Behind the interfaces of the digital libraries and institutional repositories we see today are carefully-planned and proactive processes that ensure that the end user can easily find what s/he needs. With digital libraries and repositories becoming increasingly popular, it is important to know what strategies work well.
As a current MLIS student at the University of North Texas (and new LITA member) with a concentration in digital content management, I had the privilege of attending the LITA session Developing a Sustainable Digital Workflow, at the ALA Annual Conference in Washington, DC. This session exhibited two very successful digital endeavors: the digital collections at my very own UNT, and RUcore, the institutional repository at Rutger’s University. I chose this session because it is my goal to work in digital services/archives once I graduate (7 weeks left!!). Below is a recap from each institution’s presentation on how they handle the management of digital content.
Dreanna Belden (Assistant Dean of External Relations and Head of IT Services) and Hannah Tarver (Metadata Librarian) from UNT Libraries brought the “mean green” to DC. Although I am a distance student at UNT, it was a pleasure to meet some of the UNT librarians who work so hard to make the library such a great resource to both on and off-campus students.
Managing multiple digital projects simultaneously
Dreanna Belden presented a very organized approach as to how the UNT libraries managed digital projects that originated from within the school, as well as other institutions in the Texas area. Using 14 full-time staff and student workers, much of the digitization processes and metadata creation are done in-house. Each collection has its own directory on the server, and within each directory are folders that house digital files at each phase in the digitization process (i.e. raw images, images that have undergone quality control processing, and OCRed images ready for upload).
One process I found particularly interesting was their use of a project wiki to track modifications to the objects. This collaborative approach enables students and faculty to all have access to what is going on in the project. Furthermore, the wiki contains training guidelines, which streamline the training process for any new students who work in the digitization lab each semester.
How UNT Handles Metadata
Hannah Tarver discussed the metadata creation process. For each project, a metadata template is created. The template is also created with a built-in controlled vocabulary, which can be accessed through drop-down menus in the metadata entry interface, or through a browsable subject vocabulary for subject terms. Furthermore, each digital object within a project has an XML template document associated with it so that metadata for each batch upload into the repository is easily searchable. By utilizing a uniform metadata template, common error patterns in the metadata can be found in one batch-proof, instead of having to go through each record individually. The team at UNT uses an open-source text editor called J-Edit, which color-codes the XML templates for readability.
Grace Agnew (University Librarian for Digital Systems) and Jane Otto (Head of E-Monographs and Multimedia Cataloging Section) from Rutger’s University Libraries presented on the innovative structure of RUcore. Like UNT, the digital services departments at Rutgers manage multiple projects for faculty, as well as grant-funded digital initiatives.
Data-Driven Design
Grace Agnew discussed the multi-faceted architecture and organization of RUcore, a repository built on Fedora. Collections are classified based on from where they originated (provenance). Agnew highlights that there is more to metadata than just the descriptive metadata; rights, provenance, and technical metadata have the potential to create richer, contextual relationships between digital objects. An object consists of 2 facets: its description, and the events with which it is associated. She mentions one research project, The Video Mosaic Collaborative, which presents videos of mathematics teaching sessions that can be reused and analyzed to improve math education. Event metadata can be created to track a particular video through its use in other education research project, any changes or interventions made, etc.
Trends and Challenges
Jane Otto further expounded on the use of more than descriptive metadata, and underscored the repository’s ability to accept digital objects in a variety of formats. As the costs of storage space decrease, an increasing trend in digital repositories is the archival of video. She discussed many emerging challenges that accompanied moving images, such as the choice of a metadata scheme that could handle both analog and digital videos; this is compounded by quickly changing standards as the profession moves into the digital realm.
In order to meet these challenges, the importance of cross-training the staff was emphasized. Cross-training not only increases efficiency, but ensures that each employee is well-rounded and flexible enough to respond to a rapidly changing landscape.
Summing it up
The digital departments at both universities took on projects not just from within the institution, but also sought grant-funding and collaboration with outside institutions as well. This underscores the importance of the library’s role in digitization efforts, which in turn provides global access to researchers and users everywhere.
The representatives from UNT presented an established workflow for handling multiple digital projects. Their presentation highlighted streamlining quality assurance, a step that can often become the most time-consuming in a digital project. The team from Rutgers presented utilizing an even richer metadata structure that incorporates more than just descriptive metadata, and presented fresh ways of looking at how objects can be linked in a repository.
It was interesting as a student to see what skills were required for working in digital content. Clearly, technical knowledge is important, especially regarding metadata standards and how they can best be applied to a collection. Yet there were also less concrete skills, such as the ability to envision a collection not just as a stand-alone island, but how it can be molded and integrated into other collections and ideas. Also, there was an importance in finding ways to constantly improve the workflow, and discover how new technologies and processes could make things more efficient.
Overall, this was a very informative session on how institutions are managing their digital content, as well as future developments and challenges to overcome. I wish everyone reading this a happy 4th of July weekend, and for those of you who attended the ALA conference in DC…get plenty of rest, stay cool, and enjoy the holiday!
For more information on the presenting institutions and their collections, please visit the links below: