Modern libraries are complex systems that consist of many procedures and functions. Traditionally, these functions (subsystems) have included acquisition of materials, cataloging and classification, circulation and interlibrary loan, serials management, and reference services. The most important function, however, has been the provision of service to the users. For centuries, librarians have managed warehouses of documents by acquiring, cataloging, and classifying books, journals, and other materials, and circulating them to their clients. Computer and telecommunication technologies have empowered the new breed of information professionals to select, organize, retrieve, and transfer (SORT) the actual information effectively and efficiently to the users.
In the Beginning
Historically, the most labor-intensive operation of a library has been circulation, the main goal of which is to retain a record for each item that is borrowed from the library. This transaction record usually contains information about the material (e.g., call number, title, and author), as well as information about the borrower (e.g., name, address, telephone numbers). The record also includes two other important items: the borrowing date and the due date. Up until the mid-1970s, the library staff, or in some cases the user, would enter the circulation transaction information on a special card for each borrowed item and then file the card in a prespecified sequence (e.g., by call number, title, or date due back to the library). When the book was returned, the appropriate circulation card would be pulled from the file, and the book would be returned to the shelves. Maintaining the circulation file, however, was a time-consuming task. Every day, the file would have to be updated, which involved merging and sorting the new transaction cards with the master file, locating overdue books, and identifying items that were requested by other users. In addition, individual overdue notices had to be written and sent, fines had to be calculated, and library users who had outstanding fines had to be identified.
To reduce the cost and increase the efficiency of the subsystem, library managers sought new approaches. While a few libraries used microfilm technology to record transactions and thereby reduce labor expenses, most libraries directed their attention to computers to automate their circulation operations. Using the batch-processing technology of the mid-1960s, the staff would record the transactions on punch cards, which would then be processed overnight, resulting in a printed master list of all the items borrowed from the library. Although the automated circulation systems were primitive by modern standards, they were a cost-effective solution that allowed a library to provide a better service to its clients.
Meanwhile, librarians were embarking upon another venture, which proved to be a pivotal point in the history of library automation. One of the most important functions in a library is cataloging and classifying individual items. Creating such a bibliographic record is time consuming because it requires professional librarians to apply the Anglo-American Cataloging Rules to each item. To curtail the costs and raise the productivity of the system, librarians and library technicians have copied the cataloging information for individual documents from the Library of Congress and other institutions. In the early 1960s, a few libraries formed informal networks to exchange their printed book catalogs to decrease the amount of original cataloging that their staffs had to perform. Toward the end of the 1960s, the Library of Congress took a leading role in using computer technology to establish a project for exchanging cataloging information. Under the leadership of Henriette Avram, MARC (Machine-Readable Cataloging) was developed as a protocol for storing bibliographic information in a standard format to facilitate exchange of cataloging records. In the ensuing decade, MARC computer tapes allowed many libraries in the United States and Canada to exchange cataloging information. It also facilitated the production of other products such as catalog cards and microfiche. The Library of Congress experiment was so successful that many other countries embraced the MARC record structure (with minor local variations), and eventually it was adopted by the International Organization for Standardization (ISO).
In the early 1970s, another important development shaped the future of library automation. Lockheed Missiles and Space Company introduced Dialog, an online information service that provided access to a variety of bibliographic databases. Many large academic libraries began to offer specialized services for scientists who required literature searches for their projects. As the demand for online access to information grew, so did the number and the size of databases. Between 1975 and 1999, the number of databases grew from 301 to 11,681, and the number of records in these databases increased from 52 million to 12.86 billion.
The technological progress occurring in the last half of the 1960s and in the early 1970s led to the introduction of turnkey systems in libraries. Computer hardware and software were combined to provide libraries with an exclusive and dedicated system to automate circulation operations. Turnkey systems usually consisted of a minicomputer, dumb terminals (i.e., stations without a central processing unit), and software to manage check-in and check-out of documents and to issue overdue lists.
By the end of the 1970s, many libraries had automated some of their functions, mainly circulation, cataloging, and, to a lesser extent, reference services. This was also the era of the development of the Ohio College Library Center (OCLC), known as a "bibliographic utility." OCLC was one of the first computer-assisted library cooperatives, in which fifty-four college libraries joined to share their cataloging records. OCLC, which was later renamed the Online Computer Library Center, Inc., provided online access for more than thirty thousand libraries in sixty-five countries to its vast MARC database and produced millions of catalog cards for libraries around the world. Shared cataloging, though relatively expensive, enabled many large libraries to begin transferring cataloging and classification duties from professional librarians to library technicians.
Libraries also began to convert their old card catalogs into machine-readable records. Many large-scale retrospective conversion (RECON) projects, while costly, were underway or had been completed by the mid-1980s. Academic institutions were in the forefront of RECON projects, creating millions of MARC records based on their holdings in their main and branch libraries.
These first automated library systems required a different type of bibliographic record for each function (e.g., cataloging, circulation, and acquisitions), which resulted in inefficiencies and delays in entering data and in a lack of quality-control standards. Technological advances and market demands required the vendors of library automation systems to develop a new generation of powerful integrated systems. These systems were designed to use a single bibliographic record for all the library functions. A unique MARC record allows every book (or any item) to be tracked from the moment that it is chosen for acquisition by a library to the time that it is available on the shelf for the user. At each subsystem, the MARC record is enhanced and augmented with the additional information that becomes available about the book (e.g., when the order is placed). Libraries, which have completed their RECON projects, can transfer their large bibliographic databases to the new integrated systems and automate all their operations.
Acquisitions and serials management were the last modules to be incorporated into the integrated systems. Procurement of library materials involves complex functions such as online ordering, invoicing, accounting, and claims for unfulfilled orders. When a book is considered for selection, the automated system allows library staff to enter minimal bibliographic information about it. The incomplete record is then augmented with new information to form a MARC format record as soon as the item is acquired and received by the library. Some bibliographic utilities offer libraries time-sharing access to their large databases for acquisition purposes. Among these is the Research Libraries Information Network (RLIN), which supports a number of functions such as preordering, standing orders, and in-process information. Serials management is one of the most complex operations in an online environment. Tracking publication patterns of individual journals, automatic claiming of late arrivals or missing issues, and maintaining binding information are a few examples of the activities performed by the automated serials management subsystem.
Perhaps the greatest effect that automation had in the 1980s (certainly the most visible) was the introduction of online public-access catalogs (OPACs). The new online catalogs quickly gained wide acceptance among the public, who preferred them to traditional card catalogs. The first generation online catalog was simply an extension of the card catalog and had limited capabilities. These OPACs provided users with a few access points (i.e., ways to find library materials), generally the traditional author/title/subject headings, using character-by-character matching to retrieve bibliographic records. Despite the limitations of the early OPACSs, library patrons preferred them to the card catalog, since these online systems provided patrons with information about circulation status (e.g., whether books were checked out or in the library).
Another significant technological development was the introduction of CD-ROM technology. Although the first bibliographic database on CDROM only appeared in the mid-1980s, by the end of the decade, hundreds of titles on CD-ROM were available in the marketplace. Libraries were among the first organizations to adopt the new technology, since librarians and information professionals realized the potential of the CD-ROM as a storage medium for vast amounts of information. This technology was used to provide access to a variety of bibliographical records, including MARC cataloging information. Many libraries used CD-ROMs to supplement or even replace the online utilities as a cost-saving measure.
As library functions became increasingly automated, libraries encountered problems in linking different computer systems, interoperability of systems, and implementing the client/server computer architecture. To alleviate these problems, ISO developed the Open Systems Interconnection Reference Model (OSI) in the early 1980s. This model consists of protocols for a layered communication system, which simplifies the movement of data between various computers. ISO also developed another set of protocols, referred to as Search and Retrieve Service Definition and Protocol Specification, to facilitate search and retrieval of information.
At the same time in the United States, the American National Standards Institute (ANSI) and the National Information Standards Organization (NISO) proposed the Information Retrieval Protocol for managing bibliographical records. This protocol was later modified and became known as the Z39.50 standards.
Z39.50 uses the client/server model to send messages or protocol data units (PDUs). The Information Retrieval Protocol has eleven "facilities": Initialization, Search, Retrieval, Result-Set-Delete, Browse, Sort, Access Control, Accounting/Resource Control, Explain, Extended Services, and Termination. This elaborate scheme, in conjunction with an abstract database model, has been developed to accommodate the differences among server databases. The client may specify the record structure and data elements to be retrieved, preferred syntax (e.g., different MARC formats), naming of the result set, and subsequent deletion of the set. The server should be able to provide access and resource control for the client. The standard also has provisions for security passwords, charging and billing, scanning terms in lists and indexes within the browsing facility, and sorting.
Several other standards that facilitated the management and communication of the digital information were proposed and drafted by the end of the 1980s. Along with information retrieval standards, new MARC communication standards were introduced (i.e., MARC 21) to include not only traditional bibliographic fields but also information about computer files, music, maps, and multimedia materials. In addition, the Unicode project, which began in 1988, responded to the lack of a consistent international character set and led to a set of standards for encoding multilingual text. Unicode is modeled on ASCII coding but uses a 16-bit schema rather than an 8-bit encoding system.
Standard Generalized Markup Language (SGML), initiated by ANSI in 1978, was designed as a way to separate content from style and as a means of marking up any type of text so it can be effectively handled and managed by any type of computer. SGML identifies and names digital information to be used in a variety of products and services, such as indexing, typesetting, hypertext manipulation, and CD-ROM distribution. SGML, which was approved and ratified in 1986 by ISO as an international standard, formed the basis for other markup languages such as Hypertext Markup Language (HTML) and Extensible Markup Language (XML).
Although computer networks were first developed in the 1960s and the first e-mail was sent in the early 1970s, it was not until the late 1980s that computer communication systems were widely used in libraries. File Transfer Protocol (FTP) was used for transfer of large data files, and e-mail service was used for fast and efficient interlibrary loans. Telnet, however, had the greatest effect on information services by allowing users to have remote access to libraries. Researchers no longer had to rely on librarians to find information in distant libraries or even travel to many locations to search library catalogs.
As telecommunication technology progressed at a rapid rate, so did computer hardware and software technologies. The introduction of graphical user interfaces (GUIs), particularly the Windows operating system, had a profound effect on library automation. Vendors promptly converted their character-based systems designed for the older generation mainframe computers to Windows-based GUI systems that used client/server architecture. Librarians began hastily to write request for proposals (RFPs) and seek funding to upgrade their outdated automated systems. The users were the real benefactors of the new systems, since they would no longer need to learn and memorize long commands to search, retrieve, and display the desired bibliographic records.
Throughout the 1990s, the pace of development in libraries matched the changes fueled by the introduction of the World Wide Web. Many library automation vendors adopted the Z39.50 standards and created web-based catalogs. New products included features of the previous generations of online catalogs, such as author/title/subject or keyword searching, Boolean operators, and truncation capabilities, but they also provided users with new ways to search for and display the requested information. The implementation of Z39.50 standards ensured uniformity among web interfaces in incorporating these features.
While web-based systems were being developed, the Library of Congress, OCLC, and other organizations sought new methods for enhancing the contents of millions of cataloging records in their databases. MARC format was augmented by adding a new field in its record structure to reflect the availability of information resources on the web. The Electronic Location and Access field was added to contain the Uniform Resource Locator (URL) of the item, linking the record to any type of digital unit. The unit may be a full-text document, still image, audio and video segment, or a software program.
The instability of URLs, however, poses a serious problem for maintaining large bibliographic databases while assuring high-quality control standards. Through the establishment of new systems, protocols, and standards, such as Persistent Uniform Resource Locator (PURL), Uniform Resource Name (URN), Uniform Resources Identifier (URI), Open URL, and Digital Object Identifier (DOI), many of these obstacles may be overcome. Another approach to providing access to the varied items on the web is the development of metadata standards, introduced in the mid-1990s as a way to describe the attributes and contents of digital items. Metadata describe the content of a document in a formalized way and have been used in different contexts. The Dublin Core Metadata Element Set, for example, outlines fifteen elements to identify a web-based document. The Resource Description Framework (RDF) is another example of metadata used to describe web resources to accommodate interoperability between various applications.
The Digital Library
The terms "electronic library," "e-library," "virtual library," and "digital library" have been used interchangeably to describe a new phenomenon— the development of digital information warehouses. The digital library, encompassing many concepts, was best defined by Christine Borgman (1999, p. 233) as "(1) a service; (2) an architecture; (3) a set of information resources, databases consisting of text, numbers, graphics, sound, video, etc., and (4) a set of tools and capabilities to locate, retrieve and [use] the information resources available."
In 1993, a consortium of several institutions, including the National Science Foundation and the National Aeronautics and Space Administration, funded the first Digital Library Initiative. The consortium sponsored a few large-scale projects to investigate the technological feasibility of developing digital libraries to contain information sources that can be accessed through the Internet. Digital Library Initiative-2 (DLI-2), which involved ten sponsoring organizations and more than twenty projects, was launched in 1998. The focus of DLI-2 was on the social, behavioral, and economic aspects of digital libraries.
Libraries have automated their subsystems to provide better service to their users. At the same time, advances in computer and telecommunication technologies, new standards for storage and retrieval of documents, and the World Wide Web have dramatically changed the functions performed by librarians. Automating libraries now signifies the transfer of digital information, regardless of the medium, shape, size, or form, from the producer to the consumer. Library automation has been transformed to information automation.
See also:Community Networks; Databases, Electronic; Internet and the World Wide Web; Librarians; Libraries, Digital; Libraries, Functions and Types of; Libraries, History of;Library Associations and Consortia; Preservation and Conservation of Information;Retrieval of Information.
Barber, David. (2000). "Internet-Accessible Full-Text Electronic Journal & Periodical Collections for Libraries." Library Technology Reports 36(5):3-111.
Barry, Jeff. (2000). "Delivering the Personalized Library." <http://www.libraryjournal.com/automated-marketplace/2000/delivering.asp.
Borgman, Christine L. (1999). "What Are Digital Libraries? Competing Visions." Information Processing and Management 35:227-243.
Boss, Richard W. (1997). The Library Administrator's Automation Handbook. Medford, NJ: Information Today.
Boss, Richard W. (2000). "Information Technology Standards." Library Technology Reports 36(4):3-110.
Breeding, Marshall, ed. (2001). "Library Technology Guides: Key Resources and Content Related to Library Automation." <http://staffweb.library.vanderbilt.edu/breeding/ltg.html>.
Cibbarelli, Pamela, ed. (2000). Directory of Library Automation Software, Systems, and Services, 2000-2001 edition. Medford, NJ: Information Today.
Cooper, Michael D. (1996). Design of Library Automation Systems: File Structures, Data Structures, and Tools. New York: Wiley.
Crawford, Walt, and Gorman, Michael. (1995). Future Libraries: Dreams, Madness, and Reality. Chicago: American Library Association.
Dzurinko, Mary, and Platt, Nina, eds. (2001). "Integrated Library System Reports." <http://www.ilsr.com>.
Evans, Peter, ed. (2001). "Biblio Tech Review: News, Analysis, and Comment." <http://www.biblio-tech.com>.
Healy, Leigh Watson. (1998). Library Systems: Current Developments and Future Directions. Washington, DC: Council on Library and Information Resources.
Leiserson, Anna Belle, ed. (2001). "AcqWeb."<http://acqweb.library.vanderbilt.edu>.
Library of Congress. (2001). "Z39.50: Gateway to Library Catalogs." <http://lcweb.loc.gov/z3950/gateway.html>.
Lloyd, Naomi, ed. (2000). "Library Automation Resources." <http://www.escape.ca/~automate/resource.html>.
Meghabghab, Dania Bilal. (1997). Automating Media Centers and Small Libraries: A Microcomputer-Based Approach. Englewood, CO: Libraries Unlimited.
Osborne, Andres, ed. (2001). "Library Automation Pages." <http://www.escape.ca/~automate/resource.html>.
Saffady, William. (1999). Introduction to Automation for Librarians, 4th edition. Chicago: American Library Association.
Scott, Peter, ed. (2000). "Lib Dex: The Library Index."<http://www.libdex.com>.
SOLO Librarian's Listserve. (2001). "Survey of Library Automation Systems in Use at Various Libraries." <http://www.alrc.doe.gov/library/autosurv.html>.
Williams, Martha E. (2000). "The State of Databases Today: 2000." In Gale Directory of Databases, Vol. 1, ed. Marc Faerber. Detroit: Gale Group.