Submit Comment

show all (0)
There are no comments. Click the text to your left to make a new comment.
1 0

The momentum of wartime research and development activities led to continuing exploration of new concepts in science and technology. The research in electronics, computers and cryptography, and the general interest in the organization and use of scientific information brought about a climate of inquiry concerning the nature of information and communication that led to the beginning of several new branches of what Fritz Machlup later termed information disciplines.1 This was also a time for introspection, a time for practitioners to reexamine the essence of their professions and, in the midst of global concerns, to probe their microcosm, a narrower and more private world.

2 0

DOCUMENTATION IN THE UNITED STATES

3 0

An engaging report from F. Donker Duyvis, secretary general of FID, describes the U.S. documentation scene as seen in the summer of 1946 through the eyes of a European documentalist. Duyvis visited a variety of libraries and institutions and met with publishers, groups from the United Nations, members of professional societies, and representatives of government offices.2

4 0

Duyvis was taken by the philosophy of personal service underlying library and archival work in the United States and by the well-organized technical and institutional support of libraries. Albeit documentation, “in the sense of collecting, organizing and distributing information,” was unfamiliar to Americans, Duyvis noted, they were “more documentally minded than their European colleagues.” American librarians and archivists were service oriented in that they considered the basic idea of documentation — to make print and non-print

New Winds Blowing 124

information available to those who need it — a “self evident task.”

5 0

Duyvis was impressed with the detailed catalogs of the Library of Congress and other major research libraries, and the tremendous amount of bibliographic work carried out in special libraries. Because the time lag between publication and citation in secondary publication was so great, librarians in special collections were preparing “home made bibliographic cards” and newsletters, mentioned in chapter 6, which entailed considerable duplication of effort, Duyvis noted.

6 0

Overall, the United States was ahead in coordinating documentation: the Library of Congress cards brought about uniform shelf arrangement throughout the country, and through the National Union Catalog the library could determine where a publication could be found in 66 percent of the cases. In addition, there were Union Catalogs in Philadelphia, Denver, and Seattle and many smaller union catalogs and union lists of periodicals throughout the country and in academic institutions to serve users. Special librarians had pooled their bibliographic efforts in many cases, and the Special Libraries Association (SLA) promoted and coordinated information regionally and within specialties.

7 0

The “scientific management” of the Library of Congress, the USDA library, and the National Archives awed Duyvis. The libraries were run with the precision of commercial enterprises and were “as cost-price minded as any modern factory”; their managers had operational cost figures at their fingertips, like “planners in the best industrial works.”

8 0

While European libraries had smaller collections and tried “to get the most out of the available material,” even if it called for “enlargement of card systems and the like,” because of the immense flow of paper in the United States the trend was to develop “documentary methods and equipment.” The philosophy of service also differed: U.S. libraries prepared fewer elaborate literature reports and bibliographies for individuals than those in Europe. But then Americans — from directors of big scientific libraries to staff assistants in small libraries — considered their job not only “to preserve and loan documentary materials,” but also to educate the users. In the United States, reference services were consistently good and the reference collections large and well organized so that patrons could find their own materials easily — although special libraries did conduct searches for the staff of the organizations they served.

9 0

Carl Milam of the ALA told Duyvis that more individualized documentation services were needed. Later that year the ALA and several library associations carried out a market study for setting up bibliographical search services in libraries for which users would pay

10 0

New Winds Blowing 125

11 0

two dollars per hour. The study indicated a clear need for such services at the Boston Public Library, the New York Public Library, and the Library of Congress, but not a large enough demand in other places to support such a service.3

12 0

Duyvis was enthusiastic about the National Archives, which, in contrast to European institutions, controlled the tremendous mass of material entrusted to it and “was a living institution serving the public.” He was reminded here that documentation extended beyond librarianship to “other professions such as. . . the archivist, or the office filer or the editor of a scientific periodical or the bibliographer or the patent specialists and to a certain extent every scientific worker.”

13 0

Duyvis also noted that American librarians were typically better educated than their European counterparts but was surprised that no documentation courses were offered in the United States. In the Netherlands, which had no library schools, his organization, the Netherlands Institute for Documentation and Registration (NIDER), offered several courses each year in documentation, classification, cataloging and alphabetizing, and copyright law. Some “non-coordinated documentary study” was going on in the United States, “practically outside the world of librarianship”: a committee of the American Chemical Society (ACS) was studying “the problems of the punch card machine” for bibliographic work, the Patent Office studied complicated classification problems and bibliographic questions, and the American Management Association studied documentation problems of current archives.

14 0

Duyvis sought to develop greater coordination between the Dewey Decimal Classification (DC) and the Universal Decimal Classification (UDC), but during his American tour he realized that this was unlikely because Europeans and Americans used classification for different purposes. Most U.S. libraries allowed open access to their collections and used DC to provide systematic arrangements of books; they assigned one classification number — which would be noted on a book’s spine — and several alphabetical subject headings to a work because Americans preferred dictionary catalogs. European libraries, with their closed stacks, preferred classified catalogs. Europeans used UDC to encode complex subject analyses of books. They assigned classification numbers for each element contained in the work, and patrons would search by classification number to gain subject access.

15 0

While in the United States, Duyvis did visit Science Service: he commented on Davis’s “impressive work” on dissemination of science and documentation, and wrote laconically that the work of the ADI for the time being was limited “to certain applications of microfilm.”

16 0

New Winds Blowing 126

17 0

INTERNATIONAL INTERESTS

18 0

Duyvis’s trip to the United States was one of many that scientists, scholars, and managers of scholarly organizations undertook in the postwar period when scientific and scholarly interchanges were encouraged by national governments.

19 0

The terror and trauma of World War II led to the organization of the United Nations (UN) to ensure peace by providing a forum where disagreements among nations might be settled. Among the supranational governmental organizations set up under UN auspices was UNESCO, the United Nations Educational, Scientific and Cultural Organization, “dedicated to the diffusion of knowledge to effect better mutual international understanding.” UNESCO’s charge included a mandate, to assure dissemination of scholarly, scientific, and technical information throughout the world. The concept Davis had advocated since the 1930s, the worldwide bibliographic control and dissemination of information to research workers wherever they might be, was now embraced officially by representatives of numerous governments.

20 0

As President Roosevelt presented the United Nations charter to the nation “as almost holy writ,” librarians were part of the elite of the country that did not want Roosevelt to fail as President Wilson had in bringing the United States into the League of Nations after World War I. Roosevelt was so successful in building a domestic consensus that those well versed in foreign affairs, such as Dean Acheson and George Marshall, were concerned that the high hopes raised by the UN charter were bound to lead to bitter disappointment.4 The concerns were well founded. The structure of the UN soon started to creak and, as it was reaching its fourth decade, serious, perhaps irreparable, cracks appeared in the organization.

21 0

In 1945, the foremost American librarians who, during the war, had experienced the toil and exhilaration of working on projects vital to the nation, wanted to rise to the challenge they felt had been presented to them. The U.S. government had given “unprecedented recognition” to the “essential nature of scientific literature,” and now, “for the first time on record,” heads of governments recognized the importance of scientific information and advocated exchange of scientific publications, wrote Robert B. Downs of the University of Illinois and Keyes Metcalf of Harvard University. They believed that librarians had an opportunity to develop programs, on a scale never hitherto considered possible, for the interchange of scientific literature.5

22 0

Several librarians were involved in UNESCO from the onset. Immediately after the war, Archibald MacLeish, assistant secretary of state and former librarian of Congress, began drafting the constitution of

23 0

New Winds Blowing 127

24 0

UNESCO and drew in Luther Evans, the current librarian who, in 1953, became UNESCO’s secretary general. MacLeish and Evans, as members of UNESCO’s executive committee, also participated in developing its program. Luther Evans, Ralph A. Beals of the New York Public Library, and Ralph A. Ulvelling, representing the ALA as its president, were members of the commission constituted to advise the permanent delegate to UNESCO. Carl Milam, executive secretary of ALA, and Waldo Leland, director of ACLS, served as consultants to the organization.

25 0

ALA’s leading members and its executive secretary, if not the entire association, were caught up in the movement to create world peace. In a generous spirit American librarians were eager to help restore library services, share their know-how of library management, and bring democracy to newly established nations. Keyes Metcalf, Florence Ludington of the Mount Holyoke College library, Milton Lord of the Boston Public Library, and Leon Carnovsky of the Graduate Library School, University of Chicago, were among those who provided such assistance and were instrumental in creating an atmosphere that promoted international relations.

26 0

The elite of the library and wider documentation community were eager to participate in UNESCO. They were also aware that with its broad mission, UNESCO first had to determine how to carry out its tasks: encouraging development of national bibliographies, coordinating international bibliographic cooperation, establishing international standards, and developing means to ensure the free flow of scientific, educational, and cultural materials across national borders.

27 0

CONFERENCES ON SCIENTIFIC INFORMATION AND BIBLIOGRAPHY

28 0

The first major professional and political event for scientific information and bibliography took place in London. The Royal Society Empire Scientific Conference, held from June 17 to July 8, 1946, brought together delegates from all parts of the empire to review the organization of science and issues of consequence to scientific work in Great Britain and the Dominions.6 To underscore the importance of this conference, the king and queen attended the opening ceremonies. A part of this meeting was devoted to discussions on the structure of organizations and a review of the tools for handling scientific information.

29 0

The Royal Society, following up on the recommendations of the Empire Scientific Conference, held a Scientific Information Conference June 21 to July 2, 1948. The conference was restricted to

30 0

New Winds Blowing 128

31 0

considering subjects only “of use and service to the scientific community.”7 Among the observers of this conference were Ralph Shaw of the USDA library and Mortimer Taube, who represented the Library of Congress.

32 0

The conference was significant because it provided a thorough and objective overview of institutions, traditional tools, and new mechanisms used in scientific communication by experienced practitioners who have been trained as scientists, engineers, or librarians — although substantially all of the subjects had been discussed by scholars and librarians over the years. The conference also considered the training of special librarians and information officers and, importantly, the problem of scientists in isolated places, who might not have immediate access to extensive library facilities; both were matters of concern also to the fledgling UNESCO.

33 0

Curiously, Davis’s idea of a Scientific Information Institute, first described in 1933, reappeared in a conference paper by J.D. Bernal. Bernal proposed that scientific papers be submitted to, and distributed by a central clearinghouse, with short versions published in specialty journals and full papers made available to scientists on request or by subscription. This was an extension of an earlier paper Bernal had prepared for the Empire Scientific Conference, in which he made no reference to Davis, although Davis’s SIT proposal was published in its entirety in Bernal’s book on the Social Functions of Science.8 Bernal’s paper generated so much newspaper publicity before the conference started and met with such violent opposition that he withdrew it from consideration during the conference.

34 0

Among the new inventions discussed at the conference were the Shaw-Bush rapid selector, based on the idea of Watson and Helen Davis. UNIVAC, the first commercial computer, not yet suitable for information retrieval because of its limited memory, was noted among the new “indexing devices.”

35 0

Exhibits at the London conference were diverse: scientific journals and abstracts, printing methods and reproduction, microfilm and microfilm reading machines, classification in science, and mechanical indexing and sorting. The latest developments of non conventional information retrieval methods were displayed. Among them were mechanisms for the retrieval of encoded information with Hollerith-type punched cards; an exhibit by G.M. Dyson, who later moved to the United States to work with Chemical Abstracts, demonstrated how information on organic compounds could be retrieved from edge-notched cards by using linear codes of the compounds. Calvin Mooers’s Zatocoding system, which used superimposed random codes to improve information retrieval from a file of edge-notched cards, was another U.S. development discussed at the conference.

36 0

New Winds Blowing 129

37 0

Final recommendations of the conference covered twenty areas and touched on all aspects of scientific communication. One of the recommendations demonstrates how the wartime experience changed attitudes about science information: the conference recommended that the scientific information specialists be “equal in standing to fellow scientists employed in research, industry and administration, and should receive comparable treatment in training facilities, and emoluments.”9

38 0

In contrast with the Scientific Information Conference that systematically explored the state of the art in scientific communication, three UNESCO conferences held in 1949 and 1950 were charting new paths.10

39 0

Two of these conferences were concerned about scientific abstracting and indexing services throughout the world. The participants discussed cooperation among abstracting and indexing services in science and in medicine and agreed that abstracts be prepared for all papers containing new scientific and technical information and that all scientists in all countries should have adequate access to these abstracts for current and retrospective searches.11

40 0

The third conference, sponsored by UNESCO’s libraries section, reviewed the global survey of bibliographic activity prepared by the Library of Congress and the reports and comments on the survey by national groups. After its deliberations the participants recommended that national organizations be set up to promote and coordinate a rational pattern of bibliographic services.12

41 0

These conferences manifested remarkable optimism, considering that they took place after a worldwide conflagration and in the midst of a Cold War, even after the outbreak of the Korean War in June 1950. But for a time international cooperation in bibliographic control and distribution of scientific and scholarly information was able to override divisive nationalism and other problems that thwart participation in international organizations.

42 0

One such problem is lack of funds. In this postwar period, however, U.S. philanthropic foundations supported numerous international conferences. Foreign librarians and documentalists could visit the United States and Americans could participate in international activities and take on leading roles in international organizations because the Rockefeller Foundation and the Carnegie Corporation provided the necessary funds.13

43 0

MECHANIZATION AND REEXAMINATION OF FUNDAMENTALS

Immediately after the war and in the decades that followed, issues related to scientific information and communication were the prime

New Winds Blowing 130

concern of U.S. information workers. Because the ADI was mostly affected by the developments in scientific and technical information, the problems in other areas of scholarship and research are not discussed in this book. It is well to remember, however, that the needs of scholars in the humanities did not diminish. Moreover, with major changes taking place in the social sciences and with a tremendous growth in the volume of social science data, the management and control of information in those disciplines also increased.

44 0

By the 1930s, as government programs expanded so did the bulk and variety of collected data. For example, data on social security, housing construction, and farm credit were collected, contributing to the body of knowledge in the social science fields. In addition, new data-dependent indicators were created — such as the gross national product (GNP) or national income — to provide planning guidelines, which began to be used in economic and political discourse.

45 0

After the war local data collection increased worldwide as global indexes and constructs to evaluate the economy also became of concern. Quantitative techniques became more sophisticated and were used more frequently in economics, sociology, public administration, market forecasting, and in many other areas:14

46 0

Thus changes in approach to information and document handling and the use of new technologies, important to the natural sciences, became just as important for the social sciences.

47 0

In the postwar period library and documentation activities in the country escalated. New libraries and information centers were created, many special libraries were established; and several large libraries in the country expanded, setting up special science and technology divisions.

48 0

One of the newly created agencies was the Chemical-Biological Coordination Center set up by the National Academy of Sciences-National Research Council. Based on wartime work on anti malarial agents, fungicides, and pesticides, the center attempted to correlate for the first time molecular structures and biological or chemical activity by extracting information from large punched-card files. The center’s activities stimulated new approaches to, and familiarized many with the challenges of, non conventional information work and had a wide impact on the practitioners in the science-information community:15

49 0

Some scientists and engineers had also become interested in making better use of the available information or retrieving more effectively the information they had collected in their own files.16 The various systems were inefficient and costly because interactions were based “on individual needs of particular groups rather than upon the needs of the community as a whole,” Verner Clapp decried. There was

50 0

New Winds Blowing 131

51 0

simultaneous activity in many groups but little collaboration or cooperation among them, resulting in duplication of effort on the one hand and a lack of coverage on the other: “failures in comprehensiveness matched by failures of selectiveness.”17

52 0

This laissez-faire system led to frustrations and confusion as well as increasing professional self-castigation. Despite intensive effort and hard work, in the absence of guidelines based on relevant experience, so much uncertainty remained that those responsible for managing information centers had to hew their paths guided by a combination of tradition and intuition.

53 0

The new information and documentation centers replaced standard library operations with industrial production methods. ADRC in London, and later CADO, applied such methods, just as Evans had done during the depression when he used unskilled WPA workers to register documents in country archives. In the post-World War II period, CADO devised general rules clerks could follow and broke tasks into simple components to speed processing. This technique made it possible for untrained people to catalog documents or perform other complex tasks. The same technique was used intensively when computers were introduced into libraries.

54 0

In the late 1940s, professional librarians oversaw staff, established subject headings, edited the lists printed out from keypunched card entries that had been sorted by IBM machines, and advised on matters of professional interest. The major information centers found it necessary to use automated equipment for the management of bibliographic control.

55 0

Punched cards for hand and machine sorting had been in the process of development since the eighteenth century when Joseph Jacquard devised control cards to store the necessary information for looms to reproduce patterns for woven fabrics. Charles Babbage’s analytical engine using punched cards was invented about 1840, and later in the century, Hermann Hollerith devised the punched card equipment used for the 1890 U.S. census. After mergers and name changes, IBM became the leading manufacturer of this equipment, although in the 1940s several companies moved into this field. Edge-notched cards for hand sorting — using a long knitting needle, ice pick, or similar device — had first been proposed after the turn of the century.18

56 0

Holes placed at appropriate places on the cards provided the means of encoding information that could be searched automatically. The Hollerith or IBM cards had 80 x 12 positions that could be encoded, with numbers and other specially designed codes identified by one punch at the appropriate place and letters by two punches in a column. Machines then could read the code and print out the information on

New Winds Blowing 132

the card. Hand-sorted, edge-punched cards provided a place for text, like ordinary file cards, and also offered greater flexibility in coding but were not suitable for large collections because mechanical handling was not practical. Several other types of cards with punches were also used for a while.

57 0

The principles for searching and encoding were similar for all types of cards. Specific categories of information, such as author, subject, and accession number, were assigned to certain areas or fields on the card. Thus, batches of cards could be alphabetized or ordered according to any of the encoded concepts, such as subject, year of publication, or chemical substructure.

58 0

In the 1930s a few libraries had automated some of their information work. The Boston Public Library used punched cards for analyzing acquisitions in 1934. Later, the Library of Congress had its personnel records on punched cards; by sorting cards mechanically, lists could be prepared on short notice of staff members serving in the National Guard or eligible for the draft.19

59 0

During the war a number of defense installations used punched cards. Some scientists and engineers were in positions where they had to devise machine-assisted information retrieval systems20 and either learned library and indexing techniques rapidly or, in many cases, reinvented them. Several librarians gained experience using the equipment for subject analysis: Joseph Becker, for instance, used punched cards to retrieve enemy ordinance data in the army, while Jesse H. Shera and others experimented with “primitive” techniques for indexing censorship intercepts of mail from abroad at the Central Intelligence Division of the Office of Strategic Services (OSS).21 This work led them to consider using punched card techniques for bibliographic work.

60 0

Computers, however, were not yet used in information work. The early computers, built in the 1940s, were suitable for “crunching numbers” but did not have large enough memories to maintain bibliographies or to retrieve information from large collections. In the United States mechanical calculators had already been used before the war to calculate the trajectories of a grenade, but it took twelve hours to determine just one trajectory. During the war the Bell Laboratories developed a much more dependable machine using standard telephone relays. Model III, finished in 1944, had nine thousand switches, weighed ten tons, and occupied one hundred square yards; it could calculate a trajectory in forty-five minutes.

61 0

The University of Pennsylvania continued the work on the ballistics problem, but its first truly functional large-scale computer, ENIAC, was finished after the war. It could calculate a thousand times faster than electromechanical devices but it required 18,000 electronic tubes,

New Winds Blowing 133

weighed thirty tons, and consumed 150,000 kilowatts of electricity, and because of its size and the heat it generated, it had a number of technical problems. ENIAC was also an analog computer, using ten digits, 0 to 9, which seemed more practical at the time but proved to be more error prone than the binary computers adopted later. Binary computers used just two numbers, 1 and 0, which could be translated to simple on/off switch positions. Up to that point, base two systems had been considered too theoretical and artificial. Binary computers were based on Alan M. Turing’s work in the mid-1930s. Turing believed that any information could be represented by zeroes and ones and proved his concept of a program-controlled computer with a paper strip containing only zeros and ones yet could carry out instructions. During the war Turing was a member of the British team that used his principle to crack the Enigma code of the encryption machine that automatically scrambled the messages the German high command sent to the troops.

62 0

Vacuum tubes continued to be used for electronic work. Hardly anyone was aware of the major breakthrough on December 23, 1947, when William Shockley and John Bardain, the two theorists, and Walter Brattain, the major creator, demonstrated their transistor at Bell Laboratories. This tiny device can switch current on and off as it moves over a controlled path through a solid block of semiconductor material. The transistor, which was faster and more reliable than vacuum tubes, made the development of the next generation of computers possible.22

63 0

Bibliographies and Information Retrieval

64 0

When technical reports inundated the document processing centers in 1945, each major agency attempted to automate some of the bibliographic work and to provide mechanized search tools with the aid of punched cards and modified sorting and tabulating equipment.23 Research libraries and several special libraries also initiated projects that applied mechanical devices. Many were inspired by Vannevar Bush’s 1945 article, “As We May Think,” and his concept of “memex,” a computer with associative capabilities.24 Technology advanced to enable man to reach the moon in 1969 and to receive information from planets at the edge of our galaxy, the technology of the late 1980s has not fully succeeded in creating a computer with the associative capabilities Bush envisioned. Through the 1950s, computers were so limited that libraries still had to hobble along with cumbersome punched card equipment to mechanize large-scale bibliographic work.

65 0

New Winds Blowing 134

66 0

Ralph Shaw’s rapid selector, one of the new tools for information retrieval mentioned at the Scientific Information Conference in London, realized Helen and Watson Davis’s concept for automating retrieval. Photographing abstracts alongside subject codes, Shaw carried out searches on large reels of film with an adapted rapid selector Bush developed at MIT before the war. Strangely, Shaw neither acknowledged the Davis’s contribution nor mentioned at what stage he took over the development of the rapid selector. Despite high hopes for its success, the rapid selector never functioned properly. Few understood at the time that even if mechanical problems could have been corrected, the machine could not have fulfilled its promise because Shaw had not grappled with the fundamental problems of indexing, so critical for information retrieval.25

67 0

Punched cards offered a number of advantages but also had their serious drawbacks. Once the processes were set up, keypunching and sorting could be carried out by clerical workers. Cards containing citations constituted a ready bibliography that could be manipulated and updated, while printed indexes, like Chemical Abstracts, could be updated only by cutting and pasting. These non conventional bibliographies could be rearranged by mechanical sorting, and coordinated bibliographic searches could be carried out by sorting on any of the encoded fields.26

68 0

But automated data processing was cumbersome. The mechanics of sorting could engender numerous problems, some caused by the physical condition of the cards. Just spilling a large batch of cards could mean additional hours of sorting. The printouts — in all capital letters — were unattractive; the large-size paper required was a drawback for catalogs. The need for more space was another. Punched cards were larger than traditional Library of Congress cards but could only hold eighty characters of information at most; thus a batch of punched cards would take at least six and one-half times the space required for regular catalog cards. Mechanical sorting was a great advantage, but the entire batch had to be put through the sorter for each number and twice for each letter in a field. Alphabetizing fast-growing collections was a time-consuming process.

69 0

Mechanization led to better cost control and measurement of efficiency. The advantages of using mechanical aides, coupled with the rigidity of operations they imposed, stimulated reexamination of the basic concepts of library practice. Librarians were re-evaluating how to “mark and park” documents,27 as were a number of scientists and engineers caught up in the challenges of mechanical information retrieval but unfamiliar with, and basically disinterested in, library techniques.

70 0

New Winds Blowing 135

71 0

Subject Headings

72 0

As mentioned earlier, speed of processing technical reports had the highest priority in the immediate postwar period. Precise subject access allowing effective retrieval of materials was critical for users, and this led to analyses of how best to represent information. Even a simple punched card could represent different kinds of information: it could be a citation, leading to the work the way a traditional catalog card would; it could either contain or lead to a surrogate for the document. A punched card could also present data directly, which, however, could be linked through access codes to the appropriate citations that could lead to the documents holding the data.

73 0

Punched cards made it possible to find information that could not have been retrieved before. For example, Claire K. Schultz, one of the early and successful users of punched cards for information storage and retrieval, recalled that in the 1940s a scientist could not retrieve papers on immune reactions from the Sharpe and Dome collection because immunity was only a subheading and not a subject heading used by the library. To find the citations someone would have had to look up various diseases and organisms and check whether immunity appeared as a subheading.28

74 0

But even with punched cards and large collections with many citations listed under each subject heading, such coordinated searches were excessively time-consuming. A major Army Medical Library study, carried out under Stanford V. Larkey at the Welch Medical Library of the Johns Hopkins University, showed that even with the most advanced automatic data-processing equipment of the early 1950s, such subject searches were uneconomical. This study of subject heading problems in the medical literature was started in 1948, and, even before it was completed, had won praise for its thorough methodology and for its improved sorting equipment, devised by Eugene Garfield.29 Such studies were important to the mechanization projects being carried out in various government agencies, in private industry, and in other organizations under government contract.

75 0

As mechanization gathered momentum, a dichotomy of interests between the managers responsible for operations and the “less encumbered” members of the information community soon became apparent. The former were most concerned about bibliographic operations, while the latter could range more freely in information retrieval research.30

76 0

Given the wartime and postwar involvement of scientists and engineers in information work, it is astonishing that coding and the terms used for retrieval — by whatever name — and their use for different kinds of collections were not approached more systematically.

77 0

New Winds Blowing 136

78 0

Substantive research in librarianship had only begun two decades earlier with the foundation of the Graduate Library School at the University of Chicago.31 But scientists and engineers, working in their own fields, tested hypotheses, analyzed data, and only moved to large-scale production after exploring uncertainties in pilot projects of gradually increasing size. Yet in information work, their approach to problems did not differ greatly from the more pragmatic librarians. Scientists and engineers may have understood better than librarians the underlying structure of science and technology and the manner in which their technically trained colleagues use the scientific literature. In information work, however, they did not use the proven approaches of their disciplines.

79 0

Librarians and the new breed of science information specialists also differed in their approach to subject headings. Librarians, “signaling through time” to an unknown user in the future, would look at a work, decide what it was about and select an appropriate subject heading.32 Selecting index terms — combined with encoding for automated data processing — that would lead specialists to the information they needed, however, required another point of view.

80 0

To retrieve the documents relevant to users’ queries, coordinated searches had advantages. But mechanized equipment required searching character by character, thus to reduce time, effort, and expense, codes instead of words or phrases were put into use. Now, in addition to selection of the most fitting subject headings, the question of how best to encode subjects had to be considered. In the 1940s Mortimer Taube, Calvin Mooers, and James Perry were the leaders among Americans who began to investigate subject analysis, the coding of subject terms, and their relationship to information retrieval. At the frontiers in a paradigm-breaking era, their approach combined a pragmatic view and abstract thinking.

81 0

The ideas of these men were hotly debated at ADI meetings in the 1950s. They discussed subject headings (or index terms) that allowed more direct intellectual access to collections; they pointed out that existing classifications had gaps and were also limited because they could express only hierarchical relationships. Because the new approaches were such a departure from the traditional patterns of thinking, the underlying issues were revealed only over time, after much thought and highly charged debates. Librarians and information specialists were preoccupied with operations and adjusting to equipment, and were concentrating on improved techniques for improved information retrieval. Thus it should not be surprising that it took long even for the leading information professionals to realize that no classification scheme, thesaurus, or indexing vocabulary could be unbiased and that even scientific classifications and subject terms reflect the

New Winds Blowing 137

philosophy and the general and social world views of those assigning them.

82 0

Searching for Fundamentals

83 0

S. C. Bradford, the former keeper of the Science Library in London, saw a bold role for documentation in this postwar period. He believed that efficient application of scientific knowledge would produce affluence and that a key “to our emergence from our present troubles” was documentation.33 But documentation itself was also expanding. The war had created new problems and intensified already troubling problems in the accumulation, organization, storage, and communication of knowledge, yet it had also “provided techniques, equipment, and processes that were formerly unknown or in rudimentary stages of development.”34 Thus, in the late 1940s, wartime developments in computer research brought forth new tools for handling information. The early computers that could not yet be used for bibliographic information retrieval from even modest collections still held out promise for the future. Academics explored the epistemology of information and communication — of concern to several disciplines — and brought about the birth of new fields of scholarly and scientific study that Machlup and Mansfield would trace back to this period.

84 0

Two seminal works were published in 1948. One was Robert Wiener’s Cybernetics, or Control of Communication in the Animal and the Machine, a comparative study of human, mechanical, and electronic control systems, based on his wartime studies.35 The second was Claude E. Shannon’s information (or communication) theory. Shannon’s unified theory of signal transmission considered the transmission of information as a statistical phenomenon. Based on Ralph Harley’s work of the 1920s on the transmission of information and on some of Wiener’s wartime research, Shannon’s theory provided communication engineers with a long-needed tool with which to determine the capacity of a communication channel. The concept was appealing and was taken up in a short time by scientists working in a number of disciplines. Shannon himself, however, considered the applications of his work to problems outside communication theory to be suspect and did not attach fundamental significance to them.36

85 0

When Shannon’s information theory was published, it did not relate to everyday operations of librarians or technical information officers. Up to this point there had been little criticism of standard techniques of classifying and indexing information; those engaged in using new technology were primarily interested in automating, not in conceptual changes. But the publication of Shannon’s theory proved to be most

New Winds Blowing 138

stimulating to the scientific information workers who wanted to penetrate below the surface of current library and bibliographic techniques and to discover principles that might help improve retrieval.

86 0

These scientific information workers were educated in the belief of the superiority of the scientific approach established in the seventeenth century. In explicit opposition to both ordinary experience and speculative philosophy, the scientific approach was based on measurement, experimental evidence, theory building, and a procedure of criticism and testing.37 The theorists generalized their observations, derived laws explaining processes and complex phenomena in science or science-based technology, and, in imitation of theoretical physics, expressed them through symbolic mathematics.

87 0

Much has been said about the importance of science and technology in winning the war. But some of the wartime and postwar discoveries had broader implications. Observation of an apparently insignificant incident led to the discovery of penicillin, the first true antibiotic drug. New antibiotics could penetrate bacteria and conquer infectious diseases that a short while ago had caused deaths. Blood that had to be transfused whole in the past could now be transformed into a dry powder through new techniques and could save lives at the front. Atoms were split. Within a few years, scientists discovered that molecules of nucleic acids were able to transform pneumococci; mere chemical compounds could influence the genetic makeup of living matter. In physics and in biology scientists had been able to pierce the smallest known constituents that hitherto seemed impenetrable. Wishing to discover fundamental laws about information, to find codes to capture the essence of words and of documents, was very much in the spirit of the times.

88 0

To some information workers the epiphany of Shannon’s theory held out hope for the early development of a theoretical foundation for information retrieval. But this was unrealistic. First, Shannon’s theory, limited to communication — or rather signal transmission — is not concerned with the content of information or the message itself and, as some have pointed out, should not have been called information theory.

89 0

Second, those involved with the organization and retrieval of information were concerned that no one understood exactly how people grasp information and were uncertain how to extract the essence of documents or provide the clues to assure that the information or document useful for their clientele could be retrieved from a collection. Expressing in mathematical formulas a relationship of barely identified factors and generalizing the complex problems related to all aspects of information retrieval — without rigorous experimentation — had its appeal but was bound to lead to disappointment. For the

New Winds Blowing 139

documentalists, Shannon’s information theory with its limited application turned out to be a mirage, although Shera and Cleveland believed that it brought some abstract thinkers into the field, which now was moving beyond documentation toward what was later called information science.38

90 0

As they were trying to penetrate the unknown, these information workers were constrained by the tunnel vision inflicted not as much by their training but by the climate of the times. Not until the late 1970s or even 1980s did they more fully realize how much social and behavioral factors affect knowledge transfer.

91 0

Taube, Mooers, and Perry — Uniterms, Descriptors, and Semantic Coding

92 0

As automatic devices were introduced, library practices were gradually modified. Mortimer Taube had already become interested in simplifying cataloging practices when he was with the navy project at the Library of Congress, and he ran into difficulties when he wanted to make changes in the library’s practices. His later experience at the AEC reinforced his views and led him to a radical revision in his effort to simplify subject headings. Struggling to gain control over technical reports at the AEC, Taube did not want to restrict future searches by limiting choices to a controlled list of headings. He first suggested that subject terms assigned to a document should be words used by the author; he called them uniterms because compound expressions were to be broken up. Each paper — or its code number — containing a specific term was listed on every card headed by a term contained in the paper. The collection of cards, with a listing of papers, constituted an inverted file. In response to a query, several cards would be compared, and the documents containing all the terms could then be identified.

93 0

Taube’s uniterm theory blissfully simplified subject indexing and speeded up document processing. The search techniques that compared accession numbers of works containing the appropriate terms could be easily understood. Taube eventually realized the shortcomings of his system and began modifying his ideas before his sudden death.

94 0

An unrestricted vocabulary had several disadvantages. One related to synonyms: a favorite whimsical example documentalists used in their debates was the problem of combining two uniterms and retrieving material on both Venetian blinds and blind Venetians. Actually, the burden on users was a more serious drawback, since they had to identify all the terms that might be related to a query to achieve fairly

New Winds Blowing 140

complete retrieval from large collections. In the early 1950s Taube founded a company, Documentation, Incorporated, to carry out work based on the uniterm approach.

95 0

Calvin Mooers’s concepts were in opposition to Taube’s philosophy. In Mooers’s view, the proper question to ask was not what a document was about but rather under what circumstances would the user want to retrieve this document. He advocated that only terms used in an organization be assigned to its document collection. A list of descriptors was to be developed jointly by the professional staff using the collections and by library or information center personnel of the organization. This approach reduced the number of subject terms that had to be considered, simplified searches, and improved retrieval of information that was of current interest to the organization.39

96 0

To emphasize the difference between his approach and traditional usage, Mooers introduced the term descriptor, to be used instead of subject heading or index term. Mooers was the first to advocate — albeit indirectly — that defining information needs was the joint responsibility of the users and their information service, which in the 1980s has broadened to encompass all of the information use within an organization. The term descriptor became part of the information science and library vocabulary; but over the years the term lost its specificity to the extent that in a 1970 compilation of terms it is defined as “a word used to categorize or index information; synonymous with key word,” the opposite of what Mooers intended it to be.40 Machine searching and sorting time were closely related to the length of the terms being used; thus coding became a matter of practical as well as theoretical interest. Codes could, but did not have to, reflect the meaning of the term they represented. Mooers advocated assigning random superimposed codes to reduce the possibility of “false drops” and thus assure greater accuracy.41 James Perry, on the other hand, wanted to incorporate all the information possible into his codes. Perry was an experimental chemist who had already done literature work before the war. He believed that one had to get “between the covers” of a report and encode all the information that could be extracted from the report.42

97 0

He stressed that information was polydimensional. The semantic code system he first developed with Madeline Berry (now, Madeline M. Henderson) and later with Allen Kent, was based on the concept that the vocabulary of subject terms should reflect this complexity. For example, thermometer and thermocouple, both heat measuring devices, had a partially overlapping code. Added to Perry’s codes were links and role indicators to express the relationship of the various terms used in a paper. Thus, through the codes, “telegraphic abstracts” could be constructed, which then provided searchable document

New Winds Blowing 141

surrogates.43 Perry’s semantic codes were complicated and often longer than the words they represented. But as a result of his work, thesauri — providing a controlled vocabulary with indications of the conceptual relationship between terms and of synonyms and near synonyms and giving also a grouping or classification of those terms — became useful tools of documentalists and information workers. The system, developed further when they moved to Western Reserve University, also included rules — a superimposed grammar — to help retrieval. Perry, Berry, Kent, and later others at Western Reserve University were trying to provide in their “telegraphic abstracts” a full encoded extract of a paper.

98 0

Machine translation was a new discipline of direct interest to those concerned with information retrieval and scientific communication. Warren Weaver, director of Natural Sciences of the Rockefeller Foundation, broached the idea in 1946 when visiting with A. Donald Booth, who was then working on digital computers in J. D. Bernal’s laboratory in London. Weaver suggested that every language might contain basic elements that could be detected by means of cryptography. In 1949 Weaver circulated a letter to about two hundred of his acquaintances; in the letter he wrote that the “speed, capacity, and logical design flexibility” in electronic computers and the cryptography for decoding enemy messages during the war led him to think that computers might conceivably be used for translation. In simple terms, he explained, one could think of a book written in Chinese as an English book encoded into the “Chinese code.”44

99 0

After a surge of interest, in about two decades machine translation melded into aspects of linguistics, computer linguistics, artificial intelligence, and other disciplines. Nevertheless, machine translation raised important questions and stimulated information workers. James Perry, for instance, who had left chemical research to concentrate on machine methods of information retrieval, worked at MIT in the late 1940s and early 1950s, the time when it became one of the centers of machine translation; he also participated in the first conference dealing with the discipline in 1952. Because of the heightened interest in foreign scientific publications, many hoped that machine translation would have immediate practical applications. As mentioned earlier, an enormous volume of materials in foreign languages was added to U.S. collections immediately after the war. With the normalization of the situation abroad, the proportion of foreign materials, which Americans typically do not read, continued to increase.

100 0

Machine translation, however, had a broader impact on information work. At the time when computers were new and their memories quite small, early research on machine translation demonstrated how computers

101 0

New Winds Blowing 142

102 0

could handle strings of words rather than one word at a time and that programs could be written to parse sentences automatically. This was perhaps as exciting to information workers as it was for microbiologists to see deoxyribonucleic acid (DNA) take on a visible form as it precipitated. As Joseph Becker later wrote, experiments in machine translation had taught us what today are elementary operations: “how to sort letters and words by computer; how to search, match, count and rank words;” or how to locate single words or strings of words. Work on machine translation also taught us how to edit natural language by computer — to correct spelling errors, to add and delete words — and how to merge new records into files or delete old ones.45

103 0

Machine translation also alerted scientific information workers to the importance of semantics and syntax. This work made them aware that more was needed for effective information retrieval than index terms for the original documents or abstracts and that it was vital to properly translate a question into index terms and codes before starting on a search.46 And for the scientists who had been unaware that classification tables had been translated in the past, it was a revelation to hear that one could provide multilingual access to information without translating documents, by translating the index terms and then using identical codes for those terms in the code dictionaries for each language. In this way, information encoded only once could be retrieved in any language for which a code dictionary had been prepared. Perry attempted to harmonize the views and experiences of several people. Among them were S. R. Ranganathan, University of Delhi, the great Indian philosopher and sage of librarianship, with whom Perry had some good discussions; Jacques Semain of France; Vernon Tate; Charles Bernier, of Chemical Abstracts; and other members of the ACS Committee on Scientific Aids to Literature Searching.

104 0

Perry recognized that cultural differences should be taken into account, although the examples he cited were relatively trivial, such as “translation” of the form of dates according to Christian, Moslem, or other calendars.47

105 0

Some of the machine translation work was supported by the Rockefeller Foundation, although most of the research in the evolving information disciplines had been funded by the Department of Defense and other government agencies.

106 0

Expanding Professional Interests

107 0

Government-sponsored research slowed down somewhat after the war, although in 1949-50 the government still awarded over $100

New Winds Blowing 143

million to research and development projects at universities and colleges alone. The National Science Foundation, proposed by Bush in 1945, was established by Congress in 1950. The military establishment, which earlier had not been particularly interested in supporting scientific information activities, in the decade after the war used the most advanced mechanical tools available for bibliographic purposes and now provided funds to study the more basic aspects of the organization of information and its retrieval. Grants were dispensed generously. To stimulate creativity, the detailed bureaucratic reporting requirements that characterized government grants in later decades were held to a minimum.48 Because of the research grants awarded to universities in the postwar period, academic libraries had to expand in the areas of government-sponsored research to provide the needed institutional support. The large-scale dissemination of wartime work, the expansion of libraries and newly established bibliographic and information centers, and the shortage of librarians provided job openings for a number of scientists. They were typically chemists or biologists who had become interested in scientific information through their wartime exposure to such work. They welcomed the challenge, as well as the new job opportunities in the postwar recession; many became technical information officers.

108 0

Some information officers grew impatient with conventional ways of producing, organizing, and disseminating documents and, in Jackson’s view, their prominence in documentation and their influence on management was out of proportion to their number. The scientists were often unfamiliar with library techniques and did not understand the special problems of servicing fast-growing collections. This led to a tongue-in-cheek definition of documentation as “library work performed by amateurs.”49 Yet librarians, on the whole, did not show deep concern about the problems to be faced because of the expanding technology. Only as funds became available for automation and for information retrieval projects through various agencies did librarians become interested in documentation.50

109 0

Working alongside scientists and engineers, librarians learned some of their methods, and, on the other hand, some technical people began to appreciate the work of librarians. As interest in bibliography and information management grew, it became necessary to provide or share the freshly distilled understanding from experiences in the field. Lectures, conferences, seminars, and group meetings were beginning to be held in the Washington area. The first significant event was a three-day conference on bibliographical control of government scientific and technical reports at the Library of Congress in the fall of 1947; the second was a lecture series in the fall and winter of 1949 on postwar

New Winds Blowing 144

library trends at the Graduate School of the USDA in Washington; from 1950 on events followed with greater frequency.32

110 0

The interest in documentation activities was growing to such an extent that in 1950, those active in ADI were “continually coming upon individuals and sometimes organizations” whose interests were nearly identical to their own.32 In 1949 the ACS had established a division of chemical literature “to hold meetings for the reading and discussion of papers and reports” for the purpose of studying and advancing “the art and science of the collection, recording, processing, exchange and dissemination of chemical information.” The concerns of the division extended to primary and secondary publication, abstracts and indexes, punched cards and electronic devices, library techniques, statistical analysis of numerical data, correlations, nomenclature, “and ideas for facilitating the use of chemical literature.”33

111 0

As documentation interests were intensifying, SLA started a Documentation Committee under Mortimer Taube’s chairmanship in the same year. The Group on Standardization of Information Services (GSIS), mentioned earlier, met regularly in 1950 to establish uniform practices in the technical information organizations within the U.S. government. The fundamental issues touching on bibliographic organization were discussed in July 1950 at a conference at the University of Chicago. In 1950-51, Washington area members of SLA, ADI, and ACS met informally to discuss documentation topics.34 An eclectic invisible college started to hold luncheons. Here librarians and scientific information workers, who were joined by scientific attaches and other embassy people, had an opportunity to keep up with the latest developments in automated information processing and documentation in the United States and abroad. These informal luncheons of the science and technology information group continued through the 1980s, albeit irregularly.

112 0

No formal organization existed in the postwar era for individuals responsible for information center operations or for those interested in information work. ALA and SLA did not provide suitable forums because the majority of librarians did not comprehend, or did not want to acknowledge, that the new challenges the major information centers had to face called for unorthodox solutions. The automatic equipment buffs and gadgeteers were technicians and typically were not concerned about the broader or more fundamental aspects of information handling. Scientists and scientific societies were only tangentially interested in these topics. The ACS had established a punched card committee that worked on codification of chemical compounds and had started a Chemical Literature Division, but its interests were too restricted.

113 0

New Winds Blowing 145

114 0

No formal training in documentation was being offered in the United States. English documentalists visiting the country in 1950, were surprised that library school curricula were limited to historical matters and the mechanics of library techniques with little or no attention being given “to study the process of disseminating information with a view to improving the prevailing arrangements.”55

115 0

In 1947, some hesitant experimentation with library school curricula was started, but the ALA, the accrediting body for U.S. library schools, soon realized that it was difficult to evaluate changing curricula properly and in 1949 placed a moratorium on new library school programs. Nonetheless, the first formal documentation courses in the United States were given in library schools: in 1949 Helen Focke taught a course on documentation at Western Reserve University and Margaret Egan taught one on bibliography at the University of Chicago. Jesse Shera gave a course on the organization of knowledge, and in 1951 Taube taught a course at Columbia University.

116 0

Still, with all these activities, communication among the widely dispersed groups interested in various aspects of information and its management was by happenstance. Little was written on work in the United States, and those articles were scattered in a variety of journals. Except for students with special interests, Americans typically do not read foreign languages, thus Otlet’s works were little known, and the issues of Otto Frank’s Handbuch der Klassifikation, published in 1946-49, were not even reviewed in this country.56 Only when Bradford’s penetrating essays on documentation were published in a slim volume in 1948 in London (and soon thereafter reprinted in the United States) did Americans have access to a comprehensive work on documentation.57

117 0

The practitioners could not agree on the definition of documentation, but in the early 1950s they did constitute a group that wanted to have an organization of its own. They came from disparate professional backgrounds but they shared the excitement and recognized the challenge created by the new technological developments, they wished to gain a better understanding of fundamental problems, and they wanted to benefit from the accumulated experiences of others in the management of scientific and scholarly information.

118 0

New Winds Blowing 146

119 0

NOTES

120 0
  1. Fritz Machlup and Una Mansfield, eds., The Study of Information: Interdisciplinary Messages (New York: Wiley, 1983).
  2. F. Donker Duyvis, “Report” (see chap. 4, n. 22). This document is cited below unless specifically mentioned otherwise.
  3. Correspondence between Stephen A. McCarthy and Luther H. Evans between 4 February 1948 and 6 June 1948, LC Research Materials 6.
  4. Rostow, United States in the World Arena, 55 (see chap. 1, n. 1); Richard Grenier, “Yanqui, Si! U.N., No!” Harpers 268 (January 1984):24.
  5. Robert B. Downs and Keyes Metcalf, cited in Conference on International Cultural, Educational, and Scientific Exchanges; Princeton University — November 25-26, 1946, ed. Edwin E. Williams (Chicago: American Library Association, 1947), ix-x.
  6. Royal Society of London, Empire Scientific Conference June-July 1946. Report; 2 vols. (London: Morrison Gibb, 1948).
  7. Royal Society of London, Scientific Information Conference; 21 June-2 July 1948. Reports and Papers Submitted (London: Royal Society of London, 1948); and Ralph Shaw, “Royal Society Scientific Information Conference,” Science 108 (13 August 1948):148-51. The conference was divided into four subject areas: publication and distribution of papers reporting on original work; abstracting and indexing services; indexing and other library services; and reviews and annual reports.
  8. Watson Davis, “Project for Publication” in Bernal, Social Functions of Science (see chap. 2, n. 3); and J. D. Bernal “Provisional Scheme for Central Distribution of Scientific Publications,” Royal Society of London, Scientific Information Conference, 252-58.
  9. Royal Society, Scientific Information Conference, 195-208.
  10. Adams, Medical Bibliography, 7-8 (see chap. 6, n. 27).
  11. Brownson, Helen, comp., “Recommendations and Results of International Conferences on Scientific Information and Bibliographic Services,” American Documentation 1 (Spring 1950):29-31.
  12. Jesse H. Shera, “The UNESCO Conference on the Improvement of Bibliographic Services,” American Documentation 1 (Summer 1950):144-46 (publication of the journal was late, hence included later material).
  13. “International Relations Board,” ALA Bulletin 40(15 September 1946):P97- P100.
  14. Morris Ullman, “Contemporary Trends in the Production and Use of Social Data,” in The Communication of Specialized Information, ed. Margaret E. Egan (see chap. 6, n. 32), 72-87.
  15. D. E. H. Frear, “Punch Cards in Correlation Studies,” Chemical and Engineering News 13 (23 November 1945):2077; and Chemical Biological Coordination Center, “The Chemical-Biological Coordination Center of the National Academy-National Research Council” (Washington, D.C.: National Academy National Research Council, 1954, 33 pp. Some of the early work is described in John C. Bailar, Karl F. Heuman, and Edwin Seiferle in “The Use of Punched Card Techniques in the Coding of Inorganic Compounds,” Journal of Chemical Education 25 (March 1948):142-43, 176; later practices, including indexing and abstracting practices, are discussed in G. Congdon Wood, “Biological Subject-Indexing and Information Retrieval by Means of Punched Cards,” Special Libraries 47 (January 1956):26-31.
  16. Some examples with references are given in Robert S. Casey, James W. Perry, Madeline M. Berry, and Allen Kent, eds., Punched Cards: Their Applications
    to Science and Industry, 2d ed. (New York: Reinhold Books, 1958). This volume is a remarkable complete reference about punched cards, their history and uses, the equipment to manipulate them, and the theoretical considerations regarding coding and the design of punched card systems.
  17. Clapp, “The Problem of Specialized Communication,” in Specialized Communication, ed. Egan (see chap. 6, n. 32), 32-35.
  18. Kent, Casey, and Perry, “Introduction,” in Punched Cards, ed. Casey, Perry, Berry, and Kent, 3-4; U.S. Patent 759,483 to W. K. Sparrow (1904), and U.S. Patent 873,305 to E. Eckert (1907), cited by Kent, Perry, and Casey, 4.
  19. Luther H. Evans, interview with author.
  20. Calvin N. Mooers, personal communication.
  21. Joseph Becker, “An Information Scientist’s View on Evolving Information Technology,” Journal of the American Society for Information Science 35 (May 1984):164-69; Jesse H. Shera and Donald B. Cleveland, “History and Foundations of Information Science,” in Annual Reviews of Information Science and Technology, vol. 12, ed. Martha Williams (White Plains, N. Y.: Knowledge Industry Publications for the American Society for Information Science, 1972), 254; and Jesse H. Shera, interview with author, Chicago, Ill., 14 November 1976.
  22. Queisser, Microchip, 75-79 (see chap. 6, n. 10).
  23. Jackson, “Unpublished Research Reports” (see chap. 6, n. 37).
  24. Vannevar Bush, “As We May Think,” Atlantic Monthly 176 (July 1945):101-8.
  25. Scott Adams, interview with author.
  26. C. D. Gull,“ A Punched Card Method for the Bibliography, Abstracting, and Indexing Chemical Literature,” Journal of Chemical Education 23 (1946):500- 507; and E. J. Crane and Charles L. Bernier, “Indexing and Index Searching,” in Punched Cards, ed. Casey, Perry, Berry, and Kent, 510-27.
  27. This expression originates with Robert A. Fairthorne, “The Patterns of Retrieval,” American Documentation 7 (April 1956):66-70. Objects must be “inscribed” in some way for retrieval, then placed in some order-a pigeon hole; thus they are “marked” and “parked.”
  28. Claire K. Schultz, interview with author, Philadelphia, Pa., 28 November 1978.
  29. Adams, Medical Bibliography, 92-94 (see chap. 6, n. 27).
  30. Ibid, 57.
  31. W. Boyd Rayward, “Librarianship and Information Research; Together or Apart?” in The Study of Information, ed. Machlup and Mansfield, 350-51 (seen. 1).
  32. The expression comes from Calvin N. Mooers in his “Information Retrieval Viewed as Temporal Signaling,” in Proceedings, International Congress of Mathematicians, Cambridge, Mass., vol. 1 (Providence, R. I.: American Mathematical Society, 1952), 572-73.
  33. S. C. Bradford, Documentation (Washington, D.C.: Public Affairs Press, 1950), 9. Bradford died in 1948.
  34. Tate, “Memorandum on the ADI,” 3.
  35. Norbert Wiener, Cybernetics, or Control of Communication in the Animal and the Machine (New York: Wiley, 1948).
  36. Claude E. Shannon, “A Mathematical Theory of Communication,” Bell System Technical Journal 27 (July and October 1948):379-423, 623-56; and Randall L. Dahling, “Shannon’s Information Theory: The Spread of an Idea,” in Stanford University, Institute for Communication Research, Studies in Innovation and Communication to the Public, Studies in the Utilization of Behavioral Science, vol. 2 (Palo Alto, Calif.: Stanford University, 1962), 119-39. Shannon’s comments are cited in Myron Tribus, “Thirty Years of Information Theory,” in The Study of Information, ed. Machlup and Mansfield, 475 ( n. 1).
  37. Joseph Ben-David, “The Profession of Science and its Powers,” Minerva 10 (July 1972):364.
  38. Shera and Cleveland, “History and Foundations” (see n. 21). The limitations of Shannon’s theory with respect to information science are expressed in Robert A. Fairthorne, “Morphology of Information Flow,” Journal of the Association for Computing Machinery 14 (October 1967): 710-19; and Manfred Kochen, “Views on the Foundations of Information Science,” in Information Science: Search for Identity: Proceedings of the NATO Advanced Study Institute in Information Science, 1972 August 12-20, Champion, Pa., ed. Anthony Debons (New York: Dekker, 1974), 171-96.
  39. Encyclopedia of Library and Information Science, ed. Allen Kent and Harold Lancour (New York: Dekker, 1970), s.v. “Descriptors,” by Calvin N. Mooers.
  40. Florence Casey, ed., Compilation of Terms in Information Sciences Technology for the Panel on Information Sciences Technology Committee on Scientific and Technical Information, Federal Council for Science and Technology, (Washington, D.C.: National Technical Information Service, April 1970).
  41. Mooers, “Descriptors” (see n. 38).
  42. Perry, cited by Madeline M. (Berry) Henderson, interview with author, Chicago, Ill., 18 July 1976.
  43. A good brief description can be found in Charles P. Bourne, “Evaluation of Indexing Systems,” Annual Review of Information Science and Technology, vol. 1, ed. Carlos Cuadra (New York: Interscience Publishers, 1966).
  44. Warren Weaver, “Translation,” in Machine Translation of Languages; Fourteen Essays, ed. A. Donald Booth and William N. Locke (New York: Wiley, Technology Press of the Massachusetts Institute of Technology, and Chapman Hall, 1955), 15-17. This is a reprint of Weaver’s memorandum of 15 July 1949, which was sent to about 200 of his acquaintances.
  45. Joseph Becker, “The Rich Heritage of Information Science,” Bulletin of the American Society for Information Science 2 (March 1976):12.
  46. Discussed in detail, for example, in Jessica Melton and James W. Perry, “Introduction to Analysis of Questions,” in James W. Perry and Allen Kent, Tools for Machine Searching (New York: Interscience Publishers, 1958).
  47. James W. Perry, “Information Analysis and Machine Searching,” American Documentation 1 (August 1950):133-38.
  48. Karl F. Heuman, interview with author, Bethesda, Md., 29 January 1976.
  49. Eugene B. Jackson, “Inside Documentation,” Special Libraries 48 (April 1954):151. This definition was also mentioned in Shera and Cleveland, “History and Foundations,” 252 (see n. 21).
  50. Kenneth W. Lowry, “Trends in United States Documentation Research,” Special Libraries 48 (October 1957):364-66.
  51. Jackson, “Inside Documentation” (n. 49).
  52. Vernon D. Tate to Charles Burton Fahs, 24 February 1950 (Carbon), SA, marked “Copy.“
  53. “American Chemical Society, A Division of Chemical Literature,“ Chemical Engineering News 27(14 February 1949):405.
  54. 54. Jackson, “Inside Documentation“ (n. 49).
  55. D. J. Urquhart, “American Impressions,“ American Documentation 2 (Spring 1954100-102.
  56. Otto Frank, Handbuch der Klassifikation, 5 issues (Berlin: Beuth-Vertrieb, 1946-49). Frank extended the work and later issued revised editions to the issues.
  57. Bradford, Documentation (see n. 33).