JISC

Start date: 1 November 2002

End date: 31 October 2005

Funding programme: Focus on Access to Institutional Resources (FAIR) programme

Project website: http://www.sherpa.ac.uk/

JISC theme(s): Information environment

Introduction

The SHERPA Project will aim over the course of three years to create a substantial corpus of research papers from several of the leading research institutions in the UK by establishing ‘e-print archives’ which comply with the Open Archives Initiative (OAI) Protocol for Metadata Harvesting (OAI PMH) using eprints.org software. The contents of these e-print repositories (and the harvestable metadata) will be freely available to all the HE and FE community and beyond. The creation, population and management of the repositories will be the core of the Project. Learning outcomes from these activities will also be shared with the community in a variety of ways, including the establishment of mechanisms to assist proactively other institutions in setting up and managing repositories. 

In addition to the creation of e-print repositories themselves, it is proposed to carry out detailed investigations into digital preservation of e-prints. Particular emphasis will be placed on the practical issues associated with the possible implementation of standards such as the Open Archival Information System (OAIS) reference model. 

The OAI PMH was originally developed to facilitate interoperability between e-print repositories. A number of successful e-print repositories, such as arXiv and CogPrints, are now OAI-compliant. However, most of the successful e-print repositories are centralised discipline-based implementations which are limited to a small number of subject communities. There are as yet few major examples of distributed institution-based e-print repositories. Yet the institutional model has the potential to promote the wide multi-disciplinary use of e-prints as a means of scholarly communication, and it therefore needs testing. The SHERPA Project aims to test the viability of the model in practical ways by deploying and promoting a number of repositories in various research institutions and then analysing the technical, cultural and economic issues that emerge. Attention will be given to applicability of the model to different subject disciplines bearing in mind their varying conventions of communication and publishing. 

Preliminary work carried out by some of the Partners in this bid indicates that e-print repository implementations will only be successful in the context of major institutional advocacy initiatives. Academic authors will need to be persuaded that the e-prints initiative is a useful way to improve scholarly communication. Assistance will then be required to help them format and deposit e-prints. They will also need advice in addressing issues such as IPR and copyright. The SHERPA Project will aim to co-ordinate advocacy initiatives, assist authors in depositing papers and provide practical advice in dealing with IPR and copyright issues. In doing this SHERPA will build on work already being carried out by groups such as the CURL Task Force for Scholarly Communication, the JISC Scholarly Communication Group and the SCONUL Advisory Committee on Scholarly Communications. For example, during the Spring of 2002, CURL mounted local advocacy campaigns directed at academics in at least 14 separate venues. These have been evaluated with a view to informing future activity. The learning experiences and advocacy materials from activities like this in SHERPA will be made freely available to the wider community. 

The preservation of the digital content made available by SHERPA is a key concern of the Project. Research will be carried out in this area to examine the feasibility of the application of standards such as the Open Archival Information System (OAIS) Reference Model. In April 2001, Peter Hirtle, Associate Editor of D-Lib wrote, “an OAI system that complied with the OAIS reference model, and which offered assurances of long-term accessibility, reliability, and integrity, would be a real benefit to scholarship.”  The SHERPA Project will investigate how such as system might be created. The work will draw heavily on expertise from the CEDARS Project, and will be influenced by continuing international developments in this area. It will be carried out in partnership with the AHDS which have already developed considerable expertise in this area. It will also be assisted by collaboration with the British Library and also the Research Libraries Group. 

More broadly, access to the papers in the SHERPA repositories will be promoted by working with available OAI Service Providers, some of whom (such as ePrints) are also to be funded as part of the FAIR Programme. Collaboration on metadata standards will for example be essential to facilitate full interoperability. 

The thrust of SHERPA is towards a vision of the academic community taking control of its own authored content in order to improve the research, learning and teaching processes. Working towards this vision will involve testing out new business models of publishing which may run in parallel with current commercial models. The aim will be to enhance the effectiveness of research communication by promoting free access to a wider range of material than is presently available. SHERPA will also aim to conform with the ideal of “a true democracy of learning opportunity” identified by JISC in its recent draft Information Environment: Development Strategy 2001-2005. By making the content available, ensuring that greater interoperability is achieved, and promoting user-friendly search interfaces, the Project hopes to bring the activity of researchers much more easily into the digital domain of undergraduate students, fusing the worlds of learning and research in ways which could benefit both. It will clearly also assist the Development Strategy aim of achieving a sustainable future for scholarly information by taking the digital preservation challenge seriously. 

Where possible SHERPA will work in partnership with publishers. We will actively seek publisher partners who wish to work with us to help to achieve a hybrid (that is, commercial and non-commercial) environment for research preservation and access. For example, we hope some publishers may agree to copyright transfer arrangements which enable copies of articles published in their journals to be deposited in SHERPA open access repositories, and will also create links to these from their own servers. The American Physical Society provides a model for the type of far-sighted publisher with which we hope to work. The APS has recently shaped its policies to accommodate e-print repositories. Alliances will also be sought with organisations such as SPARC. Our aim in SHERPA is to work towards the rebalancing of the ‘for fee / for free’ scholarly information environment, but we expect that academic libraries will continue to work within a hybrid environment of commercial and freely published research for the foreseeable future, with the commercial content retaining an advantage in the quality of its presentation and its ‘value-added’ functionality.

Aims and Objectives

The SHERPA Project aims to develop and promote an environment in which the research output from several of the UK’s largest research-led institutions is freely available on OAI-PMH-compliant e-print institutional repositories to the rest of the HE and FE community and beyond. 

The institutional repository model will be tested by SHERPA mainly in CURL university institutions. As well as including single-institution repositories, SHERPA will also test out the model where several institutions share a single server. The ‘White Rose’ partnership of the universities of Leeds, Sheffield and York will provide this dimension to the Project. In addition, the British Library will establish a server to host papers produced by ‘non-affiliated’ researchers (those using the BL and others working in research institutes or independently). 

The work of SHERPA will not, however, be limited to the initial Development Partner institutions. It is envisaged that the Project will explicitly investigate issues associated with enlargement. During the project seven additional partners will join as Associate Partners: these partners will be within the project structure and will draw on the Development Partners experience in creating their own repositories. Bringing the Associate Partners on board will be used to test out enlargement models. They will not funded to the same level as the original Development Partners. Further enlargement (beyond the Associate Partners) will also be investigated. In addition, SHERPA will actively encourage other institutional implementations by providing advice and assistance to others in setting up repositories. 

The proposed model of the provision of free versions of papers will need to be evaluated. The SHERPA evaluation study will examine the culture change dimensions of the Project, as well as measuring the success of the technical deliverables. It will also look at usage of the services in order to assess the possible long-term viability of the system. Key to the evaluation will be an assessment of the value of the system to researchers in institutions outside the partner sites, with particular attention paid to the value to new universities and those still building up their research profiles. 

Objectives

• To construct six exemplar institutional e-print repositories in Development Partner sites using eprints.org software and to share widely the skills gained.
• To augment these repositories with at least seven other institutional repositories during the course of the Project.
• To investigate key issues in populating and maintaining e-print collections and share widely lessons learned.
• To specify and assign high quality OAI-compliant metadata to the items in e-print collections which facilitate wide public access to their contents, and to share the standards used.
• To investigate the feasibility of the long-term preservation of e-prints, with particular emphasis on the application of the OAIS Reference Model, and to share the lessons learned.
• To effect a culture shift among academic staff regarding the deposit of papers, both pre-refereed papers and those already accepted for publication in academic journals.
• To evaluate the process of creating, populating, and managing e-print repositories, and assess its value to the entire UK research community and beyond.

Key Deliverables and Outputs

• Thirteen institutional OAI-compliant e-print repositories populated by research materials and freely available on the web.
• Published guidelines for setting up e-print repositories covering issues such as metadata, file format and copyright.
• A set of activities to assist other institutions in setting up repositories for the duration of the Project.
• A detailed report on setting up and populating a standards-based long-term digital archive.
• Publicly available advocacy material for communicating with researchers.
• An evaluation report on the technical, cultural and economic implications of the Project.
• Contributions to the ongoing debate on the future of scholarly communication within the information and academic communities.

Overall Approach

The project is structured to investigate the issues surrounding the installation and population of institutional repositories to assist scholarly communication. To do this, it is staged in three levels of build within the Development Partners: first, initial installation to investigate technical issues and those IPR/Copyright, metadata and licensing issues which need to be clear before any systematic population can take place: second, the population with a small number of papers of each repository to test the process of deposition and throw light on difficulties and issues of the process; third, to proceed to the substantial population of each repository. 

In tandem with this, the parallel tasks of effecting cultural change within participating institutions will be addressed, as the process of population and then use of the repository is a change from the current working practices of researchers, publishing academics, information administrators and other stakeholders. To address the needs of this diverse body, advocacy materials and strategies need to be developed and tailored to individual institutions and circumstances. Such work will be informed by a process of iterative and collaborative development between the Project Officers. The project devotes a considerable amount of effort to this work. The Project Officers will collaboratively generate, evaluate and implement advocacy strategies and materials over the course of the project, through successive stages of refinement and updating. 

Project Officers will be faced with specific circumstances within their own institutions through local structures or specific personnel needs which will require localisation or customisation work of centrally available SHERPA material.  Such work can be fed back for further re-use or amendment at other locations, increasing the utility and scope of available materials. 

The next stage is to use the lessons learned and experience gained, to guide the installation and population of a further 7 repositories in Associate Partner institutions, while still working within the project and an overall controlled analytical and evaluated environment. 

This will act as the test for the further expansion out into the wider HE/FE community using project outputs beyond the life of the project itself to facilitate and inform this process. 

The project team is distributed across its partner institutions, with Project Officers in each (the Project Manager also essentially taking the role of a Project Officer for Nottingham). These Officers will be responsible for the work within their own institutions on installation, population and advocacy. In this work they will be assisted by collaborative efforts from the other Project Officers, co-ordinated through project management. Project Officers can compare experiences and share work to generate efficient advocacy strategies and useful advocacy materials. Material generated can be exchanged and held centrally for localised editing and implementation. 

There is a Technical Officer for the project, based at Nottingham, who will be available to offer technical support to the Project Officers and the technical personnel in their own institutions. 

Also working as a central body is the AHDS, with a Preservation Officer undertaking investigative and advisory work for the project. This work will inform each institution in setting up preservation strategies and producing documentation and processes to ensure preservation of the materials in their repositories. 

Due notice will be taken of the work of other projects in similar areas, notably other JISC FAIR projects, and contact with appropriate projects and special interest networks will be maintained. It will be the responsibility of each Project Officer, co-ordinated through project management, to identify and pursue opportunities for appropriate advocacy, liaison and furtherance of the project's aims. 

Overall, the project work will be collaborative and will support each individual institution in its own efforts to found their repositories. Project Officers will be supported by their peers and centralised co-ordination. After their own institution's repositories are established, as part of the project they will pass on their experiences and knowledge in project outputs and personal assistance, for use and testing in the wave of expansion into Associate Partners institutions. Finally, this work will be used to generate material to assist and inform the wider HE/FE community in setting up institutional repositories after the project's end. 

SHERPA does not intend to act as an OAI Service Provider. An installation of eprints.org software provides the ability for searching that single installation at an institutional level. It is assumed that Service Providers will be available to work with SHERPA (some also funded as part of the FAIR programme) in order to ensure that the created repositories are widely searchable across institutions. Preliminary discussions have already been held with the ePrints UK  Project also funded by FAIR. Registering the repositories as OAI compliant will automatically bring them within the search space of current OAI harvesters and service providers, such as Oaister.

Project Consortium

• The University of Nottingham (lead site)
• University of Edinburgh
• University of Glasgow
• Universities of Leeds, Sheffield and York (‘White Rose’ partnership)
• University of Oxford
• The British Library
• AHDS
• CURL