much focus on the repository as a system rather than on the ‘resource’ (content).
Given the level of activity around repositories it is not surprising that different models of repository interaction are emerging or are being advocated. Little has been said about how institutional, funder and subject repositories can or should interact. A concentrated Web based solution, advocated by some respondents, would be another model. In addition the role of aggregators is unclear, e.g. Intute in the UK, DRIVER at the European level, and OAIster at the international level. There is tension between these different repository content gathering solutions, although they are not necessarily incompatible.
Currently the model for deposit differs between institutional deposit in contrast to deposit in funder repositories (UK PubMed Central or the ESRC Awards and Outputs Database). Wellcome mandates deposit of the publisher’s version of journal articles in UK PubMed Central at the time of publication, with a fee to the publisher paid by the institution as part of the indirect costs of the research grant. ESRC requires funded researchers to deposit in their own ESRC repository. The SWORD deposit protocol might assist in enabling a unified deposit to multiple repositories. The White Rose Increase project is investigating the use of SWORD4 to deposit in a local IR and the ESRC Awards and Outputs Database.
Various areas require more work. There is uncertainty about the best way to disclose the content of repositories to search engines (Search Engine Optimisation). There is limited work progressing the potential for repositories to contribute to linking data, whether linking different versions of journal articles, articles with associated data, articles and presentations.
Digital preservation is not an organisational priority. Current repository software platforms are not particularly geared up to preservation, rather the focus has been on access. It is unrealistic to expect small institutional repositories to undertake costly preservation audit exercises. Further exploration and practical use of solutions is required.
Currently many institutional repositories are at project status and are managed as pilots. Institutions will need to address moving repositories from pilot to sustainable service status. As part of this process institutions need to decide who owns the repository: the library or the research support division? Institutions will need to decide whether to establish departmental repositories or a single institutional repository.
Institutional repositories providing open access to journal articles are the most mature of repository types in the UK, and have been the focus of much of the funded activity over the last few years. These repositories co-exist with activity in new areas of interest (e.g. re-use and preservation of data, REF, Web 2.0, linked data, national repositories and publishers’ initiatives). In effect there is a ‘mixed-economy’ where the role of different types of repository is still evolving and there can be tension between different approaches. In this situation lack of clarity needs to be addressed wherever possible whilst accepting that future change is not always predictable.
Milestone 1: Repositories need to clarify their roles in relation to other repositories, whilst acknowledging they exist within a ‘mixed economy’.
Regarding research outputs (in particular academic papers), the proposed way forward for institutional repositories suggested by the consultation can be summarised as: population, integration, adding value.
Population of IRs is a priority. This will involve ongoing advocacy. Case studies are considered to be a valuable approach. There needs to be greater evidence gathered of the benefits of disseminating and sharing open access digital content using the IR as a well-managed source of content supporting preservation and life cycle management for digital content. Case studies need to demonstrate how open access content available for text mining and re-use enables new discoveries to be made. The emphasis needs to be on how to manage digital content rather than on the repositories themselves. Demonstrate benefits repositories can contribute to other processes, demonstrate novel ways in which content can be used.
Clear guidelines should be drawn up covering the full range of methods for populating repositories including deposit by intermediaries, harvesting from other repositories, searching and downloading from other existing online services such as the Web of Science. For one account of how to pre-populate a repository see Mark Leggott’s description of the RIB project,5 University of Prince Edward Island, that aims to present faculty with a 90% populated IR.
Encourage further mandates for deposit, particularly institutional mandates and encourage enforcement of these mandates. JISC project funding for repositories might be made conditional on institutions establishing such mandates. Engage with EPSRC in their discussions on mandates at the forthcoming EPSRC Council meetings.
These various approaches to populating IRs need to be underpinned by inclusion of open access in institutional strategies. Institutions also need to consider strategic issues associated with good management of the life-cycle of digital content (preservation policies will flow out of this).
Milestone 2: Support population of institutional repositories by advocacy, case studies, guidelines on best practice, encouraging institutional mandates, encouraging inclusion of open access and management of the life-cycle of digital content in institutional strategies.
Develop demonstrations of embedding repositories in the scholarly workflow e.g. making deposit easy from content creation tools; enabling more automated means of grant reporting and building personal portfolios; assisting funder organizations with metrics and administration.
The emerging model of distributed preservation services needs to be explored further and uncertainties resolved about interaction between different types of repository.
Milestone 3: Explore and address integration between institutional repositories and: other institutional systems, other types of repository, funders’ systems.
Provide added value services layered on repositories: tools for deposit, search, re-use, metadata enhancement, linked data, citation metrics, publication lists. These services might be provided at the ‘network level’ by the university sector or not-for-profits, by publishers or others in the commercial sector; alternatively software solutions might be provided as integrated add-ons to existing repository software platforms.
Customer relationship management (CRM) is a vital and local institutional activity that may be supported by centrally produced services. CRM involves outreach and the understanding of particular research and learning needs at a local level, there may be specialist clusters that can be called on. Local requirements may be met by centralised services, or by enhancing existing repository software platforms.
The role of publishers in supporting the research process needs to be addressed. Publishers add value to the research process incurring costs. Funders must be explicit on how their grant holders can pay these costs. Open access advocates need to develop an agenda for the UKs research community’s activities allowing them to define the scope of the discussion with publishers, something like a researchers Brussels Declaration remembering that Open Access is just the foundation for a new scholarship.
Milestone 4: Develop added value services layered on repository content (e.g. tools for deposit, search, re-use, linking data, metadata enhancement, citation metrics, publication lists).
There is a need for an agency/lead body within the JISC HE community to address and take a lead on legal issues connected to copyright and publisher polices. Copyright and contractual issues with publishers need to be standardised. The current disparate approach makes ingesting of papers and data too complex and slow
Milestone 5: Establish an agency/lead body within the JISC HE community to take a lead on legal issues connected to copyright and publisher policies.
- Increase. Increasing repository content through automation and services: questionnaire key findings. http://eprints.whiterose.ac.uk/increase/Increase_Questionnaire_Findings.pdf []
- http://www.dcc.ac.uk/scarp/ []
- Research Information Network. Disciplinary case studies. http://www.rin.ac.uk/case-studies []
- Increase. ESRC scenarios. http://eprints.whiterose.ac.uk/increase/esrc.html []
- Mark Leggott, Repository in a Box. http://loomware.typepad.com/loomware/2008/09/ribrepositor.html []