The University of Pennsylvania's Online Computing Magazine

PENN PRINTOUT March 1997 - Volume 13:7

[Printout | Contents | Search ]

Major changes to Franklin online catalog and Library databases

By Paul H. Mosher

Since its creation in 1985, Penn's online library catalog, and later the Library's core information bases, have relied on 1970's mainframe technology to deliver growing amounts of information electronically. Today, mainframe costs and capacity limitations are hampering our ability to support the changing information needs of teaching and research. To address this problem and provide a more secure foundation for Penn's growing Digital Library, the Library plans to retire the outdated mainframe system. By early summer, Franklin and the other mainframe-mounted information bases will reside on new computing platforms searchable on the World Wide Web.

The Library will report in greater depth on the reasons and benefits underlying the computing migration in the coming weeks. It will also work to reduce the uneasiness the migration may cause users. To begin this process, the Library invites you to watch for reports -- published on the Library Web site and in print -- and to provide input into system designs and goals, when, in a few weeks, a preview of the new Franklin is available.

Inevitably, a few potholes will mark the transition to more powerful, distributed computing systems. But the investment in new technologies will pay dividends long after the Library corrects any initial problems that may arise -- problems staff are working overtime to anticipate and minimize. In the short term, the new systems will enable the Library to take fuller advantage of the Web and the current move to client/server computing on campus. They will provide greater capacity to store and deliver electronic information at lower costs than presently possible. Over the long term, the new systems will allow the Library to respond quickly to software and hardware innovations, networking improvements, and the emergence of new digital media. And they will aid in converting redundant paper-based activities to more effective and cost-beneficial electronic processes, thus changing the Library work place for the benefit of users. Technology will free librarians to increase instructional outreach and help faculty and students negotiate the expanding universe of print and digital information. lt will also help the Library deliver many frequently-requested paperless services to the desktop, as the new systems mature.

The new Franklin will contrast sharply with the mainframe system. Some of the major differences include:

  • the graphical Web interface, featuring icons, pull-down menus, and point-and-click commands
  • the ability to recall books using Web pages
  • the ability to view your borrower record and monitor books you currently have on loan or recalled
  • a new keyword search capability that ranks by relevancy, in addition to traditional author, title, subject, call number, and full Boolean keyword searching
  • the ability to limit searches by date, library location, and type of material
  • hypertext links on subjects and authors
  • hypertext links to online resources, including more than 1,200 electronic journals

The software supporting Franklin is marketed under the name Voyager and engineered by Endeavor Information Systems. Endeavor is unique in its willingness to partner with a group of academic libraries in developing the Voyager catalog. As a contributor to this partnership, the Library's goal is to achieve the most effective online catalog possible -- one that combines the best elements of known systems with features customized for the specific information needs of the Penn community. If you'd like to explore a similar implementation of Voyager today, visit the University of Rochester Library.

For the citation bases that also reside on the mainframe, the Library is presently testing a Web-searchable system designed by OVID Technologies. OVID will feature a sophisticated Java interface which operates within the Web environment to accelerate and enhance the functionality of database searching. Over time, OVID also will provide the capacities needed to increase the number of networked bases offered. The files migrating to OVID include ABI/Inform, Current Contents, PsychInfo, and the five Wilson bases. The transition to OVID will be phased in over the summer months, with parallel access to the mainframe versions of ABI/Inform and other citation files available through September 1997. Please note that the Library will continue to provide character-based, telnet access to both Franklin and the OVID files for users who cannot or prefer not to search over the Web.

The Franklin migration to Voyager will take place in June. Given the size of Library systems, some level of service disruption is inevitable as staff convert large data sets, and transfer acquisitions funds, account ledgers, in-process files, and circulation records between platforms. To minimize inconveniences for users during this complex migration, the Library will be sensitive to the academic calendar and the need for continuous, reliable service. The Library will also make communication a high priority, and disseminate regular updates of progress and advance notice of potential service interruptions. Training is another important focus. Penn librarians are developing outreach programs now to teach students and faculty to make optimal use of the changing resources.

If you have questions, staff in all of the Libraries are prepared to provide informative answers. Alternately, please contact me at 898-7091 or send e-mail to

PAUL H. MOSHER is Vice Provost and Director of Penn Libraries.