Washington, D.C.
October 15-17, 1997
Jan Olsen
Associate University Librarian for Sciences and
Director, Mann Library
Cornell University
Good morning. We will be presenting a case study that I believe is an application of the concepts and ideas that you have heard presented this morning. I will give an introduction to set the stage and then Evelyn Frangakis, Preservation Officer at the National Agricultural Library, will follow up with greater detail.
Several years ago, the Mann Library at Cornell University and the United States Department of Agriculture formed a partnership to deliver to the nation digital information that the Department of Agriculture was producing. In particular, it involved three economic agencies within the Department: the Economic Research Service (ERS), the National Agricultural Statistics (NAS), and the World Agricultural Outlook Board (WOAB). These three provide the crux of the economics work done in the Department of Agriculture. They produce a great deal of information of tremendous vitality and use to the United States and whose native form is digital. However, when the information aged a little and when storage space was running out, it was dumped. Important public information was lost.
Rather than lose that information, Cornell has agreed to take that data as it is produced, and deliver it across the United States to anyone who wishes to use it, free of charge. We receive extremely time-sensitive data, re-format it, and within 30 minutes have it available on the Web, at http://usda.mannlib.cornell.edu/usda/.
Let me define what kind of data is involved. It is data that was produced for public use. A lot of the information is statistical, and much of it is textual. In the last calendar year, there were 2.1 million files of this data downloaded from the Mann Library site. It is popular and used heavily.
The USDA provides a grant that purchases storage space. Mann Library contributes the staff time, programming, and reference support for the data. In any given year, Mann receives about 150,000 information requests. The contract has just been renewed with the Department of Agriculture. In this coming year the system will be enlarged and improved so that it is a comprehensive economics information system.
Of particular concern with this system is the fact that the data is aging. As a library, we must not just dump the data. I mentioned that these are natively digital; there is no printed equivalent. So the question of archiving, maintaining, and sustaining this data is a very large and important one.
This problem is a national challenge. It is an example of government data to which this nation should have access, but it is government data in need of systematic preservation. While that is true for the USDA, it is also true for other government departments. We decided to treat this as a national issue. As a first step, we worked with the National Agricultural Library and planned a national, invitational conference. We invited approximately 70 people who had a vested interest in the need to preserve this data, and held a two-day conference at the Smithsonian Institution.
There was a good cross-section of participants. Attendees included USDA and other government people, those who had explored the issue of preserving digital information in the past, academics, and average users. The purpose of the conference was to encourage the USDA to acknowledge the need to preserve such data and to begin to work out a plan for preservation. I think it’s fair to say that we achieved this objective. The National Agricultural Library accepted responsibility to get planning underway and to put in place strategies and policies, ultimately passed by the USDA, to archive digital data in a systematic fashion.
The final outcome of the conference was that the USDA recognized at the very highest level the need to work on this problem in a rigorous way, and subsequently a report with a plan for preserving digital data.
Evelyn will now describe to you the projected next steps.