JISC Final Report Project Document Cover Sheet

Project Acronym: KulturVersion: FinalContact: whw@soton.ac.ukDate: 30.03.093. Aims and Objectives3.1 The aims and objectives agreed at the start of the project were to:i.Establish a model of shared practice across the sector between a mature repository in theresearch sector and other HE institutions.ii.Use EPrints software to set up two pilot multimedia, multifunctional repositories for theUniversity of the Arts, London, and the University College for the Creative Arts, and makethem available early on in the project to aid advocacy and provide the basis for metadataanalysis.iii.Establish an acceptable use model for rights issues related to artistic and cultural outputswhich can be included in institutional repositories.iv.Investigate a metadata, preservation, curation and access framework as an exemplar formanaging material in the visual and creative arts compatible with evolving internationalstandard and the work of a national datacentre (the Visual Arts Data Service).v.Ensure that the repository is effective in reaching out to audiences both within HigherEducation and to potential cross-domain users and assess author behaviours in order todevelop policies suitable to the management, promotion and populating of the repository.3.2 The aims and objectives changed very little during the project, with two exceptions. Instead of twopilot repositories being set up for UCA and UAL, a single joint demonstrator repository wasestablished which was later split into separate repositories. The joint repository model had theadvantage of enabling the project to focus initially on disciplinary rather than institutionalrequirements. It concentrated a bigger critical mass of material in one place, assisting in theadvocacy process and enabling more effective testing of the repository model.3.3 There was a slight change in the objective concerning audience engagement. As the primary usergroup, the focus has been on getting the repositories promoted within each of the hostinstitutions. Consequently, less has been done to actively promote it to potential cross-domainusers, such as the gallery and museum sector. This will be a key area for future developmentonce the repositories have been launched at the institutions. However, through theirrepresentation on the steering group, cross-domain users have played an important role in thedevelopment of the repository model. Representatives from the V&A, video art archives, andcultural industries all had input on the repository’s user interface and metadata workflow.4. Methodology4.1 Kultur has been a community response project and this methodology has been supported by amodel of shared practice. Southampton has contributed technical expertise and the experience ofdeveloping and embedding its own institutional repository, while UAL, UCA and WSA haverepresented the needs of the arts community, with their combined strengths indicative of a verybroad spectrum of creative and applied arts research activity. Through its experience of curatingdigital art collections for use in research and education, VADS has provided expertise in imagedata and metadata standards, user interfaces and IPR issues for the visual arts sector. All of theproject partners have collaborated in undertaking an extensive user analysis which has informedthe technical, advocacy and policy development of this project.4.2 On the technical side, the project has extended the EPrints digital repository platform to develop aflexible arts repository structure, based on expert input from the disciplines. Southampton hasacted as a development and testing hub, hosting a common demonstrator repository for thepartner art institutions and WSA. This has allowed work locally to focus on advocacy and thepopulating process. The common demonstrator model has been populated with records ofexhibitions, installations, videos, performances and artefacts, and the use of real research outputsPage 5 of 29

Project Acronym: KulturVersion: FinalContact: whw@soton.ac.ukDate: 30.03.09has enabled the project team to continually test and refine the repository model through aniterative process. From an early stage in the project, the demonstrator has been used to elicitcommunity feedback so that various stakeholders – individual researchers, research groups,students, librarians, teaching and learning staff, research managers – have all had some inputinto the repository’s structure and its specialist metadata.4.3 The project addressed interoperability issues by developing a shared demonstrator repository.This was split into separate institutional repositories once the metadata and workflows weredeveloped. The project worked with core standards, such as Dublin Core for metadata and JACSfor its subject classification, but its user-driven approach meant that standards were developed10and adapted in line with user needs (see the Kultur Metadata Report ). Engaging with therequirements of a cross section of the UK HE arts community was essential to the development ofa scalable working model of a repository for use by this discipline. But this ‘ground up’ approachto standards is also due to necessity, and is indicative of the broader difficulties this communityfaces in addressing appropriate standards and workflows, which remain at an early stage. At theoutset of the project, application profiles for non-text work were still very much underdevelopment, and the timescale of the project meant that it had to consider what was currentlyavailable. Although the Images Application Profile was completed part way through the Kultur11project, work on a Moving Images Application Profile is still underway.4.4 At a cultural and policy level, user engagement and shared practice have again been central. Inorder to create working repositories as exemplars for the discipline area, the project hasinvestigated the ways in which institutions, research groups and individual academics within thissector can be encouraged to deposit their work. The user analysis and usability tests identified anumber of barriers and incentives for academics to deposit their material. Solutions for dealingwith many of these issues have been formulated through the collaborative efforts of the specialistarts institutions and a mature repository with a history of successful advocacy, combined withtechnical innovations. Policy to support the management of the institutional repositories wasinitially developed through the project consortium, before being customised to suit therequirements of each institution, in conjunction with individual repository advisory groups.5. Implementation5.1 The core project team comprised a project manager and a technical EPrints officer based atSouthampton, and two project officers, one based at each of the arts institutions. All project staffwere in post by November 2007. The team met fortnightly, and progress was overseen by theProject Management Group, who met every 2-3 months. While the core team were responsiblefor technical development, advocacy, and user analysis, the larger project management group(which included representatives from VADS, the School of Electronics and Computer Science atSouthampton and Library liaison staff from all institutions) played a key role in developing policyand establishing institutional infrastructures. At Southampton, fortnightly e-Prints Steering Groupmeetings provided the opportunity to situate the project within a broader landscape of repositorydevelopment, and to exchange experience with staff involved in other projects.5.2 Early project activity focused on an environmental assessment. Institutional profiles of UAL andUCA were written by the project officers, which scoped the range of research activity carried outat each institution, outlined the research and management structures and considered any existingprovision for the management and/or promotion of digital outputs. The project manager wrote aresearch and projects review to identify useful outputs and lessons from previous repository andarts digitisation projects that Kultur could build upon. Although the original aim had been toincorporate user case studies into the environmental assessment, it was decided that, given theirimportance in leading the design of the repository, case studies should instead form part of aseparate, more substantial user analysis.10Kultur Metadata Report (2009), http://kultur.eprints.org/documents.htmImages Application Profile, Images Application Profile11Page 6 of 29

Project Acronym: KulturVersion: FinalContact: whw@soton.ac.ukDate: 30.03.095.3 While the environmental assessment was in progress, the technical officer set up an EPrints 3.1demonstrator repository. This initially had the standard Eprints 3.1 metadata fields, with theaddition of two new fields to describe the material and dimensions of artefacts. The aim was topopulate the demo with as broad a range of research outputs as possible, so that a clearer pictureof the format and metadata requirements for this sector could be built up, allowing the repositorymodel to be continually tested and enhanced. The project team contacted researchersindividually to request material for the demo. One ‘quick win’ for populating the demo was tocontact individuals who already had some work available online – whether on university, onpersonal or gallery websites, or in the case of WSA, in Southampton’s IR – to request permissionto add it to the repository. The depositing was all done by the project team. Advocacy events werealso used as a means of requesting material for the demo.5.4 This early stage of populating the demo raised a number of issues. One problem concerned thescope of the project and highlighted the need to define an inclusion policy for the repository.Some researchers/research groups saw the repository as a useful service for storing largecollections of material, such as special collections or research data. Although this could well be afuture use of the repository, the aim of the project was to develop an IR model tailored primarily toaccommodate research outputs, although it was acknowledged that the boundaries betweenprocesses, data and outputs in practice-led research are not always straightforward to define.Another issue concerning scope was that researchers did not always have digital records ofexhibitions and artefacts, and some hoped that the project was going to assist in capturingoutputs. This suggested that in order for an arts repository to be successfully embedded within aninstitution, some extra provision would have to be made to support researchers in digitising theirwork.5.5 At this initial stage of the demo development, there was also a tension between the need tosecure content, and the development work itself. The immediate priorities for the technicaldevelopment strand of work had been the functionality and the metadata model of the repository,but it soon became clear that the aesthetics of the user interface needed some work in order toengage arts researchers early on in the project, and to encourage them to contribute material.The aesthetics proved to be a much bigger concern than originally anticipated, somethingcontinually reiterated by feedback from the user community (the steering group, user analysis andindividual feedback from depositors). In response to this, early enhancements to the demofocused on developing a much more visual abstract page for end users, which offered a range ofviewing options, and which prioritised previews of the work over the metadata. The project teaminvestigated the usability of websites and arts databases familiar to the arts community and drewon these in enhancing the demo interface. For example, the list-like layout of repository pageswas replaced with a grid model, and a new lightbox preview feature was incorporated into theabstract page.5.6 During the first few months of the project, advocacy focused largely on internal audiences withinthe partner institutions. In November 2007 an advocacy workshop was organised to formulatestrategies for engaging researchers in this sector. This brought the project team together withother representatives of the arts sector. One of the recurring themes of this event was theimportance of addressing users’ reservations directly and also of tailoring advocacy effectively12through preliminary research into the user group and research culture. This was taken on boardin the decision to extend the user analysis strand of project activity. In the early stages of theproject, it was publicised to various stakeholders through presentations at relevant meetings.Target audiences included Research Co-ordinators, IT and Web Teams, Libraries, ResearchPolicy and Development Committees, and Teaching and Learning teams. A series of printed13materials were also developed introducing the project to different audiences.5.7 In early 2008, a user survey was sent out to academic staff at UAL, UCA and WSA as the firststage of the user analysis. It was designed to gather information on researchers’ working12For account of advocacy workshop, ry%2021%20nov.pdf13Some of these are available in the ‘Advocacy and Dissemination’ section of the project e 7 of 29

Project Acronym: KulturVersion: FinalContact: whw@soton.ac.ukDate: 30.03.09practices, the formats of their work, uses of IT and web services, and attitudes towards openaccess. A social surveys researcher at Southampton was commissioned to assist with the designof the questionnaire, and to advise on the methodology and analysis. Hard copies of the surveywere put in staff pigeonholes and made available at all library sites, while an online version(created using Survey Monkey software) was emailed out on the relevant staff email lists (a copyof the paper survey is included in the appendices). The surveys were anonymous butrespondents were invited to leave contact details if they were willing to be contacted for a followup one-to-one interview, which several did. The project received 199 responses, from which threebroad user groups could be distinguished, as discussed in the Kultur User Survey Analysis14(2008).5.8 Follow-up one-to-one interviews were conducted which enabled the project team to draw outmore detailed case studies of individual researchers, to map the processes involved in creatingand publicising research outputs, and to gain an overview of the research culture. They alsoprovided an opportunity to elicit more detailed thoughts on the perceived uses of a creative artsinstitutional repository, and to gather some valuable feedback on the Kultur demonstratorrepository. The interviews also turned out to be an effective means of advocating the project toindividuals and of demonstrating that the project was keen to listen to and respond to user needs.An unexpected but beneficial outcome of the interviews was that many of the intervieweessubsequently contributed material for the demo, and also went onto publicise the project toothers. The project team interviewed 15 artists and researchers from UCA, UAL and WSA. Theinterviews lasted for an hour and followed a semi-structured format, as outlined in the User15Analysis Interviews Report (2008). The results enabled the project team to build up a detailedpicture of the barriers and incentives for encouraging take-up of the repository, and to develop 5use case scenarios to inform the direction of the repository development.5.9 The user analysis highlighted that IPR was one of the most prominent concerns and potentialbarriers to encouraging deposit in a repository. Because of their experience in the arts digitalenvironment, VADS were commissioned to take on additional responsibility for the IPR workpackage. VADS surveyed existing IP processes at each institution and investigated theapplication of ‘fair use’ for arts in a repository setting. Based on this they made a series ofrecommendations for a common IPR policy, including the establishment of a diligence process. Atthe same time, as part of the repository enhancement activity, technical solutions for protectingthe copyright of a deposited work were investigated. An option for presenting nondownloadable/preview-only images and film was developed, and the automatic generation of lowresolution versions of images was explored. There is a tension here between the open accessdrive of repositories and the development of mechanisms to limit re-use of material. However,based on community feedback, it was felt it was important to at least have least have options forsuch mechanisms in order to encourage take up.5.10Feedback on the demo during the second half of the project led to a final stage ofenhancement work. Sources included a second round of feedback from the Steering Group,comments from the Repositories Support Programme, and from Research Centres. The evolvingmetadata schema was presented to Research Policy Groups/Research Offices for comment (seeKultur Metadata Report), and the demo was also presented and commented on at a number ofexternal events targeting different audiences. These included the Open Repository 08Conference (Apr 08), the JISC Innovation Forum (Jul 08), the ARLIS Conference (Jul 08), theRepositories Fringe Event (Aug 08), Art Libraries Conference on Art Documentation (Aug 08), andththe DCC 4 International Digital Curation Conference (Dec 08). In October 2008, usability testswere conducted with a small number of testers. Candidates were selected from the three partnerinstitutions and were all practice-based arts researchers and/or working in an Arts Libraryenvironment. They were observed carrying out tasks to locate information and depositing a14Kultur Project User Survey Report ort%20final%20Aug%2008.p

Project Acronym: Kultur Version: Final Contact: whw@soton.ac.uk Date: 30.03.09 Page 1 of 29 Document title: JISC Final Report Last updated: March 2008 JISC Final Report Project Document Cover Sheet Project Information Project Acronym Project Title Kultur Start Date March 2007 End Date March 2009 Lead Institution University of Southampton Project