PART 2

INFORMATION NEEDS AND NEW TECHNOLOGIES

CHAPTER SEVEN

INFORMATION AS A COMMODITY

We have suggested elsewhere in this study that libraries and the books they contain are products of a culture of print. Until very recently, scholarly information needs have been served almost exclusively by the technology of printing developed in Europe in the late Middle Ages.[1] In all of its essential characteristics, that technology was simply a different means of doing what had been done in Europe for more than a millennium: the recording of text (and visual images, musical notation, and so on) on sheets of material, whether parchment or paper. The great virtue of printing, of course, was that unlike manual copying it permitted multiple copies of a text to be produced with almost trivial ease.[2] It permitted the creation of the first media products with the characteristics described in Chapter 6: relatively high cost associated with producing the first copy of the product, relatively low cost associated with producing each subsequent copy. It also ensured that all copies of the text could be virtually identical, which gave it considerable advantages over manual copying in that the variant readings resulting from scribal error or the effects of deliberate emendation or interpolation could be eliminated.[3]

The technology of printing, of course, has other characteristics that are either virtues or limitations depending on one's perspective or on the specific information needs at issue. Its principal characteristic, which it shares with manual copying, is that it produces a physical object---a book, a journal, a magazine, a newspaper---containing a fixed, immutable text permitting the reader to interact with it only in limited ways, as contrasted with conversation, for example;[4] the information flow, that is, is one-way. The information, moreover, is universal---it is not tailored to the specific information needs of particular readers---and in almost all instances is arranged in some kind of linear sequence. Absent the kinds of devices developed in a print culture to facilitate access to information in printed form (for example, tables of contents and indexes), readers with particular needs are compelled to ferret out the pertinent information by systematic reading. The means of facilitating access to printed information are therefore to be contrasted with those permitted by electronic information technologies, which afford almost instantaneous, random access to any portion of the text.[5]

FROM A PRINT TECHNOLOGY TO AN ELECTRONIC TECHNOLOGY

Inevitably, the medium of print in which text-based information has traditionally been disseminated has shaped one's most fundamental understanding of the nature of text-based discourse and communication. To some extent, that is, terms rooted in the nature of the medium---print products---rather than in the nature of the resource---the intellectual content---shape the discourse. The nonmateriality of text-based information as exemplified by the technologies of the late 20th century, in contrast, entails different terms. Texts are no longer necessarily immutable; rather, they are dynamic. Given that characteristic, interactivity is eminently possible; readers can alter the received texts and reformat the information they contain to suit individual information needs by means of various scanning and sorting mechanisms. Late 20th century technologies, in short, uncouple the material object---the book, the journal, the newspaper---from the intellectual content---the information the material objects contain.

Such fundamental changes in our perceptions of the nature of information were described a decade ago by Harlan Cleveland in his essay in The Futurist:

[w]e have carried over into our thinking about information ... concepts developed for the management of things---concepts such as property, depletion, depreciation, monopoly, market economics.... The inherent characteristics of information now coming into focus give us clues to the vigorous rethinking that must now begin:

Information is expandable.... [T]he facts are never all in....
Information is compressible. Paradoxically, this infinitely expandable resource can be concentrated, integrated, summarized---miniaturized, if you will---for easier handling....
Information is substitutable. It can replace capital, labor, or physical materials....
Information is transportable---at the speed of light....
Information is diffusive. It tends to leak....
Information is shareable.... [I]nformation by nature cannot give rise to exchange transactions, only to sharing transactions. Things are exchanged: if I ... sell you my automobile, you have it and I don't. But if I sell you an idea, we both have it....

So it has to be a mistake to carry over uncritically to the management of information those concepts that have proved so useful during the centuries when things were the dominant resources and the prime objects of commerce, politics, and prestige. These concepts include scarcity, bulk, limited substitutability, trouble in transporting them, and the notion of hiding and hoarding a resource....

Furthermore, Cleveland observes, these changes in our understanding of the nature of information will have widespread consequences, reaching beyond print and libraries, into political economy and the law.

In political economy, won't the concept of market "exchange" have to take account of the fact that more and more of our economic activity now consists of what are by nature "sharing" transactions?...
In law, how should we adapt the concept of property in facts and ideas when the widespread violation of copyrights and the shortened life of patent rights have become the unenforceable Prohibition of our time? Aren't we going to have to invent different ways to reward intellectual labor that are compatible with a resource that is both diffusive and shareable?[6]

One observer has suggested that the technology of print, as contrasted with the information technologies of the late 20th century, has had even more fundamental effects on the social and intellectual experience of modern society:

Television is not symbiotic with literature in the way that print was. Literary values---authors, great works, deep meanings---fitted hand-in-glove with print, but television both weakens literacy (the skill on which literature depends) and undercuts literature's basic function. The replacement of the printed word by the image and the voice substitutes immediate, powerful one-dimensional pictures and simple continuities for the ironies, ambiguities, and complex structures fostered by print and idealized in literature. Where the fixity of the printed book encouraged the conception of masterworks and permanent truths so central to literature, databases in which items easily intermix and television programs that flicker fleetingly past make literary ideas like originality, form, and permanence seem quaint ideas of another age.[7]

Literature, that is, is perhaps "so much a product of print culture and industrial capitalism, as bardic poetry and heroic epic were of tribal oral society, that, like chivalry in the age of gunpowder, it will simply disappear in the electronic age."[8] As a kind of discourse, it is an expression of the technology of print, and new technologies may ultimately spawn a new kind of discourse with fundamentally different features.

The characteristics of print therefore have had profoundly important implications for the storage and dissemination of information, including scholarly information, and thus for the most fundamental aspects of the processes of scholarly activity and communication. The essential distinguishing characteristics of research libraries are themselves expressions of the technology of print, as are those of the various publishing industries that have grown up over the past half millennium. Because printing produces physical objects, libraries, in fulfilling their role as participants in the process of scholarly communication, have accordingly acquired certain fundamental characteristics determined by the nature of the technology and appropriate to the nature of their role in the process of scholarly communication, as currently defined. Libraries have been and continue to be physical spaces where printed materials are collected, classified, and stored in a way that facilitates access to them. They contain spaces where readers can consult materials in the collection rather than take them elsewhere,[9] and the proximity of members of the library staff---specialists in information management---similarly facilitates access. (Members of the library profession have observed that theirs is one of the few professions identified with a particular facility; librarians ordinarily work only in libraries, whereas attorneys are not exclusively identified with any one kind of facility.) There are constraints of space and time (and of other kinds) resulting from the nature of the prevailing information technology that limit readers' access to scholarly information: the only such information they have immediate access to is local information that the local research library has been able to acquire, and they have access to it only when the library is open.

The technology of printing, further, has defined the role publishers play in the process of scholarly communication. Indeed, publishers became players in the first instance in part because of their professional expertise in the technical aspects of publishing and because of the economies of scale resulting from centralization and specialization in that function. It is important to observe, however, that publishers make other critical contributions as well, including coordinating the peer review process, termed the "gate-keeping" function by some observers. They solicit opinion regarding the quality of manuscripts proposed for dissemination and make judgments about the importance of their contribution to scholarship. The information technologies of the late 20th century, some argue, may transform publishing in that the publishers' role in the actual process of dissemination may change; there will presumably continue to be a need for the gate-keeping function, however, and the new technologies will not obviate that need.

For the moment, the technology of print requires that both publishers and libraries anticipate demand. It is cost-ineffective for publishers to print either too few or too many copies of a particular title. Given the logistical complexities of disseminating information in printed form (printing and binding, transportation of the resulting material objects to clients, whether individuals or institutions, libraries or vendors), publishers project demand in an attempt to ensure that titles will be available to clients at the moment when they are in need of them.

A similar set of assumptions underlies the acquisitions practices of research libraries. Scholarly publications often go out of print quickly; research libraries therefore wish to acquire them as soon as possible after their publication so as to ensure access to the information they contain. Moreover, the same concern about the ready availability of material governing publishers' behavior is also operative here. Although interlibrary loan services afford access to material owned elsewhere, for many readers such services seem inefficient. The most desirable option is ownership, and for that reason acquisitions have attempted to be as comprehensive as institutional resources have permitted, so as to build a self-sufficient collection with all the advantages of ready access it entails.

This model has been described as the "just-in-case" model;[10] libraries acquire materials in anticipation of readers' needs, in accordance with an assumption that a particular reader may at some future time wish to consult a particular volume. Given the prevailing information technology, this model in many respects has indeed been the most appropriate. It may be added that for many scholars a rich, self-sufficient collection of millions of volumes has another, critically important advantage: it permits serendipitous, potentially interesting discoveries that result when scholars chance upon titles while browsing in the stacks.

The information technologies of the late 20th century compel us to rethink the most basic assumptions underlying the processes of research and scholarly communication. They affect not only the nature of scholarly activity in the first instance but also the nature of the contributions of other agents---publishers of scholarly materials, academic booksellers, research libraries---participating in the process of scholarly communication.

Some of these technologies have already been effectively employed to streamline and improve various library functions, especially acquisitions, cataloging, and circulation. The cataloging function in particular has been transformed as a result of the new technologies. As we saw in a previous section of this study, individual institutions contribute catalog copy to databases maintained collaboratively and can retrieve and easily reformat records contributed by other institutions, so that there is a degree of uniformity previously unachievable. The automation of cataloging has had the added beneficial effect of permitting collaborative collection development; the catalog copy in the database serves as a record of other institutions' holdings, so that individual institutions, building on local strengths, can make informed decisions about acquisitions that do not replicate decisions made elsewhere within the consortium.[11] While the implementation of collaborative collection development schemes is not perfect, the new technologies make these efforts feasible on a scale that would have been impractical in an earlier era.

THE RECONFIGURATION OF SCHOLARLY COMMUNICATION

Until very recently, however, the new technologies have been employed simply to automate existing functions. They hold enormous potential for a much more fundamental reconfiguration of the entire process of scholarly communication and for libraries' role in that process. Nina W. Matheson, professor of medical information and director of the Welch Medical Library at Johns Hopkins, has written of the "de-materialization" and "de-institutionalization" of information:[12] it need no longer be made available to us in printed, immutable forms collected by libraries, where access to the universe of scholarly information is governed by local constraints affecting the size of the acquisitions budget and the physical plant where the print products are stored. Rather, just as automatic teller machines have revolutionized banking (an individual's banking needs are now met by machines that are located everywhere, function 24 hours a day, and afford access to global information),[13] so the information technologies of the late 20th century facilitate access to an ever-larger universe of scholarly information beyond that contained in one's own local research library.

To anticipate the content of much of the remainder of this section, we might summarize some of the characteristics of the new technologies. There is, first, the possibility of ever-greater bibliographic control over the professional literature. The automation of cataloging and the availability of catalog records on the Research Libraries Group's Research Libraries Information Network (RLIN) afford scholars remote access to an extraordinarily rich store of information about the existence and location of scholarly materials held elsewhere and bibliographic information on those materials. As various observers have suggested, however, library catalogs ordinarily contain complete bibliographic information solely on the monographic literature. Since the most current information is often contained in the serial literature and intellectual advances often occur on the basis of interpretations argued in that literature, especially in particular disciplines, it is a limitation that the bibliographic record ordinarily does not extend to the level of the individual article.[14] Increasingly, however, there are bibliographic services available in electronic form that index and abstract the serial literature. By no means are there adequate services of this kind in all disciplines, and the existing ones are expensive to use. Nonetheless, scholars in some fields are certainly closer than before to being able to achieve relatively complete bibliographic control over the literature of their disciplines.

Increasingly, the technologies are being applied not solely to problems of access to information about information but to problems of assembling and ordering the primary information itself and of providing access to it. In all disciplines, including the humanities, the advantages to particular kinds of scholarly activity of the availability of electronic versions of texts and data are clear. As many observers have suggested, such databases are dynamic phenomena; because "the facts are never all in," in Harlan Cleveland's words, it is useful to be able to assemble them in a form that allows one to make additions and refinements easily and manipulate the texts or data in various ways. For many scholars the new means of storing text not only facilitates traditional kinds of research but also permits one to ask new kinds of questions that would have been literally impossible to pursue with text and data in printed form. And although the availability of the full texts of secondary literature---works of synthesis and interpretation---in electronic form is still a very recent phenomenon, there can be no question that such material will increasingly be available.

The potential utility for libraries of these means of capturing text-based information is obvious. "Information is transportable---at the speed of light," Cleveland has written. The dematerialization of information may ultimately permit specialization in collection development and collaborative collection development in that the full texts of materials not owned locally would be readily available from other institutions within the consortia to which individual libraries belong. It would permit the ideal of resource sharing, which depends upon more-or-less immediate access to materials owned elsewhere, to become a reality. That these technological developments are occurring at a time when resources do not permit the traditional model of the self-sufficient library to be sustained is perhaps fortuitous, perhaps not.

One must not underestimate the difficulties involved in realizing such a reconfiguration, however. Some of them will be explored in greater detail in the chapters that follow. There are, first, enormous cost implications. The sharing of information in electronic form assumes greatly upgraded computing and telecommunications networks, and many institutions will simply not be in a position to absorb their share of the capital expense. Moreover, electronic versions of material challenge some of the most fundamental assumptions underlying copyright legislation. There are, further, issues of standardization. Over the course of the past half millennium, we have become accustomed to using text-based information in printed form and are conversant with its conventions, while text-based communication in an electronic environment will require different conventions and protocols that have not yet been settled upon.

It is important to add that print products have some considerable advantages over electronic products, especially for certain purposes. Alvin Kernan has observed that print yields particular kinds of text-based discourse, and although his observation pertained principally to literature, it might be extended to include certain kinds of scholarly discourse as well. The utility of electronic versions of primary texts and data is, for many scholars, unarguable. Using search engines one can readily locate references and patterns in the texts or data, conduct particular kinds of analyses, and retrieve virtually all the pertinent material. Works of synthesis and interpretation based on the underlying data or texts, however, especially in the humanities, might share with literature some of the ambiguities, deep meanings, and complex structures for which print is a more appropriate medium. Print's suitability to some kinds of scholarly purposes should not be underestimated, and one needs to be attentive to differences among disciplines. Humanists work differently from scientists and may therefore have some different kinds of information needs. A great advantage of the present situation is that a choice between print products and electronic products need not be made, at least on technological grounds. Cost factors will, however, force almost all institutions to make certain choices irrespective of technological constraints or possibilities. Such choices can be made on the basis of the suitability of various options to specific scholarly objectives.[15]

These late 20th century technological developments have still another implication for libraries. Once the preeminent information service for research and scholarly communication, the library is now complemented by an entirely new set of information services provided by computing, each being the expression of a particular technology.

Librarians have experience in thinking about the nature of information as a commodity, about how one establishes its authenticity and orders and classifies it so as to facilitate access. Some institutions have undertaken an administrative re-organization that may reflect their belief in the importance of integration of computing and library services.[16] Information is regarded as one of the institution's most important resources, like its financial and human resources. At some universities, accordingly, a single vice president for information services, comparable in stature and in the scope of his or her responsibilities to the vice presidents for finance and human resources, has responsibility for both the library and the computing and telecommunications services. Such an administrative structure permits decisions about the allocation of resources for information services to be made in a centralized, coherent way.

We turn now to a more complete consideration of some of these developments and their importance both for scholarly communication and for libraries' role in the process. To many observers it is clear that we are in a period of transition. What many envision, ultimately, is a situation in which the full range of information services and products would be available to the individual end-user at his or her own workstation:[17] fully machine-searchable bibliographic services that abstract and index the existing printed literature;[18] databases of primary material; the full, machine-searchable texts of works of analysis with primary material integrated with it through sophisticated windowing and hypertext functions (these would lead the reader to the entire literature and substantiating primary material on any point he or she wishes to pursue); downloading and print options that would permit the end-user to excerpt and reorder portions of the full range of material available and print it locally; flexible protocols for communicating among heterogeneous systems, what one member of the library profession has called "systems with rich and varied access vocabularies [that address the] individual needs, sophistication level, and viewpoint of the user."[19] One cannot know precisely where in the transition we presently are, though we are surely much closer to the beginning than the end. The objective in Part 2 is to describe some elements of the transition and assess their potential utility.[20]