All histories are partial and provisional PDF document - DocSlides

All histories are partial and provisional PDF document - DocSlides

2014-12-13 117K 117 0 0


They are always told from a particular point of view In part this is determined by the particular time at which they are told in part by a partic ular point that is being made Who is telling the story Why What are they trying to con vince their rea ID: 23199

Direct Link: Embed code:

Download this pdf

DownloadNote - The PPT/PDF document "All histories are partial and provisiona..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.

Presentations text content in All histories are partial and provisional

Page 1
All histories are partial and provisional. They are always told from a particular point of view. In part, this is determined by the particular time at which they are told, in part by a partic- ular point that is being made. Who is telling the story? Why? What are they trying to con- vince their readers about? These are questions all historical discourse raises. In histories of the kind being excavated here, there is a strong sense of the present, a profound human desire to understand some aspect of the pres- ent, to create narratives that contextualize it in ways that make it seem comprehensible and manageable. How did we get to where we seem to be now and what happened on the way? This can lead to much that is insightful and compelling. It can also lead to oversimplifica- tion and special pleading. Some historiographical issues In a historical account of something that seems so limited, specialized, and apparently straightforward as a history of library applica- tions of computing, none of these historio- graphical issues can be avoided. Such accounts must describe patterns or trends of develop- ment and establish the backgrounds against which such developments occur. They must select evidence to support particular explana- tions of what happened and why it happened as it is argued that it did. They inevitably imply particular hypothesized futures or explicitly seek justification in them. Finally, they try to convince the reader to accept their particular point of view. In attempting to prepare an introduction to the history of the applications of computer technology in libraries, one is confronted with a number of practical as well as historiographi- cal issues. There are a great many libraries in the world. They are of different kinds in terms of the institutions within which they are locat- ed, the clienteles that they serve, the resources that they command, their aims and objectives, their size, and the sociotechnical complexity of their organizational arrangements. Their indi- vidual histories vary and reflect the manifold ways that they are stitched into the complex, historical, linguistic, and sociocultural fabric of their societies. So a first practical decision in relation to the articles offered here is to restrict the history that they discuss to North America. Within this geographic frame there is, even so, a rich diversity of libraries and library history. To be clear, my aim in this introductory essay is not to provide a systematic history of library-based automation. It is, as one of the guest editors of the articles that follow in this and the next issue of the IEEE Annals of the History of Computing , to provide a framework and context to assist readers who may have lit- tle background knowledge of libraries. Few his- torians of libraries and librarianship have attempted systematically to analyze the histor- ical record involved and to explore possible approaches within which to problematize or theorize aspects of this record. Because there is so little already written on this subject to guide readers, I alert them here to identify who is responsible for the accounts being offered in each of the articles, to ask what their points of view might be and the interests that might be at stake in their accounts. In the following, I offer a kind of integrationist historical argu- ment of library development that is tentative and open to dispute and for which the evi- dence offered is by no means complete. It draws heavily as points of reference, however, IEEE Annals of the History of Computing 1058-6180/02/$17.00  2002 IEEE A History of Computer Applications in Libraries: Prolegomena W. Boyd Rayward University of Illinois Modern libraries are constituted within and by a tradition of techniques and practices that represent 100 years of codified professional knowledge. This article provides a historical overview of this tradition that created a complex environment of expectation and misunderstanding for introducing library automation. A generation of systems development was needed to assimilate and further develop this tradition.
Page 2
on the articles that Rebecca Graham and I have included in these two library automation issues of Annals. A historical argument The argument offered here could be extend- ed in various ways to place the development of library automation and its role in the transfor- mation of the modern library within the framework of grand theories about Fordism and post-Fordism, progressivism, modernism and postmodernism, and the information soci- ety. The emergent practices, processes, tech- nologies, and ideologies that constitute the modern library could be problematized in terms of Taylorian theories about scientific management. The professionalization of librar- ianship might well be discussed in a Foucauldian framework of the disciplinary regimes within which the subject is created and controlled or in terms of Weberian theories of bureaucratization and organizational rational- ity. The library is, after all, an important com- ponent of the modern social apparatus for the management of public knowledge and there- fore of the citizen and the state. We lack sophis- ticated analyses of these kinds of the history of library automation, which is a matter of great regret and opportunity. Such analyses, how- ever, lie beyond my competence. In this intro- duction my aim is simple and pragmatic. It is to present an argument that is speci cally lim- ited to attempting to explain why library automation was difficult when it was first undertaken, to suggest the stages that it went through, and what some of the implications have been for the development of modern libraries and librarianship. I argue that the history of computerization in US libraries is intelligible only against a back- ground of the professionalization of librarian- ship in the US during the latter part of the 19th and early part of the 20th centuries. I take the view that professionalization was intimately bound up with the creation and maintenance of a range of technologies that were speci c to libraries. Long before the advent of computers, I suggest, the sophisticated and evolving tech- nologies that the modern library deployed were highly adapted to its needs and purposes. Distinctively library social roles and func- tions and the emerging professional values of librarianship were established by and through these technologies. The technologies helped to standardize and structure work practices in libraries. By the early 1930s, libraries across the country were organized in similar ways to sup- port common systems and processes. They offered variants of the same kinds of library service to what had become conventionally dif- ferentiated clienteles adults, children, school or college students, university and college fac- ulty, corporate management, the blind, immi- grants, the illiterate, and so on. The details of what libraries were, what they would do, how they would do it, and for whom have never been xed, of course, but have nec- essarily been subject to constant elaboration and negotiation. But the elaborations and negotia- tions through which professional change has occurred have been conducted within a histori- cally emergent discursive framework within which any profession presumably must be con- stituted. Such a framework involves common assumptions and a shared body of established knowledge that determines what can be talked about, how it is to be talked about, and who can engage in the discussions. For librarianship, like most other professions, this discursive frame- work is typi ed by the meetings and publications of professional associations and by programs of professional education. The framework to which I refer has ensured that change has proceeded in an orderly fashion within a broad consensus that had, as I say, become established by the 1930s as to the nature of the structures and patterns of library organization and operation. This frame- work has proved exible enough to contain most of the dissension and con icts that invariably accompany change. Further, modern libraries were quickly rec- ognized as labor-intensive work environments in which acquiring and processing large num- bers of individual items for their collections or circulating them to their clienteles involved repetitive, essentially clerical tasks. I have argued that modern libraries have similarities to the bookstore, the supermarket, even the fac- tory, yet they were different. Although the tools and practices they created to perform their functions were much concerned with ef ciency and economy, for librarians these con- cerns led to a deeply rooted belief in the value of cooperation. Cooperation meant not only personal cooperation between librarians work- ing within the structure of national and state library associations but also organizational cooperation between their libraries. It was the essentially satisfactory nature of the technolo- gies of librarianship and the discursive process- es in which they were embedded that made librarians little interested until after World War II in new technologies, such as those based on Hollerith and other punched cards. (See R. Williams, The Use of Punched Cards in Libraries and Documentation Centers in the AprilJune 2002
Page 3
United States, 1936 1965, in this issue.) One exception to this generalization may well be micro lm technology. When existing library technologies began to break down under economic, social, and bibli- ographical pressures emerging in the late 1950s and early 1960s and librarians began seeking relief in automation, the introduction of automation involved profound misunder- standings between librarians and early systems developers about the nature of the professional knowledge and traditions of librarianship on the one hand and of the capabilities of the new technology and what was required for its implementation on the other. An outline of library history In this section, I discuss some highly select- ed events in the history of the emerging pro- fession of librarianship by means of which my argument will be further elaborated. I stress here the evolutionary continuities in profes- sional practices and technologies that charac- terized librarianship at the point it was confronted with the early possibilities of automation. These possibilities were both ini- tially embraced and resisted as the librarians began what may be regarded as their ongoing struggles to master the new technologies avail- able to them. None of the developments that I do little more than list below is simple or straightforward. Each is embedded in a variety of interconnected social, intellectual, institu- tional, and professional circumstances that cannot be dealt with here. Each involves a voluminous literature. Some of the develop- ments have already been the subject of excel- lent historical studies; others have not. Underlying the developments that I outline is a search for economy and ef ciency of library operations to which cooperation between librarians was seen as indisputably necessary. Associations The public library, a major component of the library system in the US, emerged both in the US and the UK at roughly the same time in the mid-19th century. With the creation of the American Library Association in 1876, a formal structure for librarianship appeared. The ALA has met annually since 1876 and later half- yearly as well. In the course of its history, the association has been organized into what were at different times known as departments, divi- sions, and most recently separate associations within the overall structure of the parent body. These structural arrangements have reflected emerging and changing emphases of interest in terms of services and clienteles in the course of the association s history. There are now 11 divi- sions and a number of smaller roundtables and state chapters. Each of the divisions publishes journals related to its specialized concerns. The Journal of Library Automation , for example, was begun in 1968 by what was then named the Information Science and Automation Division. In 1982, the division and the journal were given their current names: Library and Information Technology Association and Information Technology and Libraries ITAL ). Of course, librarianship in the US is also rep- resented by state library associations and a number of specialized associations as well as the ALA. All of these associations hold their own meetings and publish journals and other material related to their elds of interest. Publications Publications, along with a structure of meet- ings, are essential to the discursive framework within which, I suggest, that librarianship in the US was constituted as a profession. Of crit- ical importance in this respect has been the publishing activity of the ALA. On formation in 1876, it sponsored a journal, but it was 10 years before it began a major, general profes- sional publishing program. It created a pub- lishing section in 1886 to secure the preparation and publication of such catalogs, indexes and other bibliographical helps as may best be produced by cooperation. Among the important professional tools it has published over a period of nearly 120 years are a series of standards for various types of libraries. These documents have speci ed the levels and kinds of support, the organizational structures and the nature and levels of service provision required for the successful operation of the type of library involved. The ALA has inevitably also played a major role in the technical discussions leading to the development of the profession- al tools discussed below and has published many of them either collaboratively with other organizations or separately. In addition to its wide range of specialized journals, ALA now sells more than 100,000 copies a year of the books (and some other kinds of library resources) published under its imprint. Dewey decimal classification For John Metcalfe, the Australian librarian and theorist of subject indexing, 1876 was an annus mirabilis in the history of librarian- ship. Not only did that year see the emergence of the ALA, it also saw the publication of the landmark US Bureau of Education Report sur- IEEE Annals of the History of Computing A History of Computer Applications in Libraries
Page 4
veying libraries in the US. Included in this report were Charles Ami Cutter s Rules for a Dictionary Catalogue 10 and an article by Melvil Dewey on his decimal classi cation. The latter was published anonymously in its rst edition in that signal year (it bore a copyright notice, however, in Dewey s name). Then a pamphlet of 44 pages, 11 it is now used throughout the world for the physical arrangement of library materials, especially in public libraries. 12 The 21st edition of the Dewey decimal classi cation, now an extremely complex bibliographical tool, was published in 1996 in four volumes com- prising 4,037pages. It is also available in a Web version. 13 Library of Congress classi cation At the turn of the 20th century, the Library of Congress, having examined the Dewey clas- sification system along with other rival sys- tems, decided that they were inadequate for arranging collections of the size and complexi- ty for which it was responsible. Publication of the schedules of its own special classification system began in 1901, and a first edition of most of the schedules had appeared by 1915. Other libraries interest in using the LC classi- fication grew slowly, and LC staff in the early years of its development made its schedules freely available. By 1920, ve of the largest aca- demic libraries in the US had adopted the clas- sification. In the period of rapid growth of library collections in the decades after World War II, this movement swelled to such an extent that By the mid 1970s, nine tenths of the largest American libraries and over half of all American libraries were using the LC sys- tems. 14 It is now in 48 regularly revised vol- umes and is available on CD-ROM. The Dewey classification has continued to be used, how- ever, by most public libraries and smaller libraries in general. Most libraries in the US (and indeed elsewhere) use one or the other of these two systems. Cataloging rules With an apologetic nod in the direction of the work carried out in the middle of the 19th century by Antonio Panizzi at the British Muse- um and Charles Cof n Jewett at the Smithson- ian Institution, one may argue that the fourth and posthumous edition in 1904 of Cutter s Rules for a Dictionary Printed Catalog sits at the source of the structure of American cata- logs and the practices of the catalogers who produce them. 15 From his Rules and the Anglo- American Cataloging Rules of 1908 based on them, in a complicated genealogy, descend subsequent developments of principles and rules for the bibliographical description of library materials, the speci cation and structure of headings as entry points for library catalogs, and catalog organization. With the advent of the machine-readable cataloging (MARC) for- mats in the late 1960s and development of the International Standard Bibliographical Descrip- tions (ISBDs) in the early 1970s under the aegis of the International Federation of Library Asso- ciations (IFLA), these developments cover the transition from card to online catalogs. (See S. McCallum, MARC: Keystone for Library Automation, in this issue.) The ISBDs now form a complex series of standards for the description of a wide range of formats of library materials. These fundamental library tools rep- resent almost a century of discussion and national and international standardization agreements. 16 The development of modern cataloging codes in the US has involved a cyclical process of meetings, negotiations, drafts, published codes, further revisions, radical critiques, more meetings and negotiations prior to the next stages of codification, and the creation and modi cation of professional structures within which to handle this process. It has taken place in an increasingly international context against a background of rapidly changing technology. This highly politicized process would make a fascinating subject for further study. Table 1 (next page) gives an outline of major develop- ments in the past 120 years. Subject headings In addition to the physical description of items was the equally vexatious and challeng- ing professional problem of nding subject ter- minologies and classification systems that could express systematically both what each item was about and the network of connec- tions between related concepts. 21 The problem of how to specify subjects, given the enormous variability of the language of documents and that of the language of library users, was recog- nized early as something that would need to be managed systematically. Fundamental to this process was the elaboration of standard termi- nology lists or what were called subject head- ings lists. In 1895, the ALA issued a List of subject headings for use in dictionary catalogs quite explicitly to form a kind of appendix to the third (1891) edition of Cutter s Rules for a Dictionary Catalogue. 22 Publication of the list was discontinued in 1911. The standard authority for subject headings, however, soon became the subject headings used in the dic- April June 2002
Page 5
tionary catalogs of the Library of Congress. The library had issued the rst edition of this list in 1897. 23 A second edition followed in a huge volume in 1919. The list has been continuous- ly under revision since then and is now known simply as LCSH. It comprised five large vol- umes in its 24th edition of 2001. It is updated weekly and is available on microfiche and on CD-ROM, which cumulates every three months. It has been the subject of a huge ped- agogical as well as critical literature. To provide a manageable subject headings guide for school and public libraries, Sears List of Subject Headings for Small Libraries was published in 1923, and reached a 16th edition in 1997. The dictionary catalog Central to the management and function- ing of the professionally organized US library as it emerged from the end of the 19th century was its dictionary card catalog. It was to perfect this instrument that the various tools men- tioned previously have been developed. The fact that the catalog was constructed from cards that were standardized physically as to size and weight was of the greatest importance. The use of cards allowed for intercalatability of entries so that catalogs could always be up-to-date. And of course, the configuration of the infor- mation inscribed on the cards was also stan- dardized according to the various codes of rules for entry and descrip- tion. Entries in the catalog were multi- ple, that is, there were as many entries as might be needed at least up to a point to represent a multi- authored, multisubject work. Important in this respect was the notion of the main entry. This was a card on which all of the informa- tion about a particular library item was recorded. Typically the heading would be for an author, where there was one, or the agency responsible intellectually for the work. This would be followed by the body of the entry in which the item was described and other highly stylized information about it recorded in the form of notes. Finally, in what were called tracing notes, the card would list the headings under which added entries for that item were to be made in the catalog typically other authors, titles, and subjects. The dic- tionary arrangement for the author and subject entries in the catalog generated intricate problems for filing cards, which were dealt with by the development of special codes of ling rules, reference to some of which is included in Table 1. Once dif cul- ties about the structure and ling order of the various headings were resolved and cards were actually filed in the catalog, the alphabetic arrangement of the different types of headings as though in a dictionary was considered to be familiar to most library users. At one level, the dictionary catalog was indeed simple like a dictionary with a special form of navigational assistance for the user built into it the see and see also references that formed what was called the syndetic structure of the catalog. On the other hand, any particu- lar catalog was also highly arti cial (in the best sense of the word). It was a complex tool emerging over the years from innumerable acts of professional decision-making. The principles underlying the decisions represented in the cat- alog were necessarily unstated in the catalog itself and in any case varied as cataloging con- ventions were modi ed and changed with the passage of time. The cataloging codes, codes for filing cards, classification systems and subject heading lists all were needed for catalog con- struction. They represented simplified but highly formalized adaptations of common des- ignations and everyday linguistic usages. As a IEEE Annals of the History of Computing A History of Computer Applications in Libraries Table 1. The creation of library catalogs 18761997. A conspectus of tools and critiques in the historical search for best practice. 17 Year Publication/Event 1876 Cutter s Rules for a Printed Dictionary Catalogue , 1st ed. 1904 Cutter s Rules for a Dictionary Catalog, 4th ed. 1908 Catalog Rules: Author and Title Entries. (The Anglo-American Cataloging Code) 1941 Preliminary second edition by the ALA of the 1908 rules 1941 A.D. Osborn, The Crisis in Cataloging 18 1942 ALA Rules for Filing Catalog Cards 1948 Rules for Descriptive Cataloging in the Library of Congress 1949 ALA Cataloging Rules for Author and Title Entries 1953 Seymour Lubetzky, Cataloging Rules and Principles: A Critique of the ALA Rules 19 1956 Filing Rules for the Dictionary Catalogs in the Library of Congress 1961 International Conference on Cataloging Principles, Paris 1965 Draft of Machine-Readable Cataloging (MARC) format 1967 Anglo American Cataloging Rules (AACR) 1968 MARC II format 1969 International Meeting of Cataloging Experts, Copenhagen 1971 International Standard Bibliographic Description (ISBD) 1974 First meeting of Joint Steering Committee for Revision of AACR (JSC) 1978 Anglo-American Cataloging Rules, 2nd ed. (AACR 2) 1980 Library of Congress Filing Rules 1988 Anglo American Cataloging Rules, 2nd edition revised (AACR2 R) 1997 International conference on the principles and future of AACR, Toronto 1998 AACR2 e (CD-ROM under constant revision by Joint Steering Committee) 20
Page 6
result, the catalog s organization and content were not necessarily transparent to the user, and its use was at times extremely dif cult. Library of Congress card distribution scheme Along with the ALA, the Library of Congress has had the greatest impact on the standardi- zation of practices in US libraries historically. OCLC (Online Computer Library Center, for- merly the Ohio College Library Center) was to join them as a third force in the years follow- ing its creation in 1971. I have described the importance of the Library of Congress s subject headings lists, its classi cation scheme, and its cataloging rules. But perhaps what most con- solidated the influence of the Library of Congress and led to the widespread adoption of the cataloging principles and practices it endorsed and so contributed to the general standardization of library operations across the nation was its card distribution scheme. This permitted libraries to order copies of the cata- log cards the Library of Congress had prepared for its own use instead of cataloging a book or other item themselves. Libraries from around the country (and eventually from around the world) could order sets of catalog cards for items that they wished to incorporate into their collections. Introduced in 1901, the card dis- tribution scheme had become more or less self- supporting by 1903, so great and immediate was the demand for the service. As Jane Rosenberg has observed, Library of Congress cards changed the American library world they standardized library processing and made it more efficient. Rosenberg notes that, by 1935, there were 5,738 subscribers to the card service. 24 Later, to deal with increasing volumes of publication and the strains this imposed on it, the Library of Congress instituted a number of schemes for cooperative cataloging. A select group of accredited research libraries, for example, already sending in cards to the National Union Catalog , contributed their cata- loging to the card distribution service. The library s cataloging-in-publication (CIP) project is a variant of this idea of centrally produced, locally consumed cataloging data: The CIP program began in 1971 as a special proj- ect, funded in part by grants from the Council on Library Resources Inc. and the National Endowment for the Humanities. Since then the CIP program has evolved from creating 6,500 pre-publication cataloging records during its rst year to creating more than 57,000 catalog records annually. 25 Publishers send prepublication copies of their books, or core selected portions designated by the library, to the library, which prepares skele- tal cataloging data that is then usually pub- lished on the verso of a book s title page. Additionally, various cooperative acquisitions schemes have been created in which cataloging is done elsewhere (in some cases outside the US) and accepted by the library for subsequent national distribution. One of the perhaps unexpected conse- quences of the Library of Congress card distri- bution scheme was that it reduced the importance of cataloging as a central disci- pline in library schools. This was especially so for those whose professional lives were to be spent outside the research library milieu, where original cataloging would not general- ly be needed. Library of Congress printed catalogs The Library of Congress s influence within the library community was further extended by two major bibliographic publications, which had become possible with the intro- duction of photo-offset printing. First, in 1942 the library published its own card catalog in 167 volumes. 26 It issued various supplements subsequently. Second, it undertook to publish the National Union Catalog, Pre-1956 Imprints from card files that had been cumulating for nearly 70 years. This required 754 large vol- umes and took 13 years to complete. 27 What might have seemed at the time an apotheosis pales, however, when compared to the extra- ordinary expression the union catalog idea has received through today s sophisticated Web- based technology. Cooperative cataloging in the print world based on the conventions, tools, and practices of the kind previously described in the com- puter world has created the enormous databas- es developed by the modern bibliographic utilities such as OCLC and RLIN (Research Libraries Information Network). These data- bases represent the latest developmental stage of speculative schemes formulated in the mid- 19th century, that were instantiated in the card distribution service of the Library of Congress at the turn of the 20th century, and 60 years later given monumental expression in the printed National Union Catalog, Pre-1956 Imprints . Now they incorporate the cataloging data produced not only by the Library of Congress but by other national libraries from around the world as well as cataloging data supplied by an enormous membership of libraries of various kinds. April June 2002
Page 7
Document supply services The dictionary card catalog s major function was to provide library users with information about books and other material in the library. The library s bibliographical reference tools, especially indexing and abstracting services for periodicals, provided information not only about what was available in the library but also about what was not. Librarians developed tools and systems to enable them to obtain such materials either on loan or, with the advent of various photocopying technologies, as copies of what was needed. Union catalogs, such as the National Union Catalog and the huge Union List of Serials published initially commercially but consolidated by the Library of Congress and continued by it as New Serial Titles , were important to this function. 28 Codes of practice for interlibrary lending were developed nation- ally and internationally through IFLA. Most modern libraries were fully accustomed by the 1930s to the idea that, through their reference and interlibrary lending functions, they pro- vided what might be called gateways to wider stores of library materials, of which their own collections were part, than any one of them could hope to possess. These were functions profoundly to be affected by computerization, especially by recent Web-based technologies. Most libraries (except the great reference and research libraries), having registered legit- imate users, permit them to borrow materials. This was, after all, a basic library function. It necessitated the formulation of rules and regu- lations about what could be borrowed, by whom (at what age were children able to bor- row adult materials?), for how long, and the consequences of delinquency. Various systems to manage records of who borrowed and what was borrowed were developed and again stan- dardized the Brown and Newark systems, for example, were widespread card systems in the days before computing. These along with vari- ous photo-charging systems identified what books were on loan, allowed the books when necessary to be recalled or to be reserved for loan to others, and of course, identi ed mate- rials whose return was overdue. As simple- minded as they might have seemed, such systems were an integral part of library tech- nology and were indispensable for the control of the extramural use of library collections. 29 It was this function because of its clerical, repeti- tive nature that particularly attracted early experimental technological applications. Libraries precomputer technological base The precomputer technology that under- girded the traditional library organization nec- essarily involved a number of discrete subsys- tems that were physically and organizationally separated from each other the catalog, shelf list, circulation system, acquisitions system, periodicals and serial check-in systems, and binding system, for example. This range of subsystems with their catalog cards, paper files, card trays and cabinets, along with the data they manipulated and the practices their effective operation required were, always with local adaptations and variations, pretty much standardized throughout US libraries. With the development of the ideal of a single integrated computer-based library system, which comput- ing made possible and which was dramatized in the mid-1970s by the experimental systems at the University of Chicago 30 and Northwestern University Library, 31 the former eventually fail- ing, the latter becoming a successfully com- mercialized product, NOTIS (Northwestern Online Totally Integrated System) this system fragmentation would be recognized as a major drawback to efficient library operations. However, it was based on organizational reali- ties that were not easy to change. The library marketplace Modern libraries became, over time, a not inconsiderable marketplace. Book trade pub- lishers, scholarly and scientific societies and commercial publishers of indexing and abstracting services, and library jobbers and suppliers were active in it, competing to pro- vide both materials and services to libraries. Of particular importance were the jobbers and suppliers from whom, in a kind of early out- sourcing, libraries obtained components of their technical processing work. Perhaps the first of these agencies was Melvil Dewey Library Bureau, formally established in 1881. Its main business, before it expanded into the general of ce market, was to supply card stock and equipment to libraries. 33 Depending on a particular library s needs and the nature of the materials involved, a jobber not only might obtain materials ordered by a library but might also process them in a preliminary way for incorporation into the library, such as by sup- plying catalog copy, inserting book pockets, laminating book jackets, and so on. A particu- larly important group are periodical subscrip- tion agents because of the range of dif culties this kind of material presents. The library marketplace was to become fur- ther differentiated when various computer- based entrepreneurs and systems developers became active in it. The dimensions of the 10 IEEE Annals of the History of Computing A History of Computer Applications in Libraries
Page 8
library systems component of the market as it has emerged and ourished and changed is in need of investigation. Such an investigation would chart the different origins of systems vendors, their number and longevity, the kinds of systems that they introduced, and the over- all economics involved. Without such an inves- tigation, a full understanding of the history of the automation of libraries is incomplete. 33 Professional education The professionalizing of American librari- anship has involved professionalizing the edu- cation of librarians. The rst library school was founded in 1884 by Melvil Dewey in Columbia College. The school was later moved to the State University of New York, Albany, when Dewey became the state librarian there. 34 Graduates from Dewey s school were important figures in creating other schools elsewhere in the country, many being set up initially in libraries, although all were eventually trans- ferred to universities. 35 In the late 1920s, fol- lowing the Williamson Report and with a grant from the Carnegie Corporation, the Graduate Library School at the University of Chicago was set up on the model of graduate schools in other professions, combining programs of pro- fessional education with research and scholar- ship in the substantive disciplines underlying professional practice. 36 This school was to become a model for several generations of library educators until its closure by the uni- versity in the mid-1990s. Concerned with edu- cation for the profession early in its existence, the ALA created a Training Board in 1925 that became its Committee on Accreditation in 1955. This continues to manage the ALA accreditation process and the standards that guide the process. 37 The arrival of computing in libraries For nearly a century modern libraries had been creating and settling themselves into a highly complex but essentially stable profes- sional world. By the late 1950s everything began to change. This was the period of Cold War rivalries, post-Sputnik scienti c and tech- nical excitements, and economic growth. Moreover, computer technology had begun to come into widespread use. It was a heady peri- od of optimistic technological expansionism that captured the imagination of the general public (see Cheryl Malone s article, Desk Set, to appear in the July-Sept. 2002 issue of Annals part 2 of the libraries theme). Libraries, howev- er, now began to experience new kinds of pres- sures. The higher education system and the research establishment, expanding rapidly, cre- ated new demands on old libraries and, indeed, demands for new libraries. (See R. Bregzis, C. Gotlieb, and C. Moore, The Beginning of Automation in the University of Toronto Library, 1963 1972, in this issue.) Libraries had huge collection gaps created by World War II. Cooperative acquisitions projects coordinated by the Library of Congress, ALA, and the Association of Research Libraries were designed to fill these gaps and to ensure that copies of anything published anywhere in the world would be available in the US research library system. Apart from these and other special projects, libraries generally were having to acquire and manage access to rapidly increasing volumes of publications. Of critical importance was the exponentially escalating volumes of scientific and technical literature. This comprised a cate- gory of materials that were formally published and fell within the purview of established sys- tems of bibliographical control, increasingly strained though such systems were becoming. But it also comprised a new and increasingly large and messy category of what was described as gray literature that fell outside the reach of these systems. This literature consisted of vari- ous kinds of working papers and reports in near- print form. The quantity, short useful life, and problematic nature of this body of literature were to be powerful incentives to the develop- ment of new techniques of information retrieval. The idea of computer-based solutions to the range of increasingly difficult problems that libraries were experiencing became potentially attractive to them. Many libraries were begin- ning to face the breakdown of the systems that had served them so well in the past and which incorporated so much of the librarian s profes- sional knowledge. Reynolds describes some of the operational crises that encouraged librari- ans to turn to automation: An increasingly large backlog in cataloging, the lack of accurate fund accounts in acquisitions, hopelessly dis- organized circulation files, or a large measure of internal inconsistency in a public catalog. Computers seemed to offer librarians the prospect of more ef cient processing, improved services to library users, saving money and con- taining costs, and facilitating resource sharing and library cooperation, all concerns central to their traditional operation. 38 These giant brains threatened to transform the nature of work. Maybe they could or maybe they could not think. 39 They operated by means of a highly systematized, routinized, April June 2002 11
Page 9
atomization of tasks. Was not the library an extremely labor-intensive organization much of whose work was accomplished with unit- based, algorithmic processes standardized in library systems across the country? Seemingly, libraries were ideally suited to the application of the new technologies. Both librarians and computer experts were eager to think so. However, library automation began essen- tially in an environment of complex misunder- standings that took some time and effort to dispel. The nature of some of these misunder- standings is revealed in Burke s account of Project Intrex (see C. Burke, Different Techno- Visions: The Ford Foundation s Search for an American Library Laboratory, to appear in the July Sept. 2002 issue of Annals ) and Bregman and Burger, Implementation of Library Automation at the University of Illinois Library at Urbana-Champaign, 1965-2000: A Case Study of Technological and Organizational Validity, elsewhere in this issue. If these mis- understandings were to be dispelled, techniques and practices representing nearly a hundred years of codi ed professional knowledge could not simply be swept away although computeri- zation might have made this seem possible. It was to become clear that this tradition had not led to the stultifying accretions of now-irrele- vant dogma. Libraries were in effect constitut- ed within and by this tradition to an extent that neither librarians nor technologists fully appre- ciated at the time. What was needed was not the replacement of the librarian s hard-won pro- fessional knowledge but its further develop- ment to deal with the opportunities presented by the new technology. This is a simple point but one that requires an emphatic statement. Library automation was not to be implemented successfully until this development had occurred, essentially signaled by the creation and widespread adoption of the MARC and ISBD standards. Librarians, on the whole, whether embrac- ing the new computer technology reluctantly or enthusiastically, did not at first realize the extent to which computer applications would result in imperfectly realized, much-criticized system products. They did not envisage how much disruption would be created in the rou- tine functioning of their libraries. Nor did they foresee the eventual structural reorgani- zations of various kinds that would have to be undertaken, reorganizations that would involve both changes in the composition of the staff and heavy requirements for staff training. And of course, there was the ultimate burden of cost. For most libraries, the costs of initial system development could only be met by access to external sources of funding. Few realized at the outset the extent to which ini- tial computerization would require not only major subsequent and continuing expendi- tures for systems maintenance and replace- ment but radical and permanent shifts of budget priorities. The entrepreneurial technologists and sys- tems developers, with their diverse origins in governmental organizations and the industrial- military complex, were perhaps misled by what on the surface seemed to be relatively simple professional tools and products. Circulation records, serials check-in files, acquisition sys- tems, even the card dictionary catalog itself arrayed in its endless banks of cabinets, did not seem to be mysterious mechanisms of library operation and management. The technologists and systems developers did not soon or readi- ly comprehend the subtlety, complexity, and variability that these surfaces concealed and that their automated systems would be required to capture. Thus the time needed for implementing the new systems and the extent of the costs involved were frequently grossly underestimated and even on occasion perhaps willfully concealed (see C. Burke, Different Techno-Visions: The Ford Foundation s Search for an American Library Laboratory, to appear in the July Sept. 2002 Annals ). Both groups, librarians and technologists alike, I suspect, were caught up in the early days of automation in what might be described as a kind of eschatological discourse lled with promises and prospects of salvation and of dis- appointed expectations. There were notable expos s such as that by Harrison Bryan, a promi- nent Australian librarian seeking to assess the library automation situation in the US in the mid-1960s. He was obviously surprised to nd so much talk and such little action. 40 The title of Ellsworth Mason s 1971 article speaks for itself: The Great Gas Bubble Prick t 41 It was clear that neither the librarians nor the systems devel- opers were able to assess realistically what could be done with what were at that time extremely limited capabilities of the technology. Initially developed for number crunching, the comput- er was ill-adapted to the linguistic or text-based requirements of libraries. (See J. Segesta and K. Reid-Green, Harley Tillitt and Computerized Searching, to appear in the July Sept. 2002 Annals .) But lying beneath the mutual incom- prehension that I think characterized so much of the early relationship between librarians and the technologists were different professional cul- tures and analytical and linguistic practices that 12 IEEE Annals of the History of Computing A History of Computer Applications in Libraries
Page 10
took at least a generation of trial and error in sys- tems development for mutual accommodation to be attained. Rapid changes in computer and communi- cations technologies was one of the engines of systems development. Reynolds periodizes these often overlapping changes for library automation essentially by decades. First there was the precomputer era of unit-record equip- ment. (See Williams article elsewhere in this issue.) Then came the of ine computerization of the 1960s and early 1970s. This was followed by the online systems of the 1970s. (See D. King and J.M. Griffith, Fifty Years of Information Retrieval Research, Development and Use, and J. Segesta and K. Reid-Green, Harley Tillitt, to appear in the July Sept. 2002 Annals .) The 1980s saw the advent of microcomputers and the emergence of CD- ROM technology. Finally, we come to the Internet revolution of the 1990s. 42 During all this time, libraries faced ever- increasing pressures on their resources. The prospects of automation remained beguiling in their promises of relief. Funding also became available to support both demonstration proj- ects and systems implementation. Notable in its support of research and development was the Council on Library Resources founded in 1956. (See D. Marcum, From Heaven Hill to the Digital Library: The Role of the Council on Library Resources, to appear in the July Sept. 2002 Annals .) Other funding came through the federal government, through foundations, and some derived from internal budget support. As libraries automated, the library systems mar- ketplace became a dynamic source of systems development and marketing that now presents a spectacle of bewildering variety. From the broad perspective I have adopted here, it seems that another way to periodize library automation history is from the point of view of library development rather than that of computer development. From this view- point, one might argue that library automa- tion falls into three general periods: pre-MARC, post-OCLC, and post-Internet. The importance and the development of MARC are thoroughly dealt with by Sally McCallum in this issue. OCLC, however, has been mentioned only in passing, both here and in the other articles in this issue. It was set up in 1971 as the Ohio College Library Center, designed to facilitate library cooperation and to reduce the costs of library processing. The idea, according to Fred Kilgour whose brainchild it essentially was, was to develop an online computerized library network having half a dozen major subsystems: (a) an on-line union catalog and shared cataloging subsystem, (b) seri- als control subsystem, (c) technical processing subsystem, (d) on-line interlibrary loan request- ing system, (e) retrieval by subject, (f) remote cat- alog access by readers and circulation control. 43 It developed rapidly, expanding its mem- bership first to libraries in Ohio and then libraries nationally and, of course, more recently internationally. Kilgour was able to boast that, by 1975, its online catalog con- tained nearly 2 million records and 8 million locations. 43 By 2002, what is now called WorldCat at OCLC recorded 833,324,664 loca- tions around the world for items represented by 46 million cataloging records. 44 As well as implementing systems related to all of the functions Kilgour listed, it has become an important center for research and develop- ment, which now includes the Dewey decimal classification as a result of the acquisition in 1998 of Forest Press. It seems appropriate to label the period in the history of library automation, and of librar- ianship more generally, from the introduction of MARC to the current Internet and Web- based developments, the era of OCLC, both for its own sake but also as representing the many library networks and bibliographical utilities that emerged at this time. Most of the articles in this and the following issue of Annals deal with events within this period, though none deals speci cally with OCLC itself. It has been, however, a central player in most of the devel- opments in librarianship since MARC, includ- ing the transition beyond online computing as it was conceived of at the time of OCLC s foun- dation to the modern digital world of the Internet and the World Wide Web. Finally, we come to the present where histo- ry is still being written and where libraries con- tinue to wrestle with the opportunities and challenges of changing technologies. In trying to come to grips with what is happening, to assess what is possible and desirable in planning for the future, libraries need to be understood in the historical context within which they have been created and developed as outlined here. Such an understanding is not so much valuable in preventing the repetition of past mistakes, one of the functions often attributed to history; it is to begin to know more fully what libraries are for and how they work. We hope that the articles in the two library automa- tion issues of Annals will make a contribution to the search for this knowledge. April June 2002 13
Page 11
References and notes 1. A. Black and D. Muddiman, Critical Perspectives, Critical Methodologies: The New Sociology of Information and Public Library Research in the United Kingdom, Library and Information Stud- ies, Research and Professional Practice , M. Beaulieu, E. Davenport, and N. Ole Pors, eds., Taylor Graham, London, 1997, pp. 204-219. Also see G.P. Radford and M.L. Radford, Libraries, Librarians, and the Discourse of Fear, Library Quarterly, vol. 71, July 2001, pp. 299-329; W. B. Rayward, Bureaucratic Organization of Libraries, Australian Library J. vol. 19, Aug. 1970, pp. 245-253, and W. B. Rayward, Librarian as Organizer and Manager, Proc. 16th Biennial Conf., 1971, Library Assoc. Australia , Sydney, 1972, pp. 61-71. 2. W. B. Rayward, Libraries as Organizations, Col- lege and Research Libraries , vol. 30, July 1969, pp. 312-326. 3. Library use of micro lm technology was much discussed and experimented with in the 1930s. Document reproduction technology was a driv- ing force in the creation of the American Docu- mentation Institute; see I. Farkas-Conn, From Documentation to Information Science: The Begin- nings and Early Development of the American Doc- umentation Institute, Am. Soc. for Information Science, Greenwood Press, New York, 1990. 4. J. Shera, The Foundation of the American Public Library , Univ. of Chicago Press, Chicago, 1949, and S.H. Ditzion, Arsenals of a Democratic Culture: A Social History of the American Public Library Movement in New England and the Middle States from 1850 to 1900, Am. Library Assoc., Chicago, 1947; A. Black, A New History of the English Public Library: Social and Intellectual Contexts, 1850 1901, Leicester Univ. Press, London, 1996. 5. W.A. Wiegand, The Politics of an Emerging Profes- sion, American Library Association, 1876 1917, Greenwood Press, New York, 1986; see also Wie- gand s Irrepressible Reformer: A Biography of Melvil Dewey, ALA, Chicago, 1996. 6. G.T Stevenson, A.L.A., Encyclopedia of Library and Information Science , A. Kent and H. Lancour, eds., Marcel Dekker, New York, 1968, vol. 1, p. 279. 7. For information on the ALA Publishing Web site, see 8. J. Metcalfe, Information Retrieval, British & American, 1876 1976, Scarecrow Press, Metuchen, N.J., 1976. 9. F. Miksa, The Making of the 1876 Special Report on Public Libraries, J. Library History , vol. 8, Jan. 1973, pp. 30-40. 10. C.A. Cutter, Rules for a Printed Dictionary Catalogue, Public libraries in the United States of America: Their History, Condition and Management, special report, Dept. of the Interior, Bureau of Education. Part II. Govt. Printing Of ce, Washington, D.C., 1876. 11. A Classi cation and Subject Index for Cataloguing and Arranging Books and Pamphlets of a Library Amherst, Mass., 1876. 12. W.B. Rayward, The Early Diffusion Abroad of the Dewey Decimal Classi cation: Great Britain, Aus- tralia, Europe, Melvil Dewey: The Man and the Classi cation , G. Stevenson and J. Kramer- Greene, eds., Forest Press, Lake Placid, N.Y., 1983, pp. 149-173. 13. Dewey Decimal Classi cation and Relative Index (DDC21), J.S. Mitchell et al., eds., Forest Press, Albany, N.Y., 1996. See also http://www.oclc. org/dewey/index.htm. 14. J. Rosenberg, The Nation s Great Library: Herbert Putnam and the Library of Congress, 1899 1939, Univ. of Illinois Press, Urbana, Ill., 1993, p. 44. 15. I am grateful to Rodney Brunt of Leeds Metropol- itan University and a former member of the Joint Steering Committee for the Revision of AACR for guidance in this complicated area. See his Organising Knowledge: Catalogues, Indexing and Classi cation as a Re ection of Changing Needs, A History of Libraries in Great Britain, vol. 3, A. Black and P. Hoare, eds., Cambridge Univ. Press, Cambridge, UK, to appear. E. Svenonius, The Objectives of the Catalog and the Means to Reach Them, Conceptual Foundations of Descrip- tive Cataloging, E. Svenonius, ed., Academic Press, San Diego, Calif., 1989, p. 2. 16. For detailed and authoritative discussion of these matters, see the essays in Svenonius, Conceptual Foundations. 17. See also E.J. Hunter and K.G. Blackwell, chapter 2, History: Chronological Chart, Cataloguing, 3rd ed., revised by E.J. Hunter, Library Assoc., London, 1991, pp. 5-23. 18. Osborn s critique of the complexity and cost of using the existing by-now-labyrinthine cataloging rules and his plea for their simpli cation was high- ly in uential in cataloging code reform. 19. Lubetzky s critique and later a draft code he pre- pared at the Library of Congress were instrumen- tal in the reconceptualization of the rules and the principles underlying them. 20. The most recent edition listed on the ALA Web site ( contains the original code and revisions plus Amendments 1999 and 2001. 21. E. Svenonius, The Intellectual Foundation of Information Organization, MIT Press, Cambridge, Mass., 2000. 22. List of Subject Headings for Use in Dictionary Cata- logs, prepared by a committee of the American Library Association, The Library Bureau for the ALA, Boston, 1985. 23. Library of Congress Subject Cataloging Division, Subject Headings Used in the Dictionary Catalogs of the Library of Congress, Government Printing Of ce, Library Branch, Washington, D.C., 1897; Library of Congress Catalog Division, Subject Head- 14 IEEE Annals of the History of Computing A History of Computer Applications in Libraries
Page 12
ings Used in the Dictionary Catalogues of the Library of Congress, 2nd ed., GPO, Library Branch, Wash- ington, D.C., 1919. For the most recent informa- tion, see 24. See J. Rosenberg, chapter 3, Organizing Services for Librarians, The Nation s Great Library: Herbert Putnam and the Library of Congress, 1899 1939, Univ. of Illinois Press, Urbana, Ill., 1993, see also pp. 59 and 143. 25. Cataloging in Publication Celebrates 30th Anniver- sary, Library of Congress Information Bull. , May 2001, 26. A Catalog of Books Represented by Library of Con- gress Printed Cards Issued to 31 July 1942 , Edwards Bros., Ann Arbor, Mich., 1942 1946, 167 vols. 27. The National Union Catalog of Pre-1956 Imprints, A Cumulative Author List Representing Library of Congress Printed Cards and Titles Reported by Other American Libraries, Am. Library Assoc., Mansell, London, 1968 1981, 754 vols. 28. Union List of Serials in Libraries of the United States and Canada, 5 vols., E. Brown Titus, ed., under the sponsorship of the Joint Committee on the Union List of Serials with the cooperation of the Library of Congress, H.W. Wilson Co., New York, 1965. 29. J. Flexner, The Circulation of Books in Public Libraries , Chicago, Am. Library Assoc., 1926; C.P.P. Vitz, Circulation Work, revised ed., Chicago, Am. Library Assoc., 1927; C.H. Brown and H.G. Bous eld, Circulation Work in College and Universi- ty Libraries, Chicago, Am. Library Assoc., 1933. 30. C. Payne et al., The University of Chicago Library Data Management System, Library Quar- terly , vol. 47, no. 1, Jan. 1977, pp. 1-22. 31. J. Meyer, Notis: The System and Its Features, Library Hi Tech , vol. 3, no. 2, 1985, pp. 81-90; see also V. Veneziano and J. Aagaard, Cost Advantages of Total System Development, The Economics of Library Automation, J.L. Divilbiss, ed., Univ. of Illinois, Graduate School of Library Science, Urbana-Champaign, Ill., 1977, pp. 133-144. 32. For an account of the Library Bureau, see W. Wie- gand, Irrepressible Reformer, A Biography of Melvil Dewey , ALA, Chicago, 1996, pp. 70-71, 235-241, and 371. 33. The series of surveys in the Library Journal are instructive in this respect. See, for example, J.R. Matthews, Unrelenting Change: The 1984 Auto- mated Library System Marketplace, Library J., vol. 110, 1 Apr. 1985, pp. 31-40. By 1993, the survey was broken into large and small system categories: F.R. Ridge, Automated Systems Marketplace 1993, Part 1: Focus on Minicomputers, Library J., vol. 118, no. 6, 1 Apr. 1993, pp. 53-63, and Auto- mated Systems Marketplace 1993, Part 2: Focus on Microcomputers, Library J . vol. 118, no. 7, 15 Apr. 1993, pp. 50-55. See also J.E. Rush, The Library Automation Market: Why Do Vendors Fail? A Histo- ry of Vendors and Their Characteristics, Library Hi Tech , vol. 6, no. 23, 1988, pp. 7-33. 34. W. Wiegand, Politics of An Emerging Profession, and W.B. Rayward, Melvil Dewey and Education for Librarianship, J. Library History , vol. 3, 1968, pp. 286-312. 35. C. Churchwell, The Shaping of American Library Education , ALA, Chicago, 1975. 36. J.V. Richardson, The Spirit of Inquiry: The Graduate Library School at Chicago, 1921 51, Am. Library Assoc., Chicago, 1982; and W.B. Rayward, Research and Education for Library and Informa- tion Science: Waples in Retrospect, The Library Quarterly , vol. 56, Oct. 1986, pp. 348-359. 37. F.W. Summers, Accreditation and the American Library Association, A Background Paper Prepared for the Executive Board of the American Library Association, summers_print.html. 38. D. Reynolds, Library Automation, World Ency- clopedia of Library and Information Services, 3rd ed., ALA, Chicago, 1993, p. 470-471. My text adapts the section headings used by Reynolds. 39. E.C. Berkeley, Giant Brains or Machines That Think John Wiley & Sons, New York, 1949. 40. H. Bryan, American Automation in Action, Library J. , 15 Jan. 1971, pp. 189-196. 41. E. Mason, The Great Gas Bubble Prick t; or, Com- puters Revealed by a Gentleman of Quality, Col- lege and Research Libraries , May 1971, pp. 183-196. 42. D. Reynolds, Library Automation: Issues and Appli- cations, Bowker, London, 1985. 43. F. Kilgour, Ohio College Library Center, in Ency- clopedia of Library and Information Science , vol. 20, A. Kent, H. Lancour, and J. Daily, eds., Marcel Dekker, New York, 1977, pp. 346-347. 44. See the OCLC Web site at about/. W. Boyd Rayward is a research professor in the Graduate School of Library and Information Sci- ence at the University of Illinois. He is currently Leverhulme T rust Visiting Professor at Leeds Met- ropolitan University. He was educated in Australia and the US. His PhD is from the University of Chicago. He has published widely on the history of libraries and the international organization and dissemination of information. Readers may contact Boyd Rayward at B.Rayward@ For further information on this or any other com- puting topic, please visit our Digital Library at April June 2002 15

About DocSlides
DocSlides allows users to easily upload and share presentations, PDF documents, and images.Share your documents with the world , watch,share and upload any time you want. How can you benefit from using DocSlides? DocSlides consists documents from individuals and organizations on topics ranging from technology and business to travel, health, and education. Find and search for what interests you, and learn from people and more. You can also download DocSlides to read or reference later.