Some of us are still digesting Google’s startling announcement in December 2004 that it will be working with five major research libraries to digitize more than fifteen million books from their collections in exchange for providing these libraries with digital copies of their books. Google will load the copies into their own digital library and make full-text versions available if they are in the public domain, or brief excerpts—snippets—if they are still under copyright protection. The project promises to cost millions of dollars—perhaps as much as a billion—and to take six to ten years to complete. Previously, the libraries involved had thought that such a digitization project would take far longer—when library staff at the University of Michigan were asked in 2004 how long it would take to digitize Michigan’s seven million volumes, “the answer was more than a 1,000 years.”1 The project was—and still is—staggering in its speed and daring.

In the two years since that announcement, a rapid succession of announcements from Google and other organizations astonished us with the scope and potential for enormous impact:

• The Seattle Times reported on October 3, 2005, that Yahoo and Microsoft would team up with the nonprofit Internet Archive as well as several other large research libraries and archives in establishing the Open Content Alliance (OCA) to pool the collections of a large number of research libraries.² According to the article, OCA will only digitize those materials in the public domain, including handwritten manuscripts, unless copyright holders give explicit permission to digitize. The complete books will be freely available in a permanent archive. Funding and support will come from Yahoo and Microsoft as well as from participating libraries and other companies, such as Hewlett-Packard Labs, LibriVox, Octavo, Lulu.com, and Adobe. It appears that everybody wants to get in on the act! That Yahoo and Microsoft jumped on this bandwagon is not surprising. Yahoo is Google’s archrival, and Microsoft’s share of the searching market is growing all the time—it is, after all, “the default search engine built into the default Web browser available right out of the computer box.”³
• The Financial Times (London) announced on November 4, 2005, that Microsoft is investing in a digitization project of 100,000 books from the British Library.⁴
• James H. Billington, Librarian of Congress, wrote an article for the Washington Post on November 22, 2005, about the Library of Congress (LC) receiving $3 million from Google to jump-start their digital archive of international cultural artifacts, the World Digital Library.⁵
• On March 7, 2006, the Australian reported that the European Commission (EC) plans to make at least six million books, documents, and other cultural works available by 2010. The EC will contribute $72 million to the digital library, and expects member states to make up the remaining $250 to $300 million to complete the project. Its goal is to combat other digitization projects that have an Anglo-American–centric view of history. The EC says it is not going after copyrighted works, but does not reveal details of the program, which has publishers worried that it might affect their own digital preservation programs.⁶
• The Boston Herald reported on June 15, 2006, that simultaneously with the Shakespeare in the Park festival in New York City, Google is launching a Web site that allows users to search all of Shakespeare’s plays and poems—which, of course, are in the public domain.⁷
• A New York Times article on August 9, 2006, reported that the University of California would join the Google project, adding millions of books from the system’s one hundred libraries. The digitization program will include copyrighted works. Google is talking to other libraries as well.⁸
• According to the Milwaukee Journal Sentinel, October 13, 2006, the University of Wisconsin has jumped on the Google bandwagon.⁹
• A few days later, the Financial Times (London) reported that Microsoft has a new partnership to scan books from Cornell University’s library. Microsoft already has partnerships with the British Library and other library members of the OCA.¹⁰

As Van Orsdel and Born observed, perhaps the good news in all of these announcements is that book digitization projects have taken over the spotlight in the past two years and upstaged the serials crisis.¹¹

We should be grateful to Google for sticking out its neck—for pushing the envelope on technological innovations, copyright, and other important aspects of digitization. At a symposium at the University of Michigan in March 2006, Google’s Adam Smith said we need to “just do it” and “not let perfection be the enemy of the good,” and that we need to “get it out there”—learn from mistakes, iterate the process, and make it better.¹²

On the other hand, Yahoo, Microsoft, the Library of Congress, and the OCA are staying in the background, which may be a good place to be. It not only is safer, but their smaller programs permit experimentation and policy setting to be done with older materials and those materials not under copyright. They are learning a lot, and introducing technological innovations with projects that do not risk lawsuits.

In regard to copyright, Google, publishers, and the participating universities all agree on the fundamental issue that intellectual property laws should be respected. Nonetheless, Google is being subjected to numerous lawsuits in the United States, France, Germany, and elsewhere because some publishers disagree on whether Google is infringing on copyright. Google is taking an extremely aggressive stance on copyrighted materials, insisting on an opt-out model that requires authors and publishers to contact Google and tell them they do not want their books included. Google says that opt-out is much easier, cheaper, and quicker than opt-in because Google would have to contact millions of copyright holders before even deciding which books they could digitize. Further, a large percentage of copyright holders would be virtually impossible to reach. This is one manifestation of the orphan works (copyrighted works whose owners may be impossible to identify and locate) problem.¹³

Authors and publishers say that Google is looking at this only from Google’s perspective. What if a lot of companies and organizations—not just Google—get into large-scale digitization? That would put a big burden on the copyright holders who want to opt-out but might not even be aware that their books are being digitized.

Electronic journals and digitized versions of older print journals have become firmly established in research libraries’ collections. Why digitize books as well? Twenty-first-century scholars are increasingly bypassing books—looking for background information in print library collections may slow down the scholar who wants to be productive. Even scholars in the humanities and social sciences are looking to their colleagues in the sciences, modeling their behavior after them because all scholars want to save time and be more productive. Initially, historians were hostile to JSTOR (a trusted archive of scholarly journals), but now most find it extremely helpful in their research.

College students use books and journals (at least if they have been trained to do so; otherwise, they simply use a search engine to find information on Web sites). For both students and scholars, however, the book is becoming increasingly irrelevant for learning and discovery.

Looking back a few centuries provides a perspective on how the pace of change is forcing us radically and rapidly to rethink our assumptions about scholarship. The transition from an oral to a written culture developed over many centuries. As Bengston said,

During this slow evolution, our way of thinking fundamentally changed, from repetitive, oral, memory-based knowledge to visual and spatial memory, based on the physical object of the book. For centuries books were simply the most efficient and usable technology for the transmission of culture and ideas. We need only reflect on the past few years to sense how quickly and radically the ways that we write and communicate have been and will be altered.¹⁴

What do modern scholars and students really want or need? Have we factored their rapidly changing needs, preferences, and habits into our preservation programs? Predicting what, exactly, will happen to print books or even e-books in this century and beyond is impossible. Many people are confident that certain kinds of books “will cease to exist on paper: directories, reference works, textbooks, travel guides, to name a few.”¹⁵ No one can say, however, how much scholars and students will care about linear, narrative, book-length treatments. We do not know how much generations to come will care about preserving words, compared to visual and multimedia documents or even raw data. The only thing we may be certain of is there will be a tidal wave of interest in networked, digital media. Are our preservation programs responding to those trends?

For some time now, libraries have paid attention to cooperating in digitization projects that focus on unique collections as a cost-efficient way to give scholars all over the world access to rich resources and to preserve those valuable print materials that were deteriorating. Just a little more than two years ago, Brian Lavoie and Lorcan Dempsey at OCLC Research admonished libraries to be very careful and wise in allocating their insufficient budgets for preservation.¹⁶ Accordingly, OCA members are carefully selecting which materials to contribute. Google’s general approach, on the other hand, has been to throw a lot of money at the problem and grab as many books as they can without selecting particular parts of collections. Daniel Greenstein, director of the California Digital Library, was quoted in January 2006 as saying that his discussions with Google officials disclosed that they are “more interested in grabbing a large quantity of materials than in carefully selecting certain collections of works.”¹⁷ Their attitude is simply, “the more of it, the better.”¹⁸ This approach is not true at all universities participating in the Google project, but it is a general strategy.

The question is whether a selective collection policy for digitization and preservation is better than a scattershot approach. It appears inevitable that the gems of our collections are not going to be exploited unless we digitize them. But should we be aiming for digitizing everything as fast as possible, so that we can provide what Mary Ann Coleman, president of the University of Michigan, referred to as “instant gratification of a one-in-a-million need,” or should we be taking a more measured approach that addresses the most likely and important needs now and in the future?¹⁹ I propose that we need to think more carefully about preservation priorities and match them to the norms of twenty-first-century scholarship. We need to spend our scarce resources for those digitization activities that not only will increase access, but will serve our long-term preservation goals as well.

At a symposium titled “Scholarship and Libraries in Transition: A Dialogue about the Impacts of Mass Digitization Projects,” held at the University of Michigan on March 10–11, 2006, Clifford Lynch, executive director of the Coalition for Networked Information, was the wrap-up speaker.²⁰ He proposed that digitization is a form of insurance—in fact, one of the best forms of insurance we have. He said it is not a replacement for the physical object, but increasingly a good (albeit not perfect) surrogate. But is it really? If a foreign army came marching through your town, would you be preserving documents by tossing them into the hayloft? They would be out of the way of the marauders, yes, but they still would be subject to thieves stealing them, mice nibbling away at them, and rain leaking through the roof. Preservation is much more than finding a compact, convenient, and inexpensive place to stash materials.

The academy has enduring values and standards of preservation. Every academic library has in its mission statement something about archiving, conserving, or preserving the scholarly record for perpetual access. For example, on the University of Maryland Libraries’ Web site is their mission statement: “Providing access to the use of the scholarly information resources required to meet the education, research and service missions of the University. The Libraries support this effort by building, organizing, maintaining and preserving these resources.”²¹

Another way to express this statement of values comes from Coleman. In a speech to the Association of American Publishers in February 2006, she said, “General Motors does not need to maintain the tools for its 1957 Chevys, and would have a hard time manufacturing a car from that year. But a university is responsible for stewarding the knowledge of 1957, and for all the years before and after—the books and magazines; the widely known research findings and the narrow monographs; the arcane and the popular.”²²

Given that academic libraries accept a staggering responsibility with limited resources to meet that responsibility, they need to win people’s trust that they will fulfill their mission to preserve evanescent digital materials.²³ They also need Google and other commercial enterprises as valued allies and partners. Karen Wittenberg, director of Columbia University’s Electronic Publishing Initiative (EPIC), affirmed libraries’ dependence on the for-profits—“We need to face the fact that commercial search engines are now the mechanism of choice for finding information, and we desperately need Google and other powerful players as valued partners with whom we will negotiate effective ways of collaborating that benefit our businesses and our users.”²⁴

Nevertheless, librarians and archivists need to be careful. They may get chummy with the staff of the for-profits—their employees are awfully friendly people, and how can you dislike a company such as Google, whose official motto is “Don’t be evil?”²⁵ But we should never think of Google, Microsoft, or Yahoo as one of us. Google has a corporate mission “to organize the world’s information,” but it is for the goal of building and sustaining a massive and highly profitable media empire.²⁶ Google may sincerely believe that it operates according to higher principles, but some of its recent actions, such as its decision to abide by political restrictions placed on it by the Chinese government, prove that it is willing to compromise its principles in order to stay competitive.²⁷ A lot of money is to be made in digitizing books—when the content moves from physical to digital, its value jumps enormously. When there is money to be made, libraries and archives should be vigilant and alert. Too much chumminess with commercial enterprises raises three basic problems or issues related to trust: quality, secrecy, and long-term stability.

A June 2004 report from the Association of Research Libraries endorsed digitization as an “accepted preservation reformatting option for a range of materials.”⁴⁷ The report conceded that “ensuring high-quality image capture and providing for the long-term viability of digital objects is an admitted challenge.”⁴⁸ The information professions, however, must take the leadership in developing standards and best practices, including developing “strategies to keep master files safe for the short-term, [which includes] the use of high-quality and reliable storage media, multiple back-up systems, periodic testing, and a schedule to refresh data.”⁴⁹ These short-term strategies will at least keep the materials safe—safer than in the hayloft—while long-term solutions are being developed. This is the proper leadership role for librarians and archivists. Are we up to it, or will we let the eight-hundred-pound-gorilla companies drive the agenda and set the priorities? More specifically, are we even concerned that the gorillas are not dealing with preservation? A search in Lexis-Nexis for articles in the general and business news sections uncovered hundreds of hits on the topic of “(Google OR Yahoo) AND digitization AND books.” However, as soon as the word “preservation” was introduced into the search string, the count dropped to zero. Shouldn’t that worry us?

We need to stop being reactive; we need to go after the preservation target in a strategic way. We own this problem of preserving books … at least for now. Ironically, many librarians were unhappy with a 2005 OCLC report because one of their key findings was that in the public’s eye, the library brand is books.⁵⁰ That finding is troubling if people think of libraries as only about books. But we will be in much more trouble if our users stop thinking even that. Do you want a book? Go to Google or Yahoo or Amazon.com. Where will libraries be then? No brand recognition at all!

A news item on October 24, 2006, reported that Google, in partnership with the Frankfurt Book Fair literacy campaign and UNESCO’s Institute for Lifelong Learning, is launching an online portal to connect literacy organizations.⁵¹ In addition to allowing “organisations, teachers and others with an interest in literacy to search online for and share literacy information,” the tool provides a zoomable, searchable map that enables users to locate literacy organizations around the world.⁵² Searchers could find information in academic articles and digitized books, and share the information they find via groups, videos and blogs.

This is leadership on a scale that only a huge organization with extraordinarily deep pockets, a focused mission, and amazingly creative ideas can hope to mount. We have to applaud Google for this leadership, but are we also a little jealous, or worried? Perhaps we should be, if not the former, at least the latter.

I have raised many questions without supplying answers. Why? Because this is all happening so fast. Research libraries with a mission to preserve collections and make them accessible to future generations will be affected, but we do not know exactly how yet. My cautions in each of the five areas are:

• Pace. We cannot slow it down; the pace car is Google, and other commercial drivers are nearly as pushy. But we must find the time to digest it all before making irreversible decisions about our precious collections.
• Risks. Let Google and others take risks if they wish, but we should not be taking risks with our collections, nor should we be risking our users’ free access to our collections in the future.
• Justification for digitizing books. Given that we all have to get on this bandwagon, and given that we all have limited resources to do so, we should be thinking of ways to maximize value for scholars now and in the future. What sorts of materials will be of the most interest into the far-distant future? Can we afford to digitize older materials that are the “Long Tail” of our collections—items that will appeal only to a small number of researchers, or only one, or maybe none at all?⁵³
• Trust. We must find partners—sister libraries, commercial entities, government agencies, and others. But we must keep a clear head about which of those organizations will be around for the long term, which of them share our values and our mission, and which truly understand preservation and conservation issues in regard to fragile, valuable, endangered, and irreplaceable artifacts. The Open Content Alliance model keeps a lot of control under the contributing libraries, where it should be.
• Leadership. The leadership needs to come from all parties in this endeavor. Because we are all mutually dependent, no one organization is in a position to dictate the discussions or the outcomes. Google and Yahoo need our content. We need stable, robust technology platforms for preservation and wider use of our collections. Scholars and students eed more access and knowledge about how to use these collections. We all need to stay in close communication and collaboration … with our eyes wide open! In the end, research libraries alone will be held accountable for fulfilling that vital preservation mission.