Digital Libraries
Collaborative Electronic Reviews
Department of Library Science and Instructional Technology
Southern Connecticut State University, New Haven CT

LSC 551-70
Fall 1999

Reviews by:
Karen Ball
Gayle Bogel
Lisa Brenner
Deborah Coretto
Sue Crego
Marilyn Geiger
Mark Gore
Dawn Higginson
Eric Jones
Dianne Lyons
Kathleen Marszycki
Robert Murphy
Kris Piecyk
Jodi Stacy
Amanda Timolat

Tennant, Roy. (1999). Access: Where the Rubber Meets the Road. Library Journal, April 15, 1999 v124 i7 p32(1).

Reviewed by Lisa Brenner

The goal of a digital library is to bring users information that they need quickly, easily and effectively. Access is an important issue when creating a digital library collection because users must be able to interact with the collection in order to have it fulfill their information needs. "A digital collection without effective means of access is nothing but a pile of bits awaiting a purpose" (Tennant, 1999). Roy Tennant discusses access to digital library materials in his article, Access: Where the Rubber Meets the Road.

Often, questions of access are overshadowed by other considerations, such as whose needs is the digital library hoping to serve or what kinds of activities is the library supposed to perform (Tennant, 1999). Yet, for a collection to be truly made available so that the collection may serve the needs of its users, it must consider how the digital library should function. One important consideration is how users will be able to interact with the digital collection. A collection could be made available through a variety of methods or interfaces (Tennant, 1999). Items can be browsed by topic, by searching or by highlighting features. For example, will the users of a digital library be able to properly construct a Boolean search query so that the user will retrieve the appropriate information? Should the collection be organized by topic if users will be mainly browsing the library? These questions need to be addressed as much as who will use the digital library and for what purposes.

Furthermore, a key issue in access is cataloguing. Should a library catalog digital items individually or as a collection? This is particularly important for archival and museum collections (Tennant, 1999). There are advantages to cataloging digital items as a collection, such as not needing item-level records and the entire collection is given context and provenance (Tennant, 1999). In fact, Encoded Archival Description (EAD) is already being used as the text markup standard for archiving finding aids. Nevertheless, it will probably be difficult to offer various mode of access to a particular digital item without item-level descriptions of digital items. Even after locating the proper collection that a particular digital library is in, a user might have to search through an entire collection without the locations or descriptions of specific digital items. Therefore, one might question the usefulness of cataloging digital items as collections only.

Designers should be aware that people need to use a digital library. Usually this requires a certain size typeface and carefully written instructions. Designers should also consider individuals with disabilities when planning a web site. Tennant suggests that designers should use "Bobby" at http://www.cast.org/bobby produced by the Center for Applied Special Technology to check for accessibility problems.

Finally, a designer of a digital library should also consider what technology should be used in designing a digital library. While the designer wants the library to be advanced, cutting edge technology may not be accessible by a large group of users. For example, requiring specific plug-ins or browsers may pose problems (Tennant, 1999). Even creating too complicated a web site may make the site inaccessible for those who do not have very fast machines with large amounts of memory.

Consequently, designers need to think about access when they are designing digital libraries. Designers should customize access depending on the needs and experiences of anticipated users of the digital library. "The more you can align your user interface to the way they (the users) think, the better they will understand how to use your site" (Tennant, 1999). Towards this goal, there are projects that allow users to provide input on their views of library resources, such as MyLibrary at http://my.lib.ncsu.edu and My Gateway at http://www.lib.washington.edu/resource/help/MyGateway.html.

Threaded discussion question:

What should designers of digital libraries be most concerned about when considering how users can quickly, easily, and effectively access the information they need?

Alper, Joseph (1998). Assembling the world¹s biggest library on your desktop, Science 281(5384) p1784(3), Sept. 18, 1998.

Reviewed by Lisa Brenner

Raj Reddy and Michael Shamos, computer scientists at Carnegie Mellon University, are working on a universal library. They envision that a universal library will be accessible from an individual¹s desktop via a simple interface. The interface will resemble search engines, such as Yahoo! Or Excite, but will encompass multiple search tools designed to convert an individual¹s request for information into formats and languages that can search a variety of individual digital libraries. The individual libraries will be able to contain text, two- and three-dimensional images, music, maps and other types of data. The "universal library" project has spread internationally through computer science laboratories. In the United States, the project is building upon other projects funded by the federal Digital Libraries Initiative (DLI-1), which developed various ideas involving how to collect, store and organize information in digital forms and make such information available through networks.

A large challenge to the "universal library" project is how to extract information from a variety of digital libraries. Usama Fayyad, director of data-mining studies at Microsoft Research, states, "We¹re getting very good at building data tombs, huge repositories in which information becomes buried forever. What we need now are ways of getting that information out of these repositories" (Alper, 1998). Currently, key word searching is the means of locating information in a text-based database. Key word searching, however, is only efficient if one is searching in a small database and knows the relevant terminology of a particular topic. A solution to the large return of key word searching in a large database is called Cha-Cha, a program that groups together information based on pathways located in the homepage. A solution to the terminology problem is called a wrapper, a program that would translate a query that is inputted into a digital library into a form that is recognizable by a particular search engine.

Searching images, such as maps, artwork, photographs, and video is difficult to locate online because there is no verbal text to search. Carnegie Mellon University is working on methods of indexing and searching video clips. Computer scientists at Columbia University have developed a tool called VisualSEEK that is a collection of three search engines that search based upon subject matter, physical properties or to a random sample of object types. Other difficulties include copyright and reimbursement issues that could restrict the availability of some works through a universal library, and issues of the required bandwidth to move such large amounts of information around. Furthermore, there would need to be a means of translation if there is to be universal access to this universal library. Ultimately, there would need to be multiple user interfaces that will be able to locate relevant information to satisfy a particular query. There will likely be several task-specific interfaces in order to conduct the tasks of searching text, images, video, music, etc.

Nevertheless, the universal library has promise. With the proliferation of individual digital libraries, there are rapidly growing amounts of information that is now available online. "Hundreds, probably thousands, of digital libraries are quietly amassing huge collections of nearly every type of data imaginable and making them accessible via the Web" (Alper, 1998). The first hurdle to overcome is to determine where all of these digital libraries are located. A worldwide registry of all the digital libraries and their content would likely be the best solution. As Reddy states, "Yes, I believe that over the next 20 to 50 years, the truly universal library will exist and, importantly, that it will be worth the effort and money that we will spend to make it a reality" (Alper, 1998).

Threaded Discussion Question:

What do you think the most important hurdle that would need to be overcome in order to establish a universal library?