Session III: Building Information Retrieval Systems for Science

• Development of an Information Retrieval System Based on Superimposed Coding (J. M. Cretsos)
• The Development of Secondary Services (Jacqueline Trolley and Jill O'Neil)
• The Creation of the SCI (Paul Wouters)

Session IV: Information Retrieval in Science: The Professional Aspects

• Wilhelm Ostwald, the "Bruecker" (Bridge), and Connections to Other Bibliographic Activities at the Beginning of the 20th Century (Thomas Hapke)
• Ralph Shaw: Librarian among Documentalists (Jana Varlejs)
• Science in Crisis: The Conflict over Information Retrieval, 1945-1963 (Mark D. Bowles)

Described is the design and historical development of an information retrieval system for company-generated scientific and technical reprots. The system is based on coordinate indexing, optical coincidence, and tri-axial (three-dimensional) coding. A vocabulary having a maximum capacity of one million terms can be manipulated using only 300 optical coincidence (Termatrex) cards. The system utilizes descriptors for total recall and descriptive phrasing to answer highly specific queries. A tri-axial coding scheme is incorporated to reduce the mechanical noise, present in densely packed systems, to a negligibly low value. Use is made of a mathematical model to predict first approximations of false drops due to mechanical noise.

Citation indexing was developed in the late 1950's as a new way of monitoring, organizing, and retrieving the published scientific literature. The Science Citation Index was one of the initial offerings of machine generated indexing, and it has sustained its premier position as an essential tool for the scientific community over the course of 40 years. This presentation will outline the history of this revolutionary tool as well as establish the appropriate context for secondary publishers in the past as well as what may be expected in the future.

Historically, bibliographic databases came into being for three specific purposes: (1) to assist the researcher in remaining current in his/her respective field of interest, (2) to assist the researcher in selecting relevant material from an ever widening body of published literature, and (3) to assist the researcher in filtering out the poor quality material in favor of the significant high quality publications.

Now technology and theory have made possible a total information solution that enables today's users to navigate and manage digital content well into the 21st century. Web-based bibliographic tools allow users to elicit relevant and current information from a wide array of electronic full text publications. As a navigational tool, these databases provide a key function that remains central to the research community.

"Research scientists will soon be consulting a more precise and specific literature index that links together subject material that would never be collated by usual indexing systems. Concerned with new starting points for scientific literature searches, the unique concept uncovers sometime-buried associations, relating important works and authors, yet keeps the researcher abreast of the masses of current published scientific information. This new approach to information retrieval is called the 'Citation Index'."

In these words, the Institute for Scientific Information (ISI), based in Philadelphia, USA, announced what would become the Science Citation Index in a press release in 1962. Two years later the first issue of the SCI was published. The SCI has influenced science but in different ways than its creators expected. It has in no small measure contributed to a shift in the way research performance is being assessed. A new specialty, scientometrics, has developed predominantly on account of SCI's database.

This paper sketches in detail how the SCI was created. It is based on archival research in the personal archive of Eugene Garfield, as well as at NSF and NIH, and on interviews with main actors involved. The paper comprises the second chapter of my doctoral thesis "The Citation Culture." Its central theme is the way the SCI has given birth to a completely novel representation of science differing from the traditional one, thereby altering some crucial features (but not all) of science evaluation and science policy.

This paper traces the origins of the citation indexing concept. It is well-known that the legal tradition and not the scientific context provided Eugene Garfield with the idea of making a citation index. Precisely how he and his co-builders, Joshua Lederberg, Gordon Allen, and Irving Sher, translated the juridical citation concept into one for the scientific environment has not, however, yet been analyzed. Yet, these early design decisions did create the main features of the SCI that are still largely responsible for its role in science communication and assessment. The paper follows Eugene Garfield, a prolific writer of letters, from the early fifties until 1965. It shows how the concept of citation indexing slowly developed to its final form. It retells the often frustrating encounters with the National Science Foundation and the National Institutes of Health. It shows how the citation concept is indebted to the concept of "information" and the early years of information science. The paper also proposes some explanation about why the reception of the SCI by the scientific community was so cool, although it had been created with the information needs of the scientists in mind.

Top of the page

This paper gives a summary of the activities of the German chemist and Nobel laureate Wilhelm Ostwald (1853-1932) in the area of scholarly information, communication, and publication at the beginning of the 20th century. In 1911 Ostwald, with others, founded the "Bruecke" (Bridge), an organization with similar aims as the famous Institut International de Bibliographie in Brussels. A look is also taken at connections to further institutions or individuals in the area of documentation and "information science," especially in Germany. These are, for example, the "Institut fuer Techno-Bibliographie" or the German librarian Julius Hanauer, one of the German promoters of the Universal Decimal Classification.

Developed by Vannevar Bush in the 1930s, the Rapid Selector represented an early attempt to automate document retrieval, using photoelectric cells, microfilm, and high-speed photography. It was not until the late 1940s, however, that a librarian undertook to adapt the machine for use in producing a major bibliographic tool, the Bibliography of Agriculture. As director of the library of the U. S. Department of Agriculture and member of the American Documentation Institute, Ralph Shaw understood the need for providing timely access to the burgeoning literature for a widely dispersed scientific and technical community. The Rapid Selector looked like a solution to the problem, but turned out to be a serious disappointment. Shaw's experience with the Selector had an impact on his thinking about automation, and about the differences between librarians and documentalists (now information scientists). Those views were widely known and stirred debate for two decades beyond his experiment with the Selector, and may still resonate today.

This paper will explore the early professionalization conflict between documentalists and librarians. The source for this conflict was the widespread concern of what was known as the "information crisis." This was the concern, initiated by Vannevar Bush and spreading to every professional discipline, that information was growing exponentially. The fear was that all disciplines, particularly science, would become "flooded" by the published literature and as a result fail to advance.

In response to this crisis emerged a relatively new professional group called the documentalist. These were typically men who trained in the sciences (often chemistry) and later in their careers turned their interests towards the problems of information and communication. Through their use of new computer technologies they believed they could solve the scientific information crisis by setting up specialized information centers. But in so doing, they attacked and invaded the space of another professional group who the documentalists believed were ignoring the problem. This group was the librarians. The conflict emerged in part from the opposition between the makeup of a documentalist (scientist, male, computer, information center) and a librarian (humanist, female, card catalog, library). It was the culturally privileged attributes associated with the documentalist which provided them the authority and power to attempt to infringe upon the librarians professional space.

As a case study to explore these themes I will examine what was known as the "world's most advanced information retrieval system" - the Center for Documentation and Communication Research housed at Western Reserve University (now Case Western Reserve University).

This paper will look at information handling in various scientific disciplines and describe the crisis in scientific communication. It will assess the significance of information conferences and argue that the 1956 Practical Utilization of Recorded Knowledge conference (attended by over 600 people from 400 different institutions) was the first major United States gathering to specifically address these issues. This paper will also explore the contributions of specific individuals. Allen Kent was a director of the Center (as well as a chemist and a documentalist) whose early significance in this field is often overlooked. A recent interview with him as well as his collected papers provides a window into this conflict.

Top of the page