In today’s ever-changing world of technology and information, a growing number of organizations and universities seek to store digital documents in an online, accessible manner. These digital library (DL) repositories are powerful systems that allow institutions to store their digital documents while permitting interaction and collaboration among users in their organizations. Despite the continual work on DL systems that can produce out-of-the-box online repositories, the installation, configuration, and customization processes of these systems are still far from straightforward.
Motivated by the arduous process of designing digital library instances; installing software packages like DSpace and Greenstone; and configuring, customizing, and populating such systems, we have developed an XML-based model for specifying the nature of DSpace digital libraries. The ability to map out a digital library to be created in a straightforward, XML-based way allows for the integration of such a specification with other DL tools. To make use of DL specifications for DSpace, we create a DL generator that uses these models of digital library systems to create, configure, customize, and populate DLs as specified. We draw heavily on previous work in understanding the nature of digital libraries from the 5S framework for digital libraries. This divides the concerns of digital libraries into a complex, formal representation of the elements that are basic to any minimal digital library system including Streams, Structures, Spaces, Scenarios, and Societies. We reflect on this previous work and provide a fresh application of the 5S framework to practical DL systems. Given our specification and generation process, we draw conclusions towards a more general model that would be suitable to characterize any DL platform with one specification.
We present this DSpace DL specification language and generator as an aid to DL designers and others interested in easing the specification of DSpace digital libraries. We believe that our method will not only enable users to create DLs more easily, but also gain a greater understanding about their desired DL structure, software, and digital libraries in general.
