A New Convergence

Academic libraries are actively building what we broadly call the digital library: course reserve documents, digitized collections of rare material, online abstract and indexes, electronic monographs and journals. At the same time, there is rapid expansion of network-enabled tools that support student instruction (learning management systems), student assessment (e-portfolios), and the intellectual output of a university (institutional or knowledge repositories). Our next challenge is to embed our digital library collections and services into the new instructional tools and reposition academic libraries and archives in the creation-acquisition-dissemination flow of our institutions' research.

The University of Connecticut is developing an e-portfolio service and upgrading its online learning management system. At the same time, we're building an understanding of the positive impact a knowledge repository could have on research and scholarship. With an eye on these activities, along with the growth of the library's own digital collections and services, we're seeing striking possibilities for integrating systems.

The systems that support digital collections, e-portfolios, electronic classroom tools, and knowledge repositories are specialized content management systems. As with all such systems, each is driven by storage of and access to digital objects (such as text, images, data sets, quizzes, and more) along with associated metadata. Each has a set of rules (the "business logic") that govern the use of digital objects and metadata.

There are similarities in the metadata associated with the digital object. Whatever the system, each object is likely to have a title and creator, a media type and size, and associated rights and permissions. Each also has specialized data to support the application's business logic. A digital collections system has more of what we call "descriptive" and "technical" metadata. Conversely, a learning management system has more metadata associated with grading and comments, student enrollment and usage, and placement within the context of the organization (in a section, department, etc.).

From this broad perspective, the key difference between these systems is the control over digital objects and their metadata. In library systems, the greatest degree of control is exercised by the library and archives staff. In learning management systems, control is with the instructor. In e-portfolio systems, control rests with the student. In knowledge repositories, a "community" of some nature governs the operation.

The differences in control dictate different business rules. For instance, a learning management system has a great deal of programming surrounding the auditing of student use and assessment. A library digital collection system, however, is likely to have a finer grain of searching and browsing capability.

Let's begin to think about how digital library services can be tightly integrated with other electronic support tools at the university. Imagine that an undergraduate takes several classes using the institution's course management system. Some of these are exemplary works that the student places in her portfolio system for others to see. In the senior year, the student binds some of these works together with additional material to represent an honors thesis. This program offers students the ability to "publish" their work in the institution's knowledge repository, giving it a uniquely citable identifier and assurance that the work will be maintained. A community of interest in the knowledge repository also recognizes the student's work and adds it to its area of the repository.

Or, consider the digitized versions of still images of locomotives and rail yards from the early 1900s in the library's archives. An instructor in engineering pulls selected images into lecture notes on a class web site to show the mechanics of a steam engine. Another instructor brings a search box for the collection right into an assignment that asks students to use the collection to analyze the impact of electrification on city utilities. [For more on integration, see "Learning Systems & Us," LJ 10/1/04, p. 34–35.]

So, will our universities buy one content management system to do all of the above? No. Nor should they. Instead, we must recognize the need for digital objects to move among these content systems. We require standards that aid in the linking and moving, translate specialized metadata attributes where appropriate, and provide a scheme of provenance that will reflect how these digital objects are used. Then we must weave our digital library's presence into the fabric of the institution's instruction and research environment.