Thursday, August 23, 2007

Quality Digital Repositories: Reviews

Book Shelf

Historical Voices

Historical Voices was created to be a robust, fully searchable digital library of 20th century spoken world collections—the first significant repository of its kind. Using audio files as a key component, Historical Voices provides storage, online exhibits, and educational sources for students, teachers, researchers and the public.

Historical Voices is part of the Digital Library Initiative II funded in part by the National Science Foundation and the National Endowment for the Humanities. It is housed at Michigan State University’s MATRIX: the Center for Humane Arts, Letters, and Social Sciences (matrix.msu.edu). Historical Voices continues to build its collection by working with universities, historical societies, archives, and research partners around the world. Its goal is to create a trans-institutional resource and a federated archive. Additionally, it strives to create best practices in sound digitization and develop a digital archive structure and a metadata format.

Currently, it provides links to featured galleries, such as “Studs Terkel: Conversations with America.” The search feature and best practices are being developed. However, Historical Voices provides PDFs of research it has conducted, such as “Digitization Projects: Key Issues for Archivists, Curators and Librarians.” In its Education section, it has a small collection of sample lessons built around audio selections.

On-Line Books Page

The On-Line Books Page offers access to thousands of free online books and serials in English and foreign languages. Substantial links to directories, archives, and featured exhibits, such as Banned Books Online, are available. Users range from scholars to the intellectually curious.

The design of the page is spartan, yet effective. With a simple search interface, the user can browse books by Library of Congress Subject Headings or by Library of Congress call number. Titles and authors are both searchable and listed alphabetically. When clicking on an icon in front of the listing, the user can learn about a particular book, find other books with the same author, title, or subject, or find out how to make a stable link to the detailed book description.

The transparency of the On-Line Books Page is helpful. For example, the criteria for a listing are explicit; it must be available at no charge, a full text of a significant book in English, and a stable, well-formatted text in a standard format. A list of requested books and books in process show what will be available soon.

The repository was founded in 1993 by John Mark Ockerbloom, a digital library planner and researcher at Penn State, where the page is hosted. Its development, maintenance, and content are all volunteered.

The Modern English Collection

This collection, arranged for browsing by the author’s last name or by category of interest, contains fiction, nonfiction, poetry, drama, letters, serials, manuscripts, and illustrations from 1500 to the present. There are 10,000 texts and 164,000 images publicly accessible. Users access a search engine for results, which can be viewed at different levels of details, such as by summary, chapter, or page.

Categories include texts by African Americans, texts by women writers, texts about the American Civil War, texts by and about Thomas Jefferson, texts for young readers, literature in translation, and items from the University of Virginia’s special collection. Each text is encoded in either SGML or XML and includes a bibliographic header with administrative data about the creation of the electronic text and its print source.

The Modern English Collection, founded in 1992, is part of the Electronic Text Center at the University of Virginia. The goals of the center are to create an online humanities archive of standards-based texts and images and to foster a community of users as a traditional library would. The Center strongly promotes the standardization of data in XML because it can reshape itself so it never becomes outdated and useless.

Project Bartleby

Bartleby.com claims to be the most comprehensive reference publisher on the web, meeting the needs of students and educators. Named after Herman Melville’s character Bartleby the Scrivener, this site features public domain literature in the classics, nonfiction, and reference. It is searchable within categories, such as fiction, nonfiction, verse, reference, and by author, title, or subject. The user can also browse by clicking on titles, authors, or similar works or authors to find more information of interest. Poetry is indexed chronologically or by first line. Biographic information on most of the authors is also available.

Unlike the other digital repositories featured in this paper, Project Bartleby is a commercial site and, probably not by coincidence, the most comprehensive. Ads frame each page but are usually ignored due to the user’s banner blindness.

Project Bartleby began in January 1993 at Columbia University as a project to convert public domain resources into digital archives. It was constructed according to four principles: accurate editions created by professional editorial standards; free public access; careful, well-researched selection of works with public domain or copyright license; and up-to-date presentation using electronic-publishing methods as they become standardized.

No comments: