Foundations for the Digital Library

Library Home | Catalogs | E-Resources | Get Help/Ask Us | Penn Library A to Z

Digital Library Projects
	New and noteworthy
	Locally developed digital collections
	Research and development
	Tools and technology
	Publications and presentations
	Contact us

Introduction	A robust digital library infrastructure depends both on people and on technology. Our people need to be able to work intelligently and efficiently, come up with new ideas to meet our scholars' needs, and be able to implement these ideas in a rapidly changing environment. At the same time we need to preserve and maintain the knowledge bases and services our scholars have come to depend on. In order to make this happen, we need a sound and adaptable technical architecture. For more information, contact John Mark Ockerbloom, our Digital Library Architect and Planner (ockerblo@pobox.upenn.edu).
Supporting smart, efficient staff	Several of our digital projects are designed to help our digital library developers and the rest of the library staff develop new services and serve our scholars more efficiently. Some of our current and ongoing initiatives are described below:
Communication forums	Staffweb is our portal for the library staff's online communication. It includes news alerts, links to support information, policies, and best practices, data we collect about library usage, and sites for development groups and committees. We are now working on turning it into a standard portal environment, which will enable us to repackage and repurpose the information in a variety of contexts. We also run a Wiki, a collaboratively edited web site. Systems and development staff use it for internal documentation of our systems, applications, and procedures. Any registered user can update or correct pages on the site, using a simple markup language. This helps ensure that our organizational knowledge gets recorded and doesn't go stale.
Planning groups	Our digital library steering committee makes strategic recommendations for digital library programs and policy in support of the library's mission and goals. The group consists of selected directors and senior digital library staff. Our developer's group oversees the progress and the documentation of our digital lihrary projects, and looks for ways in which developers in various parts of the library can learn from and help each other. It consists of technically savvy lead developers from all portions of the library.
Helpdesk software	Library staff of all types handle a constant stream of requests. Public service librarians handle requests for information and help from their patrons. Developers handle bug reports and feature requests for their software systems. System administrators also must deal with urgent requests to fix systems when they fail. Our methods for handling these requests have grown up in an ad-hoc manner, and do not always scale well or behave well in "crunch" times. We are currently evaluating "help desk" software to see if we can make our online notification systems more easily used by our patrons, to make sure that the right people are notified of requests, and to make sure that requests get fulfilled in a timely fashion. For more information, contact Michael Winkler, our Web Manager (winkler4@pobox.upenn.edu).
Other efficiency initiatives	Other staff efficiency initiatives we are working on include: a fund accounting "widget" to streamline some of our business and development operations a revised label / shelf list / ID generator to aid in shelving and circulation
Foundational systems	Our conceptual architecture has three levels: information bases designed around knowledge resources, descriptive metadata, and basic management; software tools used to access, manipulate, and administer the resources; services designed around the activities of library users, such as searching for resources, finding related resources, and organizing resources. Tiering our architecture allows the particular technologies in the middle layer to change rapidly, while still providing consistent management of our information resources, and consistent, user-oriented services. These layers are being built, and distinguished, over time as our databases, middleware, and client interfaces develop. Not all of our systems have three explicit layers to them, but the conceptual separation of tasks into these layers allows us to more efficiently plan, evaluate, and reconfigure our systems to meet new needs of our students and scholars. For more information, contact John Mark Ockerbloom - Digital Library Architect and Planner (ockerblo@pobox.upenn.edu). Some of our current and ongoing work in foundational systems is described below.
E-Resources	Our E-Resources database is a collection of metadata for our digital resources. We have written programs that make this database searchable and browsable by our users via our web site. Our home page also has a fast E-resource locator for quicker access to electronic resources. In our Franklin to Web (or Web-to-Franklin) initiative, we export the metadata as MARC records so that scholars can find these resources in our our Franklin catalog database as well. (This export process is largely but not entirely complete at this writing.) For more information, contact Michael Winkler, our Web Manager (winkler4@pobox.upenn.edu).
Uniform Resolution Service	Our Uniform Resolution Service allows us to use more stable, high-level references to digital resources than the fragile URLs of the World Wide Web, while at the same time managing user authentication, and measuring patron usage of our electronic resources. The service uses Handles (developed by CNRI) to identify, and help locate and describe, many of our locally managed resources. Our E-Resources database links Handles to specific Web locations. We also keep track of which resources require authentication, and keep count of aggregate usage of these resources. These URS handles are being integrated into our web pages and library catalog. We hope that this service provides more reliable links to users and better usage data to librarians.
The Typed Object Model	The Typed Object Model allows us to describe the structure and behavior of a wide variety of data formats and information services. In 2003 and 2004, we are undertaking a Mellon-funded initiative to apply this model to support online learning and digital preservation. TOM can be used to document data formats and services, assist in data format migration and other conversions, and provide a uniform application-level interface to heterogeneous data services. TOM's core software is being released as open source. For more information, see our TOM web site.
Other systems initiatives	Other foundational systems initiatives we are working on include: A common authentication, authorization, and access system (integrating PennKey, Penn Portal, and other campus authentication, and which can interoperate in the future with Internet-wide authorization systems like Shibboleth) LDAP directory services for the library A redesign of the SCETI site, bringing it more in line with our core digital library systems A shared development server for building and testing digital library services in a common environment Clustering and updating of our web, database, and courseware systems Readiness plans for prevention and recovery for security compromises, data loss, and service failures