The Prints and Photographs collection is located in the History of Medicine Division of the Library. This unit contains prints and photographs documenting the social and historical aspects of medicine from the Renaissance to the present. The History of Medicine Division Archival/Retrieval Picture Project (HARPP) is a system for electronically retrieving images; it becomes operational in December 1992.
Patrons of the collection are usually doctors and others in the health fields who are writing articles and books and staff from publishing companies searching for illustrations. However, it is not unusual to find individuals with commercial interests ranging from someone creating wall hangings for doctors' offices to the writer of a television documentary. The traditional approach, typical of many still picture collections, requires anyone searching for a specific print or photograph to journey to Bethesda, Maryland. Once there, the researcher goes through a card catalog containing a brief description of the item, card by card, looking for what is needed. When an item in the card file is selected, the researcher asks a librarian to retrieve the original. Once the item has been viewed, the researcher may request a photograph. If the item has been requested before, a photograph is made from a negative; otherwise the original is set aside and, once a week, a photographer under contract comes to the library to make several negatives of the item; afterward, prints are made. The whole process takes a minimum of two weeks.
There are several obvious disadvantages to this system. First, it requires a researcher to travel to Bethesda without knowing beforehand whether or not the required image even exists. Then it is necessary to look through a card catalog to select an image based on a description of a few words. A librarian must spend time retrieving the item requested; then the item must be handled by several people, leading to wear and tear of the artifact, some of which date from the fifteenth century. If the item is selected, a researcher must then wait for a photograph to be made from a negative or perhaps from the original; the latter generally necessitates mailing the copy. In the early 1980's, someone at the Library suggested putting the collection on videodiscs to reduce handling. In 1982, a pilot project was undertaken in which 1000 pictures were stored on a disc. The disc and a catalog were sent to approximately 100 libraries, and the response was universally favorable.
Not only is the collection going to be available to anyone with the necessary equipment; there will now be a way for anyone, located anywhere, to use the collection. When an item is desired, it will be ordered from an organization contracted to store the negatives and positives and make prints from the 35mm negative or positive used to create the videodisc. This will allow the library to concentrate on its primary mission of obtaining and maintaining items for the collection.
Those who obtain the package of discs will not be precluded from copying noncopyrighted material; however, the quality of the copied image will be not that good, and most users will lack the sophisticated equipment to enhance the copy. Also, each of the stored images has been reduced in size, but the true proportions are preserved. There are no close-ups; instead, the full image is shown with a mat around it.
In the process of capturing the collection on discs, the library has succeeded in reducing the amount of staff time needed to serve patrons, since it is no longer necessary to search for and retrieve items from the archives or to assist patrons in formulating a search strategy. Because actual handling of items in the collection has been reduced, there is less opportunity for misfiling and less time spent searching for misfiled items. Wear and tear on the collection caused by handling has been almost eliminated, and the library will shortly be freed of the laborious task of providing copies of images to patrons. Furthermore, the project will offset its original cost by reducing overhead. It should not be forgotten that this has all been accomplished using an approach that will produce some revenue, albeit small, due to the expected low price of the discs.
The process of photographing the images was done in the following way: An item was placed on a camera stand and a bar code reader, run by the operator, recorded the number. The bar code was then automatically integrated with the picture by using a special computer. Next, the operator used a laser pen to mark the edges of the item. The computer forwarded the dimensions of the picture to the camera, which was then automatically focused and the picture taken. Originally, it was hoped that 500 items could be photographed daily, but it turned out that 300 images were deemed a good day's work.
A third person was employed full-time for an equal period to create the catalog records. The source for this information was the cut line (i.e., caption) for the image and other information contained in the card catalog. The library staff provided whatever assistance was needed. The videodisc was then mastered by an outside firm using the 35mm film, and up to 54,000 images stored on each side of the disc. The control program that integrates the videodisc and CD-ROM was written by library personnel.
A four year old could learn to operate the system, which is as easy as changing channels on a television set. For example, an image having the cut line "U.S. Veterans Administration Hospital, Muskeegee, Oklahoma," could be located and retrieved by entering "Oklahoma, Muskeegee, Veterans U.S., Veterans Administration," and so on. There are plans to enhance each catalog record with complete information, including standard subject words and terms. The ability to perform compound searches enables the researcher to isolate topics quickly; for example, a search of only the first 13,000 images using the term "nurse" produced 542 possibilities. Searches based on the terms "nurse" and "Russia" yielded two images. With help from the collection's curator, who is familiar with keyterms, the search process can be refined considerably to locate images that are particularly difficult to uncover.
For the foreseeable future, information will be distributed to users via the package of discs. However, it is anticipated that, some day, cataloging information and perhaps even images will be available on-line through Medlars or a similar dial-up access. Due to the bandwidth requirements for transmitting images, this is unlikely to occur soon, but it would be possible for organizations with suitable networks in-house to allow users to access contents from those locations served by the network. But even without on-line access, HARPP is a tremendous boon to researchers and an excellent example of how an organization can extend the scope of its services through technology.