Developing an Institutional Repository:
an insiders look at the University of Utah IR

Abstract

The University of Utah libraries (J. Willard Marriott Library, S.J. Quinney Law Library and Spencer S. Eccles Health Sciences Library) have embarked on a campus-wide project to develop an institutional repository. In response to widespread use of the Web and readily available—albeit somewhat new and untested—digital technologies, librarians determined that the University of Utah needed a digital collection to truly represent the scholarly creations of its faculty, researchers and students. This essay focuses on the initial stage of the project and also offers reflections on its current status as well as possibilities for the future.

In the Beginning

The project commenced in 2004 with the creation of a logic model to guide the phases of development. The logic model delineated three distinct phases and allotted each phase two years for completion. We are currently in Phase 2. Phase 1 encompassed the building of infrastructure and the writing of policy. This included deciding upon servers and their location, software, collection development policies, metadata standards, staffing and workflow.

At the bimonthly, IR-devoted meetings, librarians determined to use CONTENTdm as the collection management tool. We primarily based the decision on the fact that Marriott Library already used the software and other academic libraries around the state had implemented it as well. (One of the larger goals of the project is to create a statewide repository and, with that in mind, it seemed economical to use software already in place.) In addition, CONTENTdm figures prominently in archiving images and multimedia; librarians felt those items would become a large portion of the collection. The software also maintains text-based items (such as PDFs) and provides the option of building compound objects which can foster 3D images and ebooks. A system manager was in place at Marriott Library and so was the server. All libraries agreed to consolidate to a single server location and to collaborate with staffing and resources.

With the server and software decisions in place, work on the collection development policy began. The policy articulated the purpose, process, content and future of the repository. Material from award-winning scholars was sought, and select departments from varied disciplines were targeted to build the content and present examples for other departments on campus. Ideally, librarians wanted initial procurement to focus on pre-published material in order to negotiate copyrights and permissions. Retrospective inclusion of already published content for the IR, especially with amiable publishers, also became part of the policy. The policy determined that information about the IR would be communicated to other departments on campus through library selectors.

As the collection development policy was being created so were metadata standards. A number of campus cataloging experts teamed up to look at Dublin Core and create a “best practices” guide. The guide covers the content of each field and the format in which data should be entered. From the larger document, librarians created a short guide for easy referencing which was also placed on the project’s website.

Servers, software, collection development, and metadata standards turned out to be the “easier” components of the project. Staffing and workflow became the more challenging issues. Before the software could be used, communications could be made, or items procured and uploaded, staff needed to be identified and a workflow needed to be developed. Librarians felt uncertain about the amount of time that would be required as this was entirely new for all involved. The IR literature spoke more to why IRs are needed, less to specifically how work patterns and workflows could be efficiently established. The IR team devised a diagram to represent ideal workflow possibilities. This became the foundation of reorganizing existing, traditional work patterns to encompass the additional material processing.

Considering the Present

Material processing, which includes uploading items to the server and applying metadata, is well in place and comfortably moving forward. The bulk of the concern still lies with the copyright and permissions workflow. Seeking permission for peer-reviewed scholarly journal articles tends to bottle-neck the process. In hopes of speeding it up, librarians and library staff created a web-accessible tool to manage all aspects of the procedure: adding the item, the authors, the publishers and their archiving policies, the library staff, and the contacts made between and amongst all possible players. The database is near completion and we anticipate it will greatly improve efficiency.

Coordinators have provided significant training for all staff involved, from metadata specialists to desk assistants participating in the permissions aspect of the process. In addition to our bi-monthly meetings, metadata subcommittee meetings have been held covering software training and addressing specific questions and concerns that arose.

We have also begun training several library specialists in the steps necessary to seek copyright permission. This has involved creation of a wiki to manage the permission templates, the IR workflow system (which contains publishers’ policies) and other pertinent documents and websites such as PubMed and SHERPA. The training is in the beginning phase, but we hope it will greatly impact the speed at which items are added to the repository.

The Flexibility of the Future

We have procured several significant collections, including that of Dr. Mario Capecchi and Dr. Homer Warner, which are providing substantial bedrock for the future growth of the project. Most intriguing is the collaboration between Eccles Library and a large publisher to provide content that the publisher could not maintain within its online textbook.

In reflection, I feel that the benefit of an institutional repository is the ability to be flexible and creative. I see those aspects becoming our convincing selling point. An institutional repository can provide access to materials that remain hidden and inaccessible on individual computers. Posters, presentations, lectures, data sets, learning objects, animations, musical scores and compositions, artistic creations, etc. can be preserved and made accessible—and this is the important part—via Google. This is essential and profound for information seekers, authors, and creators. And, of course, for libraries.

Author's Bio

Allyson is the Institutional Repository Coordinator at Spencer S. Eccles Health Sciences Library at the University of Utah. She also attends the distance MLIS program at the University of Washington's Information School.

Go to Top