Document View

Real libraries constitute the only means society has for overcoming the what and who restrictions of "virtual libraries," thereby making copyrighted texts freely available. Attempting to create whole digital libraries comparable to real libraries cannot succeed until human nature itself changes and copyright vanishes.

What is the proper relationship of real libraries to the Information Superhighway? If our profession does not define that relationship correctly, and if we fail to put its elements in a proper order of importance, then everything else we do is going to be off the track. To deal with this issue, I need to consider some of the philosophies of human nature and the nature of copyright, as well. We have all read too many articles in the library literature, and heard too many talks, that conclude by saying, "copyright needs to be studied further," or "copyright problems still need to be worked out." If we're going to talk about reference service and digital libraries, it just won't do to keep deferring this issue. There are enough elements in place right now that solid guidelines for planning can be discovered.

Before I get into that, however, I first want to protest the casual assumption that this is a "digital age." We need to stop throwing that term around so cavalierly. According to the last couple years of the Bowker Annual, the book title output for the United States alone was 46,743 in 1990; 49,276 in 1992; 51,863 in 1994; and 63,689 in 1996 (the last year for which final figures are available). Over seven years, the total book production in the United States alone has grown from 46,000 titles to over 63,000-an increase of 36 percent. The number of serials in paper format is also increasing. The 1992-93 edition of Ulrich's directory lists 140,000 titles; the 1994-95 edition lists 147,000; and the 1998 lists 156,000. According to the latest Ulrich's, fewer than 8 percent of these have electronic equivalents or are available only electronically.

Virtually none of this printed material is freely available in cyberspace. If people want free access (without individual admissions fees or direct charges at the point of use) to the vast bulk of this huge and growing print production, they still have to come inside real libraries. So, obviously, we are living in a mixed paper and digital age; and the ever-increasing mass of books and serials still needs to be collected somewhere and made freely available. Across the board, commercially published books and serials have a much higher claim to the designation "knowledge records" than selfmounted Web sites on the Internet (not that there aren't tens of thousands of wonderful sites on the Web). All of this has a direct bearing on reference service in the real world-whatever term we use to designate that world.

To begin, we must grasp why the bulk of reference service cannot be provided in a virtual library environment without radically dumbing down the whole idea of what constitutes research, scholarship, and reference service. Most of what is written here will apply primarily to research and public libraries-not so much to special libraries, which are a breed apart and which do not offer an acceptable model for research and public libraries to imitate.

Most Books and Journals Cannot Be Digitized

Why is it that most books and journals cannot be digitized? Primarily because of copyright realities. Many digital library enthusiasts regard copyright protection as a "problem" that will be "solved" in the online environment, that it will be solved at "summit" meetings of publishers, authors' groups, government representatives, and librarians, and that librarians will call the shots at these meetings. Well, there have been any number of summit meetings already, and the bottom line, that they never get around, is that copyright is not the problem. Piracy is the problem. Copyright is the solution. The concept of copyright itself is meaningless if it does not entail some very real restrictions. Restrictions are not going to vanish-nor can they realistically be considered "problems" capable of being circumvented by technological devices or by clever discoveries of legal loopholes. Why not? The answer lies in human nature itself. History and political philosophy, especially after the events in Eastern Europe of 1989, can be important guides to library and information science here. These other disciplines show that large-scale enterprises simply don't work out in the long run when they are based on the assumption that human beings will voluntarily forego the advantages of private property (including intellectual property), and that they will, instead, happily contribute their work product, freely, to a larger collective good. Human beings seem especially predisposed not to abide by such selfless goals, in the long run, when they are given the perception of alternative economic arrangements, and the freedom to pursue them.

In other words, the digital age faith that copyright problems will be worked out so that everybody has free fair use access to "everything" in virtual libraries is based on essentially unworkable Marxist assumptions about human nature. Marx thought that if private property were abolished, then there would indeed be a change in human nature itself-and that the new citizens, no longer alienated by class distinctions of haves and have-nots created by property ownership, would voluntarily contribute their work products to the collective good. Allocations of resources would be made fairly and equitably among all people based on the selfless benevolence of citizens rather than through market mechanisms of supply and demand. The verdict of history is in, however. This is not an assumption about human nature on which we can rely.

I am not saying that there is an alternative "capitalist" view of human nature. There isn't, although capitalism can be reconciled with several different views of human nature that are not reconcilable among themselves-Freudian, Existentialist, or Christian views, for example. But the view of human nature that I think is most "bankable" is the one underlying The Federalist Papers and the Constitution of the United States. Instead of assuming the selfless benevolence of human beings as the operating mechanism of a free society, the founders of this country assumed the opposite. They assumed that citizens needed to be protected from the inevitable infringements of other citizens. And so they developed an elaborate mechanism of divided representation, separation of powers, and checks and balances, that would be sustained precisely by the persistence of individual and selfish interests competing with each other. As Federalist 51 notes, "What is government itself but the greatest of all reElections on human nature? If men were angels, no government would be necessary."

What does this mean for librarians in a digital age? If men and women were angels, no copyright protection would be necessary-there would be no piracy, no sense of infringement to begin with, and no perception of unfair use. But protection is necessary because people do recognize infringement and unfair use when they see it, and these perceptions cannot be wished away or circumvented by legal loopholes. If we librarians assume the opposite, that people will selflessly contribute their private property to the Internet for the good of all and that everything will be freely available to everyone in cyberspace, then our plans for the future are built on a foundation of sand. We cannot assume that some legal loophole will be found that will enable intellectual property to be socialized so that everyone will have free and equal access to everything on the Internet. Human nature itself will have to change before that happens; and that's why copyright problems cannot possibly be worked out in cyberspace.

Notice, however, that it is the conflicting interests of intellectual property, on the one hand, and free and equal access to information (whether copyrighted or free), on the other, that I argue cannot be solved in cyberspace (synonymously, the Information Superhighway). But while problems of intellectual property and free access cannot be solved in that world, there is a solution to these problems if we simply step outside cyberspace, rather than try vainly to force everything into cyberspace. Perhaps the configurations of that "outside" world are so obvious that we're too close to see them. And it doesn't help that our profession is wearing digital age blinders that prevent us from recognizing the permanent importance of real libraries with printed books and serials-and sitelicensed electronic resources.

Those who maintain that copyright will disappear in the digital age are overlooking the obvious solution that publishers and authors are already endorsing on a wholesale basis; namely, that those people who wish to maintain hassle-free control of their intellectual property will simply avoid putting it on the Information Superhighway in the first place and confine publication to paper formats.

The Inevitability of Restrictions

Given that copyright inevitably entails restrictions, there are only two kinds of restrictions that are possible in the electronic world: either you restrict what is made available to begin with-mounting only copyright free texts that are in the public domain-or, if you do mount copyrighted texts on the worldwide network, and wish to profit from them, then you must restrict who has access. This means that only people who can pay high subscription costs for passwords, or who can use credit cards at the point of use, will be able to get into the most desirable, copyrighted parts of the virtual library. If you do mount copyrighted texts without encryption or access restricted via passwords, then, realistically, you simply no longer have copyright protection. The obvious example is that right now there are more than a thousand journals that are full-text searchable in NEXIS, but how many people can afford to pay the access fees of about four dollars per minute if you search the whole database?

There are two middle grounds in the electronic world. The first lies in large organizations subsidizing costs for their members. Some law schools, for example, subsidize access to LEXIS for their faculty and students. But membership in such a group, with restricted passwords, is actually just another form of the who restriction. Can access be similarly subsidized for the whole country, so that everybody has equal access to all information electronically-both public domain and copyrighted information? Think what this actually means: It's tantamount to the complete socialization of intellectual property in the United States! This simply cannot happen. Note: this does not preclude the possibility of some form of subsidized access for educational institutions to the basic Internet system (those parts of the Internet that are freely available to everyone who gets in). But subsidizing initial access to the basic system is not the same as subsidizing entry to its thousands of copyrighted sites that have who restrictions on them.

The second middle ground in the electronic world is to restrict access to some databases to a limited geographic area-that is, to a certain number of terminals available only in designated locales. Individual passwords would not be needed for access to these sources, and anyone who goes to those few terminals could indeed have free access to whatever is on them. But since the physical location of the designated terminals in this model is severely restricted, these site-licensed electronic resources are, in a very real sense, not part of cyberspace. They are not on the Information Superhighway to begin with-that is, they are not accessible from anywhere at anytime by anyone. You have to be inside a real library with walls in order to use them. And, in digital age thinking, having to go inside a real library with walls, during limited hours of opening, is the very antithesis of being able to tap into a virtual library from anywhere, at any time. The major implication here is that we have to go outside the Information Superhighway-outside cyberspace-in order to escape its inevitable what and who restrictions. This is precisely what can be done in real libraries. And, physical libraries with walls can do much more than offer free access to site-licensed electronic resources; they can offer free access to the vast bulk of humanity's memory contained in books and journals-a memory that is not available on the Information Superhighway. In real libraries, the full texts of every book and every journal in their collections can be freely read by anyone who comes in the door. Further, real libraries are not bound to offer only public-domain material, to restrict who can come in, or to charge patrons any direct (and prohibitive) point-of-use costs for reading anything at all in their print holdings or their site-licensed electronic resources.

It's not uncommon in our profession for librarians to attend daylong seminars that consist, in the morning, of having to take the Myers-Briggs personality assessment; and then, in the afternoon, of being told that we need to think "outside the box." Well, maybe we actually do need to think outside the confining box of the Information Superhighway and Internet, with its massive and unavoidable what and who restrictions, if we're going to find a solution to copyright problems.

If we librarians want to promote free access to the knowledge records of the world, we have to be consciously aware that overcoming the what and who restrictions of digital libraries necessarily entails a where limitation-that is, a restriction to a library place that actually has walls. This particular limitation, however, is mitigated by the lending practices of most real libraries; much of their material can be loaned or checked out for use elsewhere and, in those other locations, can be used at any time of the day or night. The where limitation is also mitigated by the fact that there are so many real libraries and that they are so geographically dispersed in so many communities. The real solution to problems of free access to copyrighted material already exists: It is to be found in the widespread geographic dispersal of real libraries with walls, and not on the Internet.

The Crucial Importantce of "Real" Libraries

Our society and our culture need a mechanism whereby everyone can indeed have free access to the intellectual work of everyone else in a way that nonetheless protects property rights by imposing a restriction. Real libraries-not virtual libraries, but real libraries-with their inherent where limitation, constitute the best and, indeed, the only possible mechanism we have for providing this free access. Although many digitally infatuated theorists criticize real libraries on precisely this point-that they restrict access to only those people who can come to a certain place-the "library without walls" that is proposed as the alternative is itself a massive tradeoff: This digital age proposal naively and routinely overlooks the fact that as soon as you eliminate the localized walls, you inevitably create other enormous restrictions of what and who.

What this means is that the where restriction that is viewed by the cyberprophets as the weakness of real libraries is, in fact, precisely the major strength of real libraries. Without this geographic restriction on access, there is a real danger that, in cyberspace, much less of the most substantive knowledge records of humanity will be freely available to citizens and researchers-for example, those sixty thousand books and one hundred and fifty thousand serials that are now being printed on paper every year.

In terms of reference service, then, if we change our priorities to focus on the nonlocalized clientele of the Internet as more important than those readers who can come into real reading rooms, there is a real danger: It is that of radically dumbing down reference service to the point that we seek to steer researchers not to the best sources available for their inquiries but rather to only those sites that they can tap into freely on the Internet. There is already a strong tendency among students in particular to limit their research to only Internet sources, to the detriment of both their critical thinking and everyone's future culture. I suggest that a major goal of reference service in the present age should be to correct rather than to exacerbate this very serious problem. To correct it, we need to emphasize the importance of real libraries primarily as destinations in themselves, not primarily as on-ramps to the Information Superhighway and not primarily as content-providers of full-text collections for the Internet. Even if we do assert the importance of real libraries as places, the problem is not solved (and is in fact exacerbated) by promoting them as places for meeting rooms, cultural programs, and coffee bars. That kind of promotion shifts the primary emphasis away from acquiring, cataloging, providing reference service for, and preserving real collections that do not have massive what and who restrictions on them.

If we justify the continued maintenance of library buildings as places only, or even primarily, in terms of meeting rooms and coffee bars-or in terms suggesting that our primary function is to provide places where people can get help with the Internet-then we are tacitly agreeing, without a shot being fired, to the concealed proposition that real collections with where limitations are in fact no longer of central importance. When our best defense of real libraries implies that their primary function nowadays is to provide services other than access to real collections onsite, then our "defense" plays directly into the hands of budget cutters who assume that real collections are indeed obsolete in a digital age. If we change our priorities to rank meeting rooms, coffee bars, and help with the Internet above the maintenance of real library collections, then we are effectively encouraging the tail to wag the dog; in the long run, this can only be destructive to the health of the dog's body.

Given the human nature we actually have, we can never entirely eliminate restrictions on access to intellectual property. We have to settle for tradeoffs among them. But the geographic restriction on access to our culture's most substantive knowledge records is a tradeoff that libraries can do more than just live with; we can thrive on it. We first need, however, to cure ourselves of the suicidal notion that has taken hold of the library profession's soul in the last decade, that providing service within walls is our primary weakness when in fact it is precisely our primary strength. This fundamental, rock-bottom misperception of reality-this habitual predisposition in our profession to look at our problems through the wrong end of the telescope-is tainting everything we librarians do. When we always frame our own questions and set up our own conference agendas automatically to include phrases like "in a digital age" or "in virtual libraries," then right off the bat we are boxing ourselves into a universe of options that is too limited; we are thereby precluding the very possibility of finding actual solutions to our most pressing problems in the real world.

If we continue to portray the provision of service within walls as something to apologize for, to be embarrassed about, and to get beyond, then we will be killing not just our own profession, but also our culture that depends on the free access to substantive knowledge records that can be provided only by real libraries with where restrictions. Substantive knowledge records, here, means not just the increasing avalanche of printed books and journals that have gone through the important filtering processes of publication and selection, but also the increasing variety of site-licensed electronic resources that cannot be tapped into from anywhere, at anytime, by anyone out in cyberspace.

Reference Service to Remote Users

A second question arises, then: To what extent can librarians-surrounded by copyrighted print sources, and by the vast bulk of public domain print sources that are too expensive to digitize, and by site-licensed databases that are not freely available in cyberspace-provide reference service to remote users who are not inside the walls? Although much current library literature seems to focus on words like "opportunity," "transition," and "transformation" in this area, I believe the genuinely relevant term here is "tradeoffs." We have to make a practical distinction among levels of service; we have to distinguish reference questions from research questions; and we have to make a further distinction within research questions themselves.

For purposes of this discussion, let's define reference questions as those that have a reasonably determinable "right" answer: "What was the population of Chicago in 1920?" "What is the zip code for a particular street address?" "Where can I locate a copy of a particular book?" Research questions, in contrast, are more open-ended: "What do you have on U.S.-Israeli relations after the Six Day War?" "What do you have on antebellum black artists?" "What can I get on the history of advertising in the United States?" Many reference questions can be handled routinely with standard encyclopedias, almanacs, directories, Web sites, and so on-although, as anyone who has to answer such questions can tell you, some of them involve very complicated and time-consuming digging. Research questions, however, usually take more time than reference questions; these can be handled in either of two ways.

The first way to handle research questions for offsite inquirers is to send printouts to them that will guide them with their own research once they do come inside a library's walls; in these cases we try to identify the best or most likely sources without taking the time to actually read and individually evaluate them ourselves. For example, a few months ago I had an inquiry from someone in California who wanted to know how to research American advertising history through the Library of Congress. I handled this one by printing out lists of relevant subject headings with the number of postings to be found under each in our book catalog; I also printed out citations to published bibliographies on advertising history, photocopied descriptions of special collections at LC that cover the subject, and also photocopied pages from Lee Ash's Subject Collections directory that identify special collections on advertising history throughout the country (since not everything is at the Library of Congress). I also printed out several Web pages on advertising collectibles, pages from advertising history associations, and pages from libraries with good collections on the subject, among them the National Museum of American History and the Hartman Center for Sales, Advertising, and Marketing History at Duke University. But I did not take the time to actually request and read any of this material myself-this is what the researcher must do to really answer his research question on advertising history. And, let me emphasize, the Web sites I found did not provide the actual materials in their collections online; the reader will still have to travel inside the walls of the Museum of American History of the Hartman Center to use their material.

The second way that reference librarians can handle a research question from offsite is to actually read, compare, and evaluate the sources themselves. Here, even a relatively uncomplicated search can still be very time-consuming. For example, a few months ago I had to answer an inquiry from the director of the Biblioteca Palatina in Parma, Italy, asking for biographical information about a New York banker named Calvin Bullock, who died in 1944. Since the necessary sources were probably not readily available in Italy, it wouldn't have done any good to just send a list of citations. I had to do the research here. It turns out there was a small paragraph about Bullock in Who Was Who in America. But I also checked the Personal Name Index to the New York Times Index 1851-1974, a papercopy set covering the bulk of the newspaper's retrospective years that are not computer-indexed, which gave me twelve more references. To follow them up, I then had to check ten individual paper-copy volumes of The New York Times Index itself-again, not computerized for the relevant volumes spanning 1930 to 1953. From these twelve citations I then picked out the most likely sounding three, one dealing with the fiftieth anniversary of Bullock's firm, and two obituary notices for him. I then had to walk to the newspaper room across the street, pull the Times microfilm, crank through it to find the articles, and read all three. Only one of them was substantive enough to photocopy. I then had to take that reel to a different machine to make the copy. Finally, I had to come back and write a cover letter.

This second way of handling a research question-pursuing the inquiry to the point of actually reading full-text sources, and comparing, evaluating, copying, and sending them off-obviously takes a lot more time than just identifying sources and directing a researcher to look at them. The point here is that for every hour that I, as a reference librarian, have to spend on this kind of inquiry from an outside correspondent, is an hour I am not doing other things that also need to be done onsite. With limited time, you have to trade off something. And what you trade off inevitably affects the quality of either the collections or the service you can provide onsite to a larger number of people.

It is a fallacy to think that opening the e-mail in box to anyone offsite who wants to write in with a question is a means to serve "more" people. The tradeoff in the real world is that reference librarians actually have to spend much more time on a much smaller number of individual questions. It's not that we should stop handling questions from outside the walls. It's a good thing to do some questions in depth-that's how we learn our own collections, and we become better reference librarians for the experience. But, librarians cannot simply do research for all outside inquirers, and then fax them the results. We cannot realistically rank the provision of this kind of time-consuming service as a priority higher than helping a greater number of researchers inside the walls. Here at the Library of Congress there is a "special library" function to do exactly this kind of research for Congress; but it takes 760 full-time workers in the Congressional Research Service to satisfy the needs of 535 members of Congress. If the same ratio of workers to clientele holds true elsewhere, it would take 369 million reference workers to provide this kind of service to the U.S. population of 260 million people.

Digitized Collections?

Can research questions be handled by digitizing the full texts of sources and distributing them so that people won't have to come into real libraries? In general, no. But, there is a minor qualification. There may be a change in the copyright law, to allow academic libraries or individual professors to, in effect, digitize very limited collections of copyrighted material comparable to what would be found on a reserve shelf. This exception, then, would be allowed only for libraries supporting formal distance education programs. The main restrictions would still be in place, however: Access would be limited to only those few students formally enrolled in the class-a massive who restriction-and only for the time period of the class.

It should be pointed out that doing research from a digitized "reserve shelf" collection alone (or even primarily) is not comparable to doing real research-it's more like playing a game in which you know the Easter eggs have been hidden inside a definite, confined area, and all you have to do is locate them. This is not the same as has having to determine if they do exist to begin with, and then also having to find out where they may be located when all fields are open. When the very questions that are asked by teachers are limited and tailored in such a way that they can be answered within the narrow confines of previously digitized collections, then some learning may be going on. But it is not a learning of how to do real research.

Indeed, such "distance" learning may even serve, unintentionally, to further confirm the students' predisposition to believe that all research can be done by simply tapping into the Internet. If they are graded favorably for using the electronic sources alone, then no matter how many times we piously encourage them to use other sources within real libraries, too, they will most remember the concrete rewards they received by ignoring such advice. And when, in a few years, those students become taxpayers, they will then recall that they could get their own degrees without using real libraries-so why should they pay taxes to support such obsolete institutions for other people when their own experience has confirmed that everything a student needs is in fact on the Internet?

Apart from such temporary and restricted reserve shelf digitizations, can libraries digitize their whole collections for anyone to tap into at any time? Of course not. The Library of Congress, for example, hopes to scan five million images into its National Digital Library by the year 2000. Most of the items, such as maps, photos, and postcards, have been selected so that one image equals one complete item. If one counts each page of a book as a scannable image, however, in 1996 alone the Library received SO million new images into the collection. With probably the largest conversion project in the country, LC is not even close to keeping up with current receipts, let alone with scanning in the 2.6 billion individual page images that it already possesses.l Even apart from the astronomical costs involved in digitizing such a collection, and even apart from the massive preservation problems created by electronic media, digital libraries in cyberspace comparable to real libraries-and that is the important phrase, comparable to real librariescannot be created in the first place without the fundamental changes in human nature discussed above. We librarians cannot freely distribute our collections electronically until a change in human nature eliminates copyright and site licensing.

Most research into the substantive knowledge records of our civilization must still be conducted within the walls of real libraries by readers themselves, using not just printed and microform sources, but site-licensed databases that they cannot tap into from anywhere, at anytime, out in cyberspace. This being the reality of the information world in an age of mixed print and digital media, it follows that the primary focus of reference service must be, no matter how "unsexy" it sounds, on the needs of researchers who come within the walls of real libraries.