“The Library of Congress, with the bipartisan support of the United States Congress, the Executive Branch and America’s entrepreneurial and philanthropic leadership,” is developing the National Digital Library (NDL). This effort is underway at a number of different organizations and academic institutions. Persons interested in this larger project should visit the NDL Federation’s Web site (http://lcweb.loc.gov/loc/ndlf). There, they will find background information and a list of the organizations involved.
As part of this effort, the Library of Congress is building what they call “American Memory: Historical Collections for the Digital Library.” A good deal has been written about the project; and a few sources are cited above. This review describes the current content of American Memory and the quality of its organization, access, and graphics. The opening page of this site does not provide a good impression. It is busy with a combination of text and graphics which are not well laid out. Furthermore there is insufficient information on the first page to give the user a sense of the scope of the project and what is available. It will take the user some time to explore the site and understand the breadth and depth of its content.
The American Memory project is the conversion to digital images of collections of material, accompanied by finding aids and indexes. The collections typically are subsets from a distinct collection given to the Library of Congress or assembled by them. The material includes photographs and other graphical images, audio recordings, film clips, and texts. All items are either in the public domain or copyright is held by the Library of Congress. Each collection is highly focussed on a specific topic. Two examples are the Daniel A. P. Murray Pamphlet Collection “...which presents a panoramic and eclectic review of African‐American history and culture;” and “California Gold: Northern California Folk Music from the Thirties,” which is a multi‐format ethnographic field collection assembled by Sidney Robertson Cowell for the Northern California Work Projects Administration. (The Web collections do not broadly represent the Library of Congress’s collections on a general topic.)
American Memory currently includes 19 other collections of which the four below are representative examples:
- 1.
1Selected Civil War Photographs, 1861‐1865;
- 2.
2Documents from the Continental Congress and the Constitutional Convention, 1774‐1789;
- 3.
3American Life Histories: Manuscripts from the Federal Writers’ Project, 1936‐1940; and
- 4.
4Votes for Women: Selections from the National American Woman Suffrage Association Collection, 1848‐1921.
These collections are listed under broad subject categories which can be approached through the “Learn” pathway on the home page. Alternatively a user can “Browse” the list by topic or media and identify appropriate collections. Finally there is the option of doing a keyword search on the text of the collections. While browsing a particular collection is a useful approach, if the user is interested in exploring a subject in depth (e.g. Eleanor Roosevelt, Afro Americans in World War II), it is likely the “Searching” option will be the most useful. A search can be performed on an individual collection, all materials in a particular format or across the entire digital library of 21 collections.
If the search includes more than one term the search results are displayed in ranked order based on the frequency and proximity of the search terms. The list provides a link to the individual documents. When the user links to the individual document, background information is provided on what collection the document is from and related information to put the document in context.
The search system is cumbersome in one significant way. If the user moves beyond the list of search results and explores a document and then tries to return to the search results, he or she will frequently get the message: “This document (referring to the search results list) resulted from a POST operation and has expired from the cache. If you wish you can repost the form data to recreate the document by pressing the reload button.” This requires the search results to be reloaded from the host site ‐ a several seconds process that delays smooth movement from one index entry to another.
In some collections, such as The Gottscho‐Schleisner Collection comprised of over 29,000 images, there is a preconstructed index which is approached through a series of menu screens.
Working with the collection at various hours of the day and night, the site was responsive and downloaded graphical, text, and audio files promptly. As an exception, the limits of technology make downloading of film clips lengthy and challenge most microcomputers typically used as Internet navigators. Viewing or film file or listening to an audio clip require a helper application. RealAudio is standard for the audio files; but the digital versions of the film clips can be in one or two of three different formats: MPEG, Quicktime, and AVI. To fully use the film clips, the user must have applications to handle all three formats.
The quality of the graphics leaves something to be desired. The thumbnail pictures are clear and understandable. However, when the full picture is displayed, it is not always clear and sharp. This is a distinct disappointment since it weakens what is otherwise a high quality site. However, without seeing the originals, which in some cases are likely of very poor quality, it is difficult to know whether they could be reproduced at a higher level of quality within the limits of reasonable price and the current technology.
As a group, the 21 collections represent a tremendous body of material that is extremely difficult to quantify by browsing the site. The sizes of some of the individual collections, for example, the Gottscho‐Schleisner Collection with 29,000 images, California As I Saw It: First‐Person Narratives of California’s Early Years, 1849‐1900, with 190 books, and American Leaders Speak: Recordings from World War I and the 1920 Election, 1918‐1920, with 59 recordings, give some indication of the tremendous amount of information included.
Clearly this is a sampling of the Library of Congress’s vast collections; and it was designed to test the various technologies, index approaches, and methods of access necessary to deliver digitally stored information effectively in a variety of formats. As such, the corpus as a whole is not an introduction to American history or even a significant representation of particular eras or themes of American history. It is a sampling of the archival and special collections of the Library of Congress.
As a pioneering effort it is unparalleled and provides a unique set of materials for studying selected aspects of American history. The index tools and finding aids are well constructed and function well, with the one exception noted. This reviewer looks forward to the day American Memory, along with the other National Digital Library partners, has assembled a fuller representation of the breadth of American history and culture.
At this writing, the Library of Congress plans to expand American Memory with the addition of 33 new collections from 1997 through 1999. These collections are listed on a page at the Web site and can be reached from the main page through “Other Previews.”
