Boston, Massachusetts
May 17-19, 1995

Realizing Digital Libraries

Libraries, Printing, and Infrastructure: A Historian’s Perspective

Bruce E. Seely, Associate Professor of History
Michigan Technological University

I want to thank Duane Webster and the members of the Association of Research Libraries for the invitation to be here this morning. I am delighted to join you in your 126th Membership Meeting. But as you all sit back to hear speakers contemplate the prospect of “Realizing Digital Libraries,” I have to open with a disclaimer: I am not an expert on libraries, except from the perspective of a user. From that position, I want to thank to a number of you who represent institutions that have I visited and worked in (at last count more than 30 different university archives and libraries at ARL member institutions) during the course of my career as an academic historian. I am grateful to all of you and your staffs who make possible the kind of work that historians and other scholars do. Beyond that first-hand experience, I bring to this talk perhaps a little more awareness than many academics have of the circumstances that confront libraries because, as Secretary of the Society for the History of Technology for the past six years, I have had the opportunity to participate in the meetings of the American Council of Learned Societies. In those forums, libraries have often been part of the discussion about the state of academia and the humanities. I will say, however, that only the need to prepare my remarks today led me to do something I doubted I ever would — page through several volumes of the Journal of Library Automation and its successor, Information Technology and Libraries. This hardly qualifies me an expert in the library field, although it will not prevent me from finding something to say!

When Duane approached me about speaking to you, it was with the idea of talking about how historians have come to view technological change. Most of you are living through it, while historians of technology, like myself, study the general processes of technological change. My goal this morning will be to suggest a couple of general patterns to think about, and then see if I can draw some parallels between the development of large-scale electronic networks in libraries and the lessons that emerge by examining the engineering infrastructure on which our cities rest.

The “Printing Revolution”

Let me start by talking a little about the appearance of the printed book in the fifteenth century. This development is the product of a set of technological innovations that even most historians of technology — who usually avoid using such terms — acknowledge as revolutionary in scope. Indeed, standard assessments of current developments in communications often use the printing revolution as a starting point for comparisons. Your own president, Susan Nutter, opened a session at your 123rd meeting with the comment, “I think it is safe to say that everyone is aware that a communications revolution is underway that is as profound as the introduction of the printing press. . . .” It seems to me worth examining why events of some five hundred years ago merit the term revolutionary, for that word springs to the lips so easily today.

We can begin with a reminder that printing is not a single technological development, but the product of a series of interrelated technical improvements, some of them originating outside Europe. Rag paper, developed in China about 200 AD, had reached Europe by the 12th century. Ink that could adhere evenly to metal was developed from the linseed oil varnish of the Flemish school of painters during the 15th century. Movable type was the key development, allowing for the fairly rapid assembly of pages of text — unlike laboriously carved wooden locks for printing, as used in China. The type had to be cast from a metal that did not contract as it cooled, yet it also had to be inexpensive, since many, many letters were needed. The famous 42-line Gutenberg Bible had about 2,750 letters per page. Since most print shops would have six pages in various stages of composition, print, or disassembly at any one time, about 100,000 letters had to be on hand.

But for movable type to be adopted, a number of other components had to be developed. These included a composing stick to hold the letters while the printer assembled the text, one letter at time, one line at a time; then a chase that locked the lines together to form a page. The press itself had to be designed to apply even pressure, quickly and smoothly. I think you get my point, which is that we are talking about working out an array of small, but serious technical details. Pictures of early print shops show small foundries for casting type, large composing tables, racks of letters, shelves of printed pages waiting to be assembled and bound, and adjacent paper-making operations with huge racks holding drying paper. Printing is a relatively straightforward technology, at least by our lights, but we should never make the mistake of assuming that it is simple. Nor should we forget that printing constitutes an interconnected system, and the successful technology required solutions for a range of large and small technical problems. This is true of all technologies — they tend to exist not as individual machines, but rather as systems.

Attention to systems makes clear another point, that this new system completely displaced the older method of hand-copying. Indeed, only rarely is it possible to substitute a new component into a process without significantly adjusting, or even abandoning, the older method. The strength of the linkages among the pieces of a system often become clear only when attempts at substitution are made. Moreover, components of this new system of printing reached well beyond the walls of the print shop, for technologies are inextricably bound up in the social and political, cultural and economic arrangements of our societies, as well. Evidence of this becomes quite clear when we shift our attention to the product of printers — the book. Rather quickly, the sheer number of books that appeared had a significant impact on European society. Reliable estimates suggest 40,000 titles appeared between 1450 and 1500, perhaps a total of 10 million books. Approximately half of those are still available in libraries. This suggests one of the most important ramifications of printing — its relative permanence.

But the impact of these books and the system that produced them was much more far-reaching, as becomes clear in a superb book by historian Elizabeth Eisenstein, The Printing Revolution in Early Modern Europe, published in 1983. Among the changes that followed in the wake of printing were new developments in scholarship, including a reduction in travel to visit collections of manuscripts in distant libraries, learning without masters, and the ability to compare ideas, identify contradictions, and create new syntheses. She noted changes in the presentation of information in books, as compared to manuscripts, that included tables of contents, footnotes, and illustrations. Also emerging was a sense of authorship and concerns about priority and individual recognition, a trend reflected in the appearance of the title page. Taken together, Eisenstein suggested, these adjustments led to a more general willingness to question the authority of ancient wisdom and established institutions. Moreover, textbooks became the keys to transmitting knowledge that was more codified and organized, as knowledge was corrected and improved from one generation to another. Eisenstein argued that these changes in the presentation of information created a culture of printing that carried over into society in many ways. The adoption of alphabetical order and the index exhibited a special emphasis on rationalization and codification — on order, in other words. “Changes in format,” she commented, “might well lead to changes in thought patterns.”

Eisenstein also identified the special importance of the printed word for history, using the term fixity to describe this point, for once committed to print, events are set in time. She suggests this development was absolutely crucial to the development of science in the West, with its emphasis on cumulative developments and the transmission or ideas between generations. Similarly, neither the Italian Renaissance nor the Protestant Reformation represented the first re-discovery of ancient learning, nor the first schism and protest against the church at Rome, but the existence of printing presses insured that these events became fixed in time. The emergence of nationalism also was connected to printing, since one could now build non-local identifications to causes and to kings. A central element was the construction of language walls around regions using printed vernacular text. All of this rested on the permanence afforded by the printed book, widely distributed. Finally, Eisenstein noted that not everyone was thrilled at this new technology. She repeated a story, possibly apocryphal, about Gutenberg’s financier, who set out with a dozen copies of the new Bible

to see for himself how he could best reap the harvest of his patient investment. And where did he turn first of all to convert his Bibles into money? He went to the biggest university town in Europe, to Paris, where 10,000 or more students were filling the Sorbonne and the colleges. And what did he, to his bitter discomfiture, find there? A well-organized and powerful guild of the booktrade . . . founded in 1401. . . alarmed at the appearance of an outsider with such an unheard of treasure of books; when he was found to be selling one bible after another, they soon shouted for the police, giving their expert opinion that such a store of valuable books could be in one man’s possession through the help of the devil himself and John Fust had to run for his life or his first business trip would have ended in a nasty bonfire.

There is much more in Eisenstein’s wonderful survey on the impact of printing. But let me draw a couple of special lessons from this historical skip backwards in time. First, it becomes very clear why we should think about technological systems, since we can miss many aspects of the technology by focusing only on the narrow definition of technology as hardware. A broader net can often let us see ways in which technology and society interact to produce not only new ways of doing things, but also new ways of thinking and approaching problems. Second, as the last story makes clear, resistance to these systems is a common theme. So, too, however, is its corollary, for much technological innovation is driven by a deep-rooted enthusiasm (at least in the West) for technology. We can’t go into the roots of this development today, but it remains a crucial factor that drives the development of technology, every bit as important in explaining innovation as concerns about profit and gain. Indeed, use of the label “printing revolution” betrays this enthusiasm, and may lead too easily to the conclusion that technologies are independent of society, rather than the product of social activity. The point for us to remember is the willingness in our culture to get excited — even over-excited — about new technologies.

Thirdly, we need to realize that most of the serious consequences of printing and books were unexpected and unanticipated. They show up clearly in the twenty-twenty hindsight of history, but none of them were predicted by those involved at the time. No one at the time envisioned the huge libraries of our day, or the explosion of information. It was a wrenching shift for civil and ecclesiastical authorities who realized that printing challenged and then usurped their easy control over information; it was not long before commercial publishers realized the financial opportunities afforded by selling titles placed on the Roman Church’s index of prohibited works! And not all of these unexpected consequences were automatically good, although enthusiasm often leads to dismissing some of the downside effects. Eisenstein, for example, notes that those people in Europe who were outside the culture of printing — people like the Basques in northern Spain, for example — have fallen out of the pages of history since they had no written language and so have not become fixed in time through the printed word.

Civil Infrastructure

Eisenstein’s work is especially useful for my purpose today because the changes are not only so clear, but also because they can be easily appreciated by librarians. Printing is, after all, the technological development that is behind the books and journals that explain the existence of your institutions. It makes it easier, I hope, to grasp the consequences of technological change when the situation is close to home. But now let me turn to another historical development that I also think has relevant lessons for us, although it may seem further away. Let me focus for a few minutes on the development of the complicated engineering systems upon which our urban existence has come to depend. The products and creations of American engineers are often labeled infrastructure today. These developments are worth examining because they shift our attention from the broad topic of changing technology to the experience of others in developing complicated technological systems.

The word infrastructure is used ubiquitously now, but it made its appearance in the language in the late 1940s and early 1950s, applied first to the permanent facilities necessary for the military-bases, airstrips, ports, dry docks, and rail lines. By the 1960s, the word was being adopted in an urban context and applied to the technological support systems underpinning modern urban society. But the systems themselves are quite older, for they began to appear in cities in the 19th century as a range of specific problems approached a certain magnitude. Philadelphia, for example, built the first waterworks beginning in 1799 after a series of epidemics were traced to polluted wells. In 1832, New York City launched a massive, $13-million water supply system that included a dam in the Catskills, a 40-mile long aqueduct, a bridge over the East River, and a 180 million gallon reservoir after a cholera outbreak and fire in the business district. Most large cities soon followed New York’s lead. But as they did so, sewage collection systems soon had to be constructed, ironically because the introduction of good water supplies led to a rising tide of waste water that existing disposal techniques (privies and open drains) could not handle. The first systems to collect sewage in this country were begun in Brooklyn in 1857 and Chicago in 1858; by 1870, most major cities had taken some steps in this direction. No city hatched as ambitious a plan as Chicago, which reversed the flow of the Chicago River so that its waste was not carried into Lake Michigan, the source of its drinking water. Instead, Chicago’s sewage flowed into the Illinois River, where it fouled other cities’ water supply! Later, cities began to build plants to treat sewage before dumping it back into rivers.

Other infrastructural developments in the 19th century included transit systems, beginning with horse cars in New York in 1832, Boston in 1856, and Philadelphia in 1858; electric streetcars after the first system was built in Richmond in the late 1880s; and in a few cases, subways, following New York’s opening of the first one after 1900. Later, gas lighting and then electric power systems, highways, and communications networks would be added to the maze of utilities on which cities depend.

These utilities are the technological foundation of modern life. We need only think about our present surroundings and how we got to this meeting to realize our dependence upon them. And increasingly, libraries, like the rest of academia, are growing similarly dependent on new infrastructure systems and networks. For this reason, I think analogies between urban and intellectual infrastructures can be informative. I will suggest some general conclusions about the development and impact of urban infrastructure, and then see if we can develop some parallels to recent developments in libraries.

First, even more clearly than in the case of printing, urban infrastructure networks are systems by definition, systems that are extensive and expensive, often covering broad geographic areas. They are complex, not only because of their geographic scale but also because of the size of the technical components and the hardware. But they incorporate much more than hardware. Most infrastructure networks came into existence through a complicated political process which significantly affected how they would be built, where they would run, how much they would cost, and who would control them. These social, political, and economic issues cannot really be separated from the technical factors. Moreover, these nontechnical components are every bit as important and should not be viewed as peripheral, even though engineers and planners have tended to consider the political and social factors to be sources of problems, delays, headaches, and obstacles to the implementation of technology. At the same time, we should recognize that the hardware is important, since such technical issues as the development of standards, reliability, maintenance, and obsolescence determine how well the system functions. Over time, these types of technical issues assume more importance.

Second, because of the size and complexity of these systems, and because of the nature of the technologies, control and authority over many civil infrastructure technologies has been entrusted to technical experts. This is not an example of technocracy as that word is often understood, for there has been no deliberate effort to subvert democratic styles of government. Rather, arguments about improved efficiency, the avoidance of corruption, and technical complexity have led government officials and Congress to trust experts in the planning and implementation of these systems. Always the buzzword was the promise of efficiency. Cities took control of many municipal services in the late 19th and early 20th centuries in part because they wished to have experts appointed through civil service examinations, not beholden to private corporations. They wanted to avoid situations like that which appeared in Washington, DC in the late 1870s, when cronies of Boss Shepherd won a contract to build the city’s sewer lines and constructed a network of collector mains that ran uphill! Likewise, highways and road building in this country came to be viewed as the domain of engineers, not politicians. Indeed, the goal of federal policy was to prevent political interference in the highway engineers’ work. But one result of this decision has been a limitation of democratic governance, so that the users of the systems often have not been involved in decision-making.

A third point to be drawn from the experience of developing urban infrastructure is that these systems, once constructed, quickly become invisible to us and soon are taken for granted. Indeed, they quickly become essential to our way of life. We expect them to work first time, every time. We assume electricity will always be available at an outlet, or water will always flow from the spigot. We tend to ignore these systems until a failure reminds us of both their importance and our tendency to blindly trust their reliability. The 1965 blackout in the Northeastern U.S., which inspired the film Where Were You when the Lights Went Out?, poked fun at this situation; we were less inclined to laugh about the flood in Chicago’s downtown a couple of years back, or the collapse of highway bridges on the interstate network. But these reactions all point to a basic expectation we seem to have — perfection is not only achievable, but required. We assume technology can easily do all we ask of it. And in this, we echo the enthusiasm of the experts charged with developing and overseeing the systems. That these expectations might be overblown is a conclusion rarely voiced.

A fourth consideration from an examination of civil infrastructure that resonates with the discussions of scholarly infrastructure concerns the developmental patterns of large engineering systems. These large technologies tend to be constructed and adopted unevenly. Put another way, not everyone gains access at the same time. Not surprisingly, the well-to-do, the politically powerful, and various elites tend to be served first, while those less well off come later, if at all. As a result, inequities are common in the spread of these systems. This is an important reminder of the social and political dimensions of these systems. It seemed perfectly understandable when electric utilities argued in the late 1920s that they could not economically expand service to outlying suburbs and into the countryside because of the cost. But a new approach, the inauguration of rural co-ops backed by loans from the Rural Electrification Administration in the 1930s, suggests how new ways of thinking, organizing, and funding systems can lead to different, more equitable results.

Finally, and perhaps most importantly, civil engineering infrastructure systems are very long-lived and enduring. Because of their cost and extent, they are not easily changed or altered after construction. In part for this reason, and in part because of the importance of the systems to modern life-styles, infrastructure systems exercise a significant influence upon patterns of subsequent development. Decisions about where and what type of system to build cannot be easily undone, and thus act to foreclose later possibilities. Obviously, planning and forethought should be essential elements in the development of infrastructure systems. Yet, in the United States at least, most efforts to think ahead might be characterized as “planless planning,” considering each system in isolation from other, even potentially competing systems. Thus, there have been many unplanned consequences of such systems. For example, we pushed ahead with massive highway and airport construction programs in the 1950s without asking how they would affect the railroads. As much as any other factor, this lack of attention to a broad view of infrastructure systems and their position in society accounts for the inability of these systems to serve our needs as well as they might.

Library Infrastructure and Technological Change

I hope you can sense some of the ways that these observations apply to libraries struggling to cope with technological changes that seem as all-encompassing as those wrought by printing 450 years ago. But let’s see if we can draw some lessons from history applicable to infrastructure systems in libraries. This effort will take me a little further from my own work, so at best I hope to strike some sparks rather than provide clear answers, much less prescriptions. But the tone of my remarks will be cautionary. The three themes I highlighted in talking about printing — technology as systems, the paired circumstance of resistance and enthusiasm, and the importance of unanticipated consequences — offer a framework for this discussion. They overlap with several of the comments just made about civil infrastructure.

I suspect that most of you fully appreciate — perhaps better than most people — that we must think about the digital technologies in libraries as systems. Whether it is OCLC or the ability provided by the Internet to access remote card catalogs, librarians recognize that we cannot talk about computer terminals, or even the Internet, in isolation; we have to consider all of the technical and non-technical components in order to deal with the changes facing libraries. The systems we are examining include, at least, FAX machines, telephone systems, software, issues of copyright, university budgets, politicians, university presses and publishers, and even, perhaps, the scholars using libraries. It is the complexity of the systems we are talking about that accounts for my cautionary tone here. Civil engineers acquired experience with systems over a span of time, beginning small and working their way up. As they gained confidence, it was easier to move toward larger projects, but, even so, many of these didn’t tax their technical and their managerial abilities. Given the amazingly rapid development of technology today, I wonder whether we are fully involving all of the parties we should. In a paper on electronic journals, Jean-Claude Guedon captured something important.

. . . in the last four or five years, we have essentially witnessed a flurry of experimental designs for scholarly electronic publishing. On the upside, these were the result of a good deal of volunteer work, patchwork financing, personal enthusiasm, sometimes laced with utopian thinking and a dash of impatience with the old ways, inventiveness, and creativity . . . . On the downside, this mildly chaotic emergence of hundred plus electronic journals in the last five years has not lent itself easily to comparative analyses and syntheses. There may not be one best way to design scholarly electronic journals, but with what we have at hand, it would be difficult to select a few “better ways.”

Clearly, we must not lose sight of technology as broadly defined systems.

As for the tension between resistance and enthusiasm, it is here that we come to the crux of the current situation libraries confront. Certainly there is a pool of resistance, a group of critics who feel libraries are moving too quickly to embrace an electronic world. I think I can imagine how many of you would react if Nicholson Baker or Clifford Stoll were speaking to this group about their opinions on where libraries are heading. That some of their arguments seem based on aesthetic sensibilities or personal preference makes it easier to dismiss them as Luddites — simple opponents of progress who do not understand the reality of shrinking budgets, rising costs, and demanding clients and administrators. Yet I would suggest that in pointing to the opposite of resistance, namely enthusiasm, these two authors and others identify a serious difficulty, perhaps one much more important than meets the eye at first glance. Ours is a culture that clearly favors new technologies. Yet I suspect that many of you do recognize some validity in the complaints that Stoll raises; that there are opportunity costs in moving toward electronic environments that take resources that could be used for other things; that there are serious questions to be answered about the archiving of electronic records; and coping with continuing rapid technological change in formats and hardware. I am more concerned about the tendency to dismiss these as minor problems, or as items not worth worrying about. Many of us can filter out the grandest claims and oversold promises, but are we still prone to assume that technological fixes hold all of the keys to problems? Yet the record of technological change in the past should make us sensitive to promises that are too good to be true.

I wonder, however, given my brief review of some of the library literature. In 1966, the Stanford University Library Director quoted the chair of his library automation committee, who noted that “glowing published reports sometimes reflect theoretical possibilities rather than practical considerations. Stanford, in designing a system, will place performance and service ahead of glamour.” Sound advice, I think, but I wonder how often it has been taken. Many books about library automation point out that ego, institutional politics, the desire to be on the cutting edge, financial circumstances and opportunities, and other factors are the real keys to automation decisions. I suspect that we should be especially aware that technological change in many libraries is driven by a swell of enthusiasm for the newest technological prospects.

Consider, for example, the argument made in a little book entitled The End of Libraries. James Thompson opened with a long list of the problems facing libraries in the early 1980s: the doubling of collections every 16 years (there were about 30 million titles by 1980); the fact that 40 percent of a collection would never circulate; the disintegration of acidic paper and the need to microfilm many books; inadequate cataloguing systems he labeled as not user friendly; and the changing economics of acquisition. With society “choking on a plethora of print,” he suggested computers were the answer. “Such are the possibilities offered that there can be no option but to embrace them. . . .Since libraries are about to lose anyway, any alteration in strategy can do no harm.” Or listen to an editorial from that same year in Information Technology and Libraries, entitled “The Last Horse Soldiers.” After listing a host of changes in communications technologies, the author concluded, “Gutenberg’s invention established text as the basis of much of our vision of the world. Our children’s view of the world is being shaped by different technologies. The question is not whether newer, richer technologies will displace text for most uses. The question is when.” More recently, librarians Walt Crawford and Michael Gorman have labeled this enthusiasm among librarians for technological solutions “Technolust.”

These attitudes — one fatalistic, the others unbridled optimism — bother me because the authors write as though changes are outside our control, moving us, rather than the result of efforts to decide where we want to go. They do not betray the approach voiced at Stanford in the 1960s, when cataloguing and data processing functions were the uses being studied. Jean-Claude Guedon also felt that new communications alternatives must be considered in the scholarly community. But he added a crucial element to the discussion. “The point here is to invent, but to invent with our eyes steadfastly focused on the current objects, and not on some mirages stemming either from too strong a nostalgia for the past or too great a fascination for technical hype.”

Are libraries moving precipitously? We have learned that technological change always has costs as well as benefits. Enthusiasts, however, are not really interested in identifying reasons why we should NOT develop their grand idea or project. And Americans, especially, have tended to assume that benefits always outweigh costs. Sometimes this unbridled enthusiasm can lead to what I consider pretty silly conclusions. Take the dream of libraries without books, for example. Such an institution is far off in the future, I suspect, for current predictions fly in the face of any realistic cost assessment. Michael Hart’s Project Gutenberg has a goal of 10,000 digitized books by the year 2001. That may seem like a lot, for it equals the number of titles held by the British Museum from the first half century of the printing press. Yet we should also remember that 40,000 new books appear every year, a number equal to the total number of books estimated to have been published during that same period, 1450-1500. This reality has led most historians to find little reason to expect the works on which we rely, the tremendous number of books and journals published before 1980, to be available on-line any time soon. We hope this fact is not lost in the rush to go digital. Too often, all we hear about are new problems and solutions, not the older ones that so bothered James Thompson.

The third concern we have been dealing with is that the consequences of changes in technology are often unanticipated and not predictable. This might lead us to conclude — as many do — that the answers to the problems will be found, or developed as needed. That optimism is not always well supported by history. Bolting after the newest technology simply because it is there and with the hope that any difficulties will eventually be solved seems to me an especially perilous position. And I think this is doubly so because of libraries are so important. As James Thompson noted, “Libraries are mankind’s memories.”

Clearly, all of you are very concerned about the issues I have raised today, and hardly need to be reminded of the problems involved. Indeed, in preparing this talk, I have become more and more concerned about the difficult task confronting libraries and librarians. Tighter budgets, rising costs, smaller staffs, pressing constraints of space, and so on all add to the difficult environment within which you are developing answers to what future libraries will look like and do. But I think the lessons gained by thinking about other builders of systems might be of some use to you. Like urban infrastructure systems, the networks and systems being developed for library use are so expensive that they will foreclose other options, forcing hard choices. Moreover, those same costs factors will create, or at least perpetuate, the differences in levels of service and availability of information that are evident to any observer of libraries today. The digital libraries will not be available to all, and that should cause some thinking about what to do about such inequities. The goal of the ARL in developing digital libraries is universal access, but how you get that is an open — and very important — question that should not get lost. Indeed, all of these circumstances mean that careful decisions are imperative. I am not suggesting you oppose new technologies because of nostalgia or because libraries have traditionally been about printed media. But you should be careful about the ways in which enthusiasm can blind the process of considering options and choices for using those new technologies. You should not give too much control or authority to experts without insuring there is significant opportunity for input from all concerned with the changes; you should ask critical questions about their assumptions and promises. These complex systems require consideration of far more than the technical details. Above all, technological panaceas offer only forlorn hopes.

Let me conclude by referring to a nice little essay by Edward M. Walters, who in 1982 was director of libraries at the University of Texas at Dallas. In “The Future of the Book: A Historian’s Perspective,” Walters tried to look ahead and guess whether, as some predicted, the demise of the book was at hand. He cautioned that “through the centuries, the profession of history has evolved an elaborate methodology for examining the past with microscopic scrutiny; but the profession lacks a methodology for looking into the future. . . . [He warned] I have been trained to look in the wrong direction for the topic under consideration.” This warning seemed especially germane to me! Yet Walters actually did pretty well as a prognosticator. First, he predicted that much scientific literature would appear online, and scholarly dispute would move there as well. Moreover, as the cost of traditional distribution of specialized scholarly information to relatively small number of individuals rose, he expected “new services to help offset the costs of the printed form. Where these new technologies may lead us is simply unknown.” He thought the disciplines that were noncumulative, as compared to the sciences, “would continue to use their books and journals in the way they always have and with library services continuing in these fields undisturbed.”

Not bad, from 13 years ago. So, given Dr. Walters’ success in looking ahead, I thought I should close by endorsing another of his comments. He noted, “I still agree with Professor Edmond Mignon of the University of Washington that the most immediate impact of the new technology is likely to be ‘the continuing diversification of the information environment rather than any radical changes in the design of primary library services.’ In this conclusion I can take some comfort from the fact that librarians have a number of years of experience living with a diversified information environment, and I do not believe that the experience of the future is likely to overwhelm us.” Perhaps today you all feel more threatened with being overwhelmed than was true in 1982, but to me, this still seems like a reasonable view of what the future holds for libraries. Electronic communication and digital libraries are not the whole story — and seem to me unlikely to become so anytime soon. The book is not dead. Maybe this fact makes the problem facing libraries more complicated, for it means that you really are going to be working in multiple media. In 1965, Paul Wasserman, dean of the Graduate Library School at the University of Maryland, published a volume entitled The Librarian and the Machine, in which he concluded, “The greatest problem of all which relates to the mechanization of library processes is the need to carefully and clearly assess exactly what it is which the process seeks to accomplish, then to reassess each component task in order to ascertain whether it leads logically and efficiently to this end. A machine is in no way a substitute for the human mind in identifying values or the rationality of procedures, for it cannot aid in the choice of alternatives nor can it discriminate except according to a prearranged set of rules which human minds evolve for it to follow.” I think this is a good point to bear in mind, so let me wish you good luck as you contemplate the future through the rest of this meeting.