Institutional Repository Platforms

Success and Challenges: An examination of platforms used to support institutional repositories (IR)

Annotations by Natalie Walker

Definition of Project

An institutional repository (IR) is defined as, “a set of services offered by a university or group of universities to members of its community for the management and dissemination of scholarly materials in a digital format created by the institution and its community members"(Reitz, 2014). Since 2000, more than 900 IR’s have been established at academic institutions. To host an IR, institutions must select a platform to support a variety of activities, including file submission, descriptions, metadata, sharing, and storage. There are many options for IR platforms including open source and proprietary software as well as tools created by universities. The increase in popularity of IR’s has shown there is a need for more information to help with the selection of a tool. This annotated bibliography focuses on research conducted to examine top IR platforms used by academic institutions. The literature selected includes case studies, comparisons of platforms, and guidelines to assist with the evaluation and selection of a platform.

Search terms - “Open-source software”, “Open-source software for repositories”, “Academic Repositories”, “Digital Academic Repositories”, “Use of institutional repositories”

Annotations

Allen, A. L. (2017). Lessons Learned in Partnerships and Practice: Adopting Open Source Institutional Repository Software. Journal of Librarianship & Scholarly Communication, 5.

This case study details the platform selection process by the University of Arkansas for their IR. The university evaluated three open-source platforms; CONTENTdm, DSpace, and Islandora. Each platform was used for a digitization project that created content for the IR. Some of the projects included the use of CONTENTdm to process collections for the archive, DSpace to digitize undergraduate theses, Islandora to digitize serials, and both DSpace and Islandora to digitize audio recordings for the music department. Allen reported this approach offered two important benefits. First, it helped to strengthen communication and collaboration among staff at the library, archive, and other university departments. Second, the university was able to assess how well each platform matched with their needs before committing to using the platforms. Based on these experiences, the university selected two platforms. They opted to continue using Islandora for the university archive. However, they selected a proprietary platform, Bepress (Digital Commons) for its repository.

Amorim, R. C., Castro, J. A., da Silva, J. R., & Ribeiro, C. (2015). A comparative study of platforms for research data management: interoperability, metadata capabilities, and integration potential. In New contributions in information systems and technologies (pp. 101-111). Springer, Cham.

This article is a summary of a study that evaluated four open-source IR platforms; DSpace, CKAN, Figshare, and Zenodo. The results of a preliminary study, the authors evaluated each on key attributes: architecture, metadata handling capabilities, interoperability, content dissemination, search features, and community acceptance (Section 3, Table 1). Table 1 provides a helpful overview of the key attributes for each platform, but it is worth noting that the authors did not define the terms assigned to each category (e.g. Installation package vs. service; local vs remote; flexible vs fixed). A second product of the comparison is the key advantages for each platform (Section 3.3, Table 2). If an institution is considering a repository because it wants to increase the visibility of scholarly works from the university and encourage reuse of the dataset, this article provides useful information on what to look for in a platform.

Amorim, R. C., Castro, J. A., Da Silva, J. R., & Ribeiro, C. (2017). A comparison of research data management platforms: architecture, flexible metadata and interoperability. Universal Access in the Information Society, 16(4), 851-862.

Amorim et al examined the features of six IR platforms: DSpace, CKAN, ePrints, Figshare, Zenodo, and EUDAT. Assessment of the platforms focused on two broad categories, architecture and metadata and dissemination. Architecture concentrated on structural-related characteristics such as deployment, storage location, customization, and support. Metadata and dissemination focused on the flexibility in descriptions and interoperability. This included metadata exporting schemas, validation, version histories, and OAI-PMH (p.856). Metadata capabilities were a key focus of this article as the authors believed this was an important feature to consider when selecting a platform. The authors also reported they believed DSpace was a good option for smaller institutions because of its flexibility in schemas, it was OAI-PMH compliant, and storage could be under the institution’s control. Similarly, they recommended ePrints because it offered many of the same options but had flexibility in storage location and supported versioning in the metadata.

Asadi, S., Abdullah, R., Yah, Y., & Nazir, S. (2019). Understanding Institutional Repository in Higher Learning Institutions: A systematic literature review and directions for future research. IEEE Access, 7, 35242-35263.

Asadi et al conducted a systematic literature review (SLR) of research conducted on IR’s. The goal of the SLR was to provide an in-depth review of current research on IR’s. Included studies were published from 2007-2018, clearly address an IR topic, written in English, and address the research questions. There were 115 studies that met these criteria. An examination The top two topics of the studies were deployment, implementation, and adoption (31studies) and benefits and challenges (28 studies) (p.35248). The top two challenges identified were poor IT infrastructure and the absence of IR policies (p.35248). The authors found two benefits of IRs from the studies where they helped to showcase the institution’s intellectual quality and it enhanced the reputation, visibility, and prestige of the university (p35351). An examination of IR platforms found there more than 80 packages used. The most common open-source platform used was DSpace. On the ROAR site, 46% (1834 of 3969) of IRs used DSpace (p. 35244). The authors concluded that successful IRs benefits the entire institution. There is also a need to increase knowledge and awareness of open access IRs as they will continue to grow.

Bankier, J., & Gleason, K. (2014). "Institutional Repository Software Comparison," UNESCO Retrieved from https://unesdoc.unesco.org/ark:/48223/pf0000227115.

With a growing number of tools available to host IRs, Bankier and Gleason saw the need to create a resource to assist with the evaluation of IR platforms. This guide is intended to be used by librarians who are either creating a new repository or considering an upgrade or change to their platform. Bankier and Gleason’s guide includes a comparison of the top five IR platforms. The platforms included in the comparison are Digital Commons and four open-source options, DSpace, ePrints, Fedora, and Islandora. They were identified based on downloads, installations, information available on the platform’s website, and interviews with users. The authors also conducted an evaluation of the key features and reviewed a list of IRs on Registry of Open Access Repositories (ROAR), The Directory of Open Access Repositories (OpenDOAR), and Repository 66 to determine the platforms used. Each platform was evaluated based on twelve categories: infrastructure, front-end design, content organization and control, content discovery, publication tools, reporting, multimedia, social features and notifications, interoperability, authentication, accessibility, and preservation. For each of the categories, Bankier and Gleason provide a brief definition alone with a table rating each platform. What this comparison shows is that there is a great deal of similarity in the features offered by the platforms. For this reason, it is important that institutions have a clear idea of their needs before they begin a search.

Brush, D. A., & Jiras, J. (2019). Developing an institutional repository using Digital Commons. Digital Library Perspectives, 35(1), 31-40. http://dx.doi.org.proxy.lib.wayne.edu/10.1108/DLP-08-2017-0028

This case study highlights some of the challenges academic institutions face when trying to balance the needs of the university with the resources available. For smaller institutions, these challenges can be exacerbated by issues related to staffing, funding, and access to resources necessary to do the work. For Rowan University, an institution with an R3 Carnegie classification for research and approximately 19,000 students, there was a need to come up with a plan to support the growing scholarly activity at the university. Taking the lead on this project, the staff at Campbell Library at Rowan University began a review of their process and developed a plan of action. A key finding of their review was their existing IR platform, DSpace, had not been properly implemented because the library lacked staff with the expertise to complete the task and access to support. For example, the OAI-PMH function that allows for the content to be discoverable to search engines was not enabled. It was also discovered the system was underutilized as its primary use was to archive thesis and dissertations. The Campbell Library staff decided to switch the IR platform to Digital Commons for their repository. Reasons for their decisions included extensive support from Digital Commons in the design and implementation of the platform, the option to customize the system without losing any of its basic functions, and a high level of use by schools and universities that have a similar level of research and staff expertise. Recognizing the expansion of the repository would require a great deal of staff time, they opted to manage the repository by committee. The committee consisted of an IR manager and librarians from a variety of departments, including public services, the medical school, and the archivist. This approach eliminated the need to assign a full-time staff member to the management of the IR and engaged important stakeholders on campus.

Corbett, H., Ghaphery, J., Work, L., & Byrd, S. (2015). 1 Choosing a Repository Platform: Open Source. Making Institutional Repositories Work, 3.

Through the presentation of two case studies, this article examines the reasons why two institutions selected an open-access or proprietary platform to support the operation of their IR. The first case study focused on Virginia Commonwealth University’s transition from open source, DSpace, to Digital Commons. The primary reason for the change was the need for a system that included a journal publishing platform. The second study focused on Northeastern University’s transition from a proprietary platform, Innovative’s Symposia, to Fedora-based Digital Repository Service (DRS), an open-source platform. The primary reasons for the change was a lack of staff and resources to fully develop the Symposia platform and the need to have more customization and maintain control over the content. The authors noted that the literature on this topic indicated that every IR is unique, and the selection of an IR platform should be guided by the unique needs of the institution. Platforms should be evaluated and selected based on how well they fit the institution's mission and goals.

Henry, D. B., & Neville, T. M. (2017). Repositories at master’s institutions: a census and analysis. Library Resources & Technical Services, 61(3), 124.

Henry and Neville conducted a study of IR at primarily master’s institutions. Their study had three goals: to determine the percentage of institutions with IR’s that contain faculty content, type of content included, and to analyze the discoverability of content. Based on their definition of an IR for the study (p.126), they found 190 master’s institutions had and active IR’s. Of those, 137 included content from faculty (p127). They found 10 IR platforms used by the institutions. Summarized in Table 5 (p. 128), the top three platforms were Digital Commons (58.4%), DSpace (26.3%), and CONTENTdm (5%). Throughout the literature, there is a great deal of focus on interoperability and the discoverability of IR content. The authors examined searchers for IR’s from four access points; Google, OpenDOAR, ROAR, and the institution’s main website. They found Google produced the top results and that 99 of 137 IRs were discoverable from at least two access points. A detailed explanation of methods used to assess the discoverability of IRs can be found on pages 126-127.

Royster, P. (2019). Evolution of an Institutional Repository: A Case History from Nebraska. Open Praxis, Open Access: Digital Scholarship in Action, ed. D. Chase & D. Haugh (ALA Editions, 2019). https://digitalcommons.unl.edu/libraryscience/382/

This article provides a historical overview of the University of Nebraska-Lincoln’s (UNL) implementation and evolution of their IR. Established in 2005, UNL’s IR is one of the oldest and largest in the United States. At the time of publication, it contained 99,500 items. Its annual usage averages 6 to 7 million downloads and accounts for 15% of all university web traffic (p. 3). Another important distinction of this IR is its use of the Digital Commons platform since the establishment of the IR. In 2004, ProQuest began to distribute Digital Commons. A faculty member of the university, Dr. Azaddine Azzam, endorsed the software, highlighting the user experience. This paired with the trust the university had in other ProQuest products resulted in them selecting the platform to create a repository. One of the first projects was to add historical dissertations digitized from microfilm (p. 5). The relationship between UNL and Digital Commons strengthen after ProQuest ceased distribution and they began working directly with vendors. This resulted in the university playing an active role in improvements of the platform. Noted contributions highlighted by the author include the addition of a ‘download’ button on the article and improvements to usage reports. Today, usage reports not only deliver real-time metrics to depositors, but it has become a powerful recruitment tool for repositories (p. 7-8). Thus, UNL views their IR as a publishing project and not an archive or IT project. Its mission is to maximize access to content created by its community. Despite the change in distributors for Digital Commons, UNL continues to use the platform as its features continue to meet their needs.

Tzoc, E. (2016). Institutional repository software platforms at undergraduate libraries in the United States. College & Undergraduate Libraries, 23(2), 184-192.

Tzoc examined the implementation of IRs at primarily undergraduate institutions. A focus that is not typical for this topic, Tzoc noted that much of the literature on this topic focuses on research-intensive graduate institutions. In a study conducted in 2007, only five of the 123 high research baccalaureate institutions reported they had implemented an IR (p185). In Tzoc’s study, he identified 573 very high research undergraduate (VHU) institutions. Just eight years later, Tzoc’s study found 67 VHU had an active IR. Similar to graduate institutions, there was a high use of open-source IR platforms. The top three platforms reported were Digital Commons (38), DSpace (21), and CONTENTdm (4). Inquiry into how satisfied institutions were with their IR platform revealed that 13 (31%) were very satisfied and 22 (52%) were satisfied (p. 189). Institutions also offered a number of suggested improves that included better support for hosting multimedia activities on the sight, the ability to customize platforms with CSS/HTML and expanded API capabilities.

References

Reitz, J. M. (2014). Online dictionary for library and information science. Danbury: Western Connecticut State University. Retrieved from https://products.abc-clio.com/ODLIS/odlis_about.aspx