Making Institutional Repositories Work

Burton B. Callicott, David Scherer, Andrew Wesolek, Burton B. Callicott, David Scherer, Andrew Wesolek

• Editorial: Purdue University Press
• Año: 2015
• ISBN:
  9781612494234
• Categoría: Filología
• Categoría: Biblioteconomía y ciencia de la información

Quickly following what many expected to be a wholesale revolution in library practices, institutional repositories encountered unforeseen problems and a surprising lack of impact. Clunky or cumbersome interfaces, lack of perceived value and use by scholars, fear of copyright infringement, and the like tended to dampen excitement and adoption.This collection of essays, arranged in five thematic sections, is intended to take the pulse of institutional repositories-to see how they have matured and what can be expected from them, as well as introduce what may be the future role of the institutional repository. Making Institutional Repositories Work takes novices as well as seasoned practitioners through the practical and conceptual steps necessary to develop a functioning institutional repository, customized to the needs and culture of the home institution. The first section covers all aspects of system platforms, including hosted and open-source options, big data capabilities and integration, and issues related to discoverability. The second section addresses policy issues, from the basics to open-source and deposit mandates. The third section focuses on recruiting and even creating content. Authors in this section will address the ways that different disciplines tend to have different motivations for deposit, as well as the various ways that institutional repositories can serve as publishing platforms. The fourth section covers assessment and success measures for all involved-librarians, deans, and administrators. The theory and practice of traditional metrics, alt metrics, and peer review receive chapter-length treatment. The fifth section provides case studies that include a boots-on-the-ground perspective of issues raised in the first four sections. By noting trends and potentialities, this final section, authored by Executive Director of SPARC Heather Joseph, makes future predictions and helps managers position institutional repositories to be responsive change and even shape the evolution of scholarly communication.

Part 1

CHOOSING A PLATFORM

Selecting a platform for an institutional repository requires a host of decisions that should be considered within the context of one’s unique campus environment. In the following chapters we see the distillation of these considerations into three major themes: staffing, purpose, and goals. First, Hillary Corbett, Jimmy Ghaphery, Lauren Work, and Sam Byrd weigh the benefits and disadvantages of several popular repository platforms, as well as discuss the process of migration from one to another. Katherine McNeill then explores the data repository ecosystem while offering insights into the role of the repository in this diverse landscape. Finally, Kenning Arlitsch and colleagues delve into strategies to effectively enhance the findability of the content hosted on an institutional repository through search engine optimization.

Corbett and colleagues outline the wide array of repository platforms and the major considerations that go into choosing one over another. One of the key themes in this evaluation is the type of investment an institution is willing or able to make in a repository initiative. Several open source platforms are evaluated that allow for flexibility in their implementation, at the cost of staff time to develop, maintain, and update them. Conversely, hosted repository platforms may be launched with as little as .25 FTE, but come with more rigid structures and limits on customizability.

Prior to selecting an institutional repository, one must consider the type of repository it is intended to be. Major funding agencies are requiring researchers to manage, and often openly share their data at a rapidly increasing rate. Is this a service an institutional repository initiative is intended to support? If so, should support come in the form of institutional repository infrastructure, or through ancillary services such as guiding researchers in the selection and deposit of materials in existing and external data-specific repositories? Katherine McNeill explores this nuanced and fluid ecosystem of repositories. Ultimately, she suggests a multifaceted approach of leveraging an institutional repository for research publications, while integrating them with datasets stored in a variety of external data-specific repositories.

Finally, what are the goals of a proposed institutional repository initiative? Is it enough for an IR to serve as an archive of the scholarly output of the institution, or does the repository initiative intend to disseminate that scholarly output broadly and openly? Kenning Arlitsch and colleagues argue in support of the latter: “discoverability of content through Internet search engines is paramount to the success and impact of institutional repositories.” The authors then outline a variety of search engine optimization techniques that librarians may apply to their institutional repositories in order to increase discoverability.

Ultimately, platform decisions for institutional repositories are highly dependent on individual institutional contexts. However, Part 1 offers a framework for selection built on the themes of the staffing, purpose, and goals of a repository initiative. This framework may then be built upon further after a careful survey of the needs, values, and culture of the institution, thus providing the first step toward a successful repository initiative.

1	Choosing a Repository Platform: Open Source vs. Hosted Solutions

Hillary Corbett, Jimmy Ghaphery, Lauren Work, and Sam Byrd

Platform selection is a concept that will be familiar to many who work in libraries, regardless of whether they have worked with an institutional repository. Selection and implementation of a new integrated library system (ILS) or discovery platform are experiences that most library staff will generally encounter more than once in their careers, and they are processes that typically represent a significant, long-term time commitment for staff across the organization. The stakes are high because so many library employees’ day-to-day work involves active and extensive use of the system that is chosen. Because of this common experience, it naturally follows that library staff tasked with choosing an institutional repository platform may approach the job with trepidation. But in reality, the selection process doesn’t have to be time-consuming or fraught with anxiety. (Indeed, a common pitfall may be to overplan for the process.)

While it’s essential to include representatives of different areas of expertise, the group tasked with selection can be fairly compact. This will help the process move more smoothly. Who should be included in this group? If there is an existing repository, its manager should be involved, of course. Staff from metadata and systems units should also be included. Even with a hosted platform, where no on-site technical expertise would be needed, the systems representative will likely be best able to evaluate its architecture and interoperability. Someone with an archival background can also provide valuable perspective on the preservation aspects of the repository platforms under consideration. Your Web developer or user experience expert can be very helpful in evaluating interfaces and their potential customizations. Above all, the repository must be usable. It can have great metadata support and elegant architecture, but if the interface is clunky, no one will use it. A team member who knows how users interact with the library’s other online resources is essential. Finally, you may also wish to seek input from a power user of your current repository, or someone who is likely to be an active user of a repository under development. If including them during the selection process isn’t feasible, such users should certainly be asked to help later with usability testing.

Your library may already have an existing repository, but try to evaluate prospective new platforms independently of whether or not they are “better” or “worse” than your current platform. In many ways, a new platform will likely just be different—and that’s going to be a combination of positive and negative. Of course, it’s important to consider your current platform in the context of how you will migrate its contents! But you’ve already made the decision to move to a new platform—strive to evaluate your choices on their own merits. The goal in your selection process is to compare new platform with new platform, not new platform with current platform (or with the absence of a platform, if you don’t currently have a repository). If your library already hosts a repository and you’re looking for a new platform, you should certainly make a list of your current platform’s pros and cons—but don’t let them influence your process too much or get bogged down with too much discussion of the current platform. Likewise, keep in mind that platforms are constantly under development, and specific features you note as absent or less well developed may be slated for future releases. Most importantly, remember this evaluation is not a mere side-by-side comparison, but needs to be tied to your institution’s repository goals and ambitions.

While this chapter discusses selection of a locally hosted, open source system (DSpace/Fedora) vs. a cloud-hosted, proprietary system (Digital Commons), it is important to note that these examples are merely illustrative. Libraries have a range of choices for repository software that includes open source and proprietary in any number of support environments, and exemplary repositories are flourishing on a variety of systems, both open source and proprietary. This chapter focuses on the differences between proprietary and open source solutions, but also demonstrates how and why libraries choose a repository system. In writing about this process, we realized that it was important to acknowledge that there are two different audiences for this chapter: those who may just be starting out with building a repository at their institution, and those with an established repository who are considering a platform change. Thus, this chapter addresses the challenges and opportunities of platform selection in both circumstances.

SELECTIVE LITERATURE REVIEW

The library literature regarding open source software has dealt with a variety of systems, including integrated library systems (ILS) and repository platforms. Pruett and Choi’s (2013) article comparing select open source and proprietary ILS software includes a thorough review of previous research, including welcome background from fields other than library science. Palmer and Choi’s (2014) descriptive literature study is also an important touchstone for an understanding of previous research on library open source software. In this review, the authors found that almost 35% of the library literature regarding open source has dealt with digital repository software, and they posit that this concentration is largely due to a preponderance of open source repository platforms (DSpace, Fedora, EPrints). Indeed, the repository market is almost an opposing image of the open source ILS market since open source solutions have defined repository solutions from the outset.

Library literature concerning the choice between open source or proprietary repository platforms reflects the multifaceted and unique circumstances that individual institutions face. Burns, Lana, and Budd (2013) reflect this reality in the conclusion of their survey of institutional repositories, stating that “the most important lesson learned from this survey is that not all institutional repositories are alike” (Discussion, section 5, para. 1). Though widely applicable evaluation methodologies and parameters for choosing an institutional repository are well documented (Fay, 2010; Giesecke, 2011; Rieger, 2007), final decisions for open source vs. proprietary platforms are most often unique to the circumstances of each institution and emerge from university-level needs assessments. Common factors cited in the case studies for choosing proprietary solutions include costs of technical infrastructure and staffing, the need for swift implementation to allow for a focus on repository population and promotion, interface branding and customization, electronic publishing options, and online discoverability of scholarly research (Bluh, 2009; Mandl & Organ, 2007; Younglove, 2013). Libraries that select open source repository platforms also note customization as a positive factor, but include extensibility, flexibility to ingest varied formats, and interoperability (Fay, 2010; Marill & Luczak, 2009). In line with these cost-benefit issues of open source, Samuels and Griffy’s (2012) case study in evaluating open source publishing solutions includes a comparative methodology that includes total cost of ownership.

Salo’s tongue-in-cheek essay “How to Scuttle a Scholarly Communication Initiative” (2013) is required reading, both for its insightful look into library culture and its very well-developed bibliography for anyone interested in starting or improving a scholarly communication program. In discussing platform choice, Salo encourages usability and beta testing as well as reaching out to colleagues who are current or former users of the systems under consideration. Salo makes her point about the pitfalls of focusing solely on platform without consideration of the larger scholarly communication goals of the organization in a particularly humorous manner: “It is particularly important to fixate on a software package before the initiative’s mission, milestones, and workflows have been decided . . . to maximize the discrepancies between necessary work and the software’s capabilities” (p. 3).

VIRGINIA COMMONWEALTH UNIVERSITY: FROM OPEN SOURCE TO PROPRIETARY

Virginia Commonwealth University (VCU) launched a DSpace instance in 2007 as a platform to support its electronic theses and dissertations (ETD) program. All systems and database administration, server maintenance, and application support were handled by library technical staff. There were no additional staff allocated for the ongoing support of the repository. The initial installation and support were carried out by the Web systems librarian, who relied heavily on the DSpace-tech listserv¹ for support and advice. Shortly after launching DSpace,...