It has been written that "Partnership is a key element in stable, long term access to electronic Government information, building on the successful model of cooperation between GPO, agencies, and local libraries for public access to information."
GPO's current emphasis on partnership grew out of the successes of the FDLP and the demands of a task larger than a single entity's capacities. One of the most successful aspects of the FDLP is that various tasks that would be far too large or prohibitively costly for an agency alone are shared among participants in the program, producing value for all. Thus, rather than building a single permanent collection of records of last resort for Government information in traditional formats, regional depositories receive and retain all documents permanently (in general) and in coordination and cooperation with selectives ensure ongoing access in a convenient proximity to end users. By engaging in the partnership type arrangement, far better service to users is provided than GPO could hope to provide on its own.
Similarly, the resources for building permanent access to online Government information are staggering, extremely difficult to manage and very costly for a single entity. So, GPO is experimenting with various models that will bring agency producers of information together with partners in various kinds of organizations and institutions, ranging from individual universities to consortia to large service providers, to seek strategies and know-how for providing storage capacity and effective access to agency electronic information.
The FDLP/ERIC Digital Library Pilot Project is one such experiment. In October of 1997, following negotiations that actually began with my predecessor in ETS, Duncan Aldrich, GPO, the National Library of Education (NLE), and OCLC, Inc. entered an agreement to conduct a one year pilot project with full text of the public domain ERIC reports long distributed in microfiche to depositories.
The project is designed to obtain information concerning electronic archiving, bibliographic access to image files, library and end user behavior and preferences, and other issues relevant to the transition to a more electronic depository program. It specifies that OCLC will provide free and unrestricted access to public domain ERIC reports via Federal depository libraries, will provide technical support for their use, will meet certain performance measures for reliability, security, and usage of the information, and will participate with GPO and NLE on assessment and evaluation of the project.
GPO agrees to coordinate pointers to the collection in GPO Access, provide general user support, and to recognize OCLC as an official site for ERIC digital information. GPO will also coordinate communication with the depository libraries in the pilot.
NLE and OCLC have entered into a separate technical agreement by which image files of the ERIC documents are provided to OCLC by NLE's contractor, ERIC Document Reproduction Service (EDRS).
With this pilot we are taking a valuable and well known product within the FDLP and moving it into the electronic arena, consistent with our goals in the Transition Plan. The pilot will provide much needed information for us on the ramifications and requirements for making large and high-demand collections available electronically through depositories. It also connects us with a very visible and forward-thinking pair of partners. The project has potential for providing a model for many other products and agencies.
I want to emphasize that our objective for this pilot is to gather information on feasibility and comparative costs for this type of data archiving. It is a pilot, and as such it has some limitations that we all need to be aware of. It has a beginning and an end. We certainly hope that there will be sufficient evidence of success and positive response to continue the project beyond its expiration at the end of December. But that is not certain at this time. Likewise, it's important to remember that this is the same subset of ERIC documents that have been available in microfiche, which are the ones in the clearinghouse that are in the public domain. We're very fortunate that this test project happens to be a set of documents that are popular and will give us good data on user behavior and handling of large full-text documents in a high-demand setting.
Let's talk briefly about the mechanics:
The pilot actually began earlier this year when OCLC began receiving digital data from EDRS. The pilot will cover reports from January 1997 forward. Image files are being converted at OCLC and beginning in late June or early July, libraries in the pilot group, the 312 depositories that select item 0466-A-03 will receive packets of information with login and passwords that will allow them to enter OCLC FirstSearch to use the full text and bibliographic data for ERIC reports free of charge.
The pilot group will be phased-in in stages, so that we can work out bugs as we go. Throughout the rest of 1998 and into early 1999 OCLC will be gathering data on use and user behavior, and all three partners will be conferring and analyzing the data for a report to appear in January of 1999.
What do we hope to know? GPO hopes that we will have data on the feasibility of using a contractor such as OCLC for this archiving activity. We need to be able to compare the costs of this route with the costs of other methods. We also will have data on demand for these reports that should be of interest to GPO and NLE, and will have good information on user behavior and expectations.
At this point I'm pleased to introduce the next member of this panel, who I consider one of the great dividends of my "tour" at GPO this year. During this project, Blane Dessy, who is Director of the National Library of Education, and I have had many excellent meetings and transacted much business over the specialties of Chef Ted at the establishment halfway between our offices. Blane and Richard Whalen of his staff are enthusiastic participants in the partnership, and GPO is very proud to be working with NLE (and with OCLC) to bring this about.