Lewis.indd A Strategy for Academic Libraries in the First Quarter of the 21st Century 1 David W. Lewis The wide application of digital technologies to scholarly communications has disrupted the model of academic library service that has been in place for the past century. Given the new Internet tools and the explosive growth of digital content available on the Web, it is now not entirely clear what an academic library should be. This article is an attempt to provide a strategy for academic libraries in what is left of the first quarter of the 21st century. There are five components of the model: 1) complete the migration from print to electronic collections; 2) retire legacy print col- lections; 3) redevelop library space; 4) reposition library and information tools, resources, and expertise; and 5) migrate the focus of collections from purchasing materials to curating content. Each of the components of the strategy and their interactions will be considered. It is hoped that the result will provide a useful roadmap for academic libraries and the campuses they serve. “You’ve got to be careful if you don’t know what investments in libraries made sense. where you’re going ‘cause you might not Lombardi made it clear that in the com- get there!” petitive environment of higher education ~Yogi Berra1 today, if libraries could not make a strong and clear case for their role, the money n September 2006 at a confer- ence on library assessment in Charlo esville, Virginia, John Lombardi, the Chancellor of the University of Massachuse s-Amherst challenged the assembled librarians. He said that despite the fact that his mother and sister were librarians, and that, as a Latin America historian, he had depended on libraries and librarians all of his profes- sional life, he did not know anymore what an academic library should be. Thus as a campus leader, he found it hard to know would go to the new student recreation center because that is what students and their parents asked about on the campus tour.2 Coming from a thoughtful and influential friend of academic libraries, Lombardi’s words should be a wake-up call. Jerry D. Campbell expressed similar concerns in his 2006 EDUCAUSE Review article when he said, “Because of the fundamental role that academic librar- ies have played in the past century, it is tremendously difficult to imagine a David W. Lewis is Dean of the University Library, Indiana University Purdue University Indianapolis; e-mail: dlewis@iupui.edu. Versions of this article were presented at “Visions of Change: Academic Librar- ies in Transition,” California State University at Sacramento, January 26, 2007, and at the Coalition for Networked Information, Spring Task Force Meeting, Phoenix, Arizona, April 17, 2007. 418 mailto:dlewis@iupui.edu A Strategy for Academic Libraries in the First Quarter of the 21st Century 419 college or university without a library. Considering the extraordinary pace with which knowledge is moving to the Web, it is equally difficult to imagine what an academic library will be and do in another decade.”3 The recent report of ACRL’s Roundtable on Technology and Change in Academic Libraries echoes Campbell. Its opening sentence speaks of the “iconographic power of a college or university library.” But in the next paragraph the report states, “The busi- ness of libraries can now be understood as one component of a rapidly evolving, almost wholly transformed environment in which information is proliferating at heretofore unimagined rates and in which the ability of academic libraries to deliver authenticated and reliable information is continuously challenged by new tech- nologies.”4 It is easy to understand why, at the end of the age of print, academic libraries— and, indeed, all libraries—are dazed and confused. The technology upon which we have built our missions over the past half millennium is being usurped. The devel- opment of print in the 15th-century and the 19th-century industrialization of print made libraries what they are today. Or, to be more precise, what they were in 1993 when the Web era began. Most of what we as librarians know about organizing information is a refinement and enhance- ment of the work of Melvil Dewey and other 19th-century library pioneers. As Google so powerfully proves every day, authority control and classification are no longer the only, or the best, answers. Academic libraries must find and articu- late their roles in the current and future information ecology. If we cannot or will not do this, our campuses will invest in other priorities, and the library will slowly but surely atrophy and become a li le used museum of the book. This article is an a empt to provide a strategy for academic libraries in the digital age or at least in its early stages. I do not believe that the transitions proposed will take place immediately, but rather that they will play out over the next fi een to twenty years. What will be important is that we manage this transition purposefully and that we not dri through it. There are trade-offs that need to be made. If we are not prepared to make them, it is unlikely that we will be able to marshal the required resources and we will fall short of what we need to accomplish. Assumptions My strategy for the early 21st-century academic library builds on several un- derlying assumptions: 1. Libraries are a means and not an end. Libraries serve as a mechanism for making knowledge available in commu- nities and organizations. More precisely, libraries are the mechanism for providing the subsidy that is required if information is to be used efficiently in communities and organizations. An economic case can be made that, without such a sub- sidy information will be underused and communities and organizations will be less successful than they should be. As technology changes, there may be other be er mechanisms for applying the sub- sidy and we should embrace and support them. 5 One example of such a mechanism is open-access publishing. 2. Libraries confront a variety of dis- ruptive technologies, and these technolo- gies will disrupt libraries. 6 The structures and practices of libraries will no more withstand the technological changes we are facing than the scribal culture withstood the changes brought on by the printing press. Change will not be instan- taneous, but it will be relentless. To take the most obvious example, Google aims to digitize and index all of the world’s printed literature. While Google and the publishing community are currently at odds, it is inevitable that economic models will be found to make copyrighted materi- als openly accessible. What will it mean to libraries when all books are potentially full-text searchable and available to every- one with an Internet connection? 420 College & Research Libraries September 2007 3. Real change requires real change. Incremental adjustments at the margins will not suffice; rather, alterations in fun- damental practice will be needed. Fortu- nately, this is not uncharted ground. There are established strategies and tactics, and we can take advantage of them. 4. We have a window of opportunity. Books and libraries are revered in aca- demic culture, and librarians in general are well thought of by faculty and even administrators. We have a reasonable measure of good will that we can spend down. If we do this wisely, we can suc- cessfully manage the transition we now face. However, this window will not stay open forever, so we cannot afford to wait too long. Parts of the Puzzle For me there are five parts of a strategy for maintaining the library as a vibrant enterprise worthy of support from our campuses. 1. Complete the migration from print to electronic collections and capture the efficiencies made possible by this change. 2. Retire legacy print collections in a way that efficiently provides for their long-term preservation and makes access to this material available when required. This will free space that can be repur- posed. 3. Redevelop the library as the pri- mary informal learning space on the campus. In the process, partnerships with other campus units that support research, teaching, and learning should be devel- oped. 4. Reposition library and information tools, resources, and expertise so that they are embedded into the teaching, learning, and research enterprises. This includes both human and, increasingly, computer- mediated systems. Emphasis should be placed on external, not library-centered, structures and systems. 5. Migrate the focus of collections from purchasing materials to curating content. In the near term, say the next decade or so, I believe that most academic librar- ies will want to pursue all five of these activities (particularly the first, third, and fourth). However, in the longer term, one can easily imagine that one or more of these activities (probably the second and fi h) will become less important on some campuses or will be more effectively managed by regional, national, or inter- national agencies. Part One: Complete the Migration from Print to Electronic Collections There are three types of material to be considered as we look at the migration from print to electronic formats: reference works, journals, and books. The migration is nearly complete for the first two and is just beginning for the third. The conversion of indexes and ab- stracts to electronic formats began in the mid-1980s with the advent of CD-ROMs and was complete by the mid-1990s when Web versions of these products were re- leased. Encyclopedias moved to electronic formats in the same way and in the same timeframe. Legal and business reference works, whose print versions required la- bor-intensive filing, soon followed. These products were clearly superior substitutes for their printed predecessors, and, in most cases, print products were aban- doned. Large aggregated reference sets, such as Gale’s Biography Resource Center or Literature Resource Center, became avail- able on the Web in the late 1990s, and in the early 2000s a wide variety of more specialized reference materials followed. It is less clear that these la er examples of electronic products were treated as substi- tutes for their print counterparts. Beginning with Lexis/Nexis and then IAC’s InfoTrac, full-text journal content started to become available electronically. With indexes and abstracts, this content moved to the Web in the mid-1990s. This journal content expanded as other ag- gregators entered the market and many individual publishers released Web ver- sions of their titles either as freestand- A Strategy for Academic Libraries in the First Quarter of the 21st Century 421 ing products or as supplements to the print. JSTOR then added large backfiles to the mix. By the early 2000s, in most disciplines, nearly all important journal content was available electronically. In most cases, libraries did not treat the aggregator’s products as substitutes for printed versions arguing that the constant changes in these collections’ content made them an unreliable and therefore unac- ceptable substitute. Substitution of the electronic version for the print was more acceptable for individual titles, especially when budgets were constrained. While librarians were moving with caution, users were not. In most libraries, the use of printed journals declined quickly and consistently. This can be tracked by look- ing at photocopying and reshelving sta- tistics. It is also likely that the ease of use and power of the Web indexes, especially when full-text collections were part of the product or where linking services, such as SFX, were employed, increased the use of the journal literature. There should be considerable savings in migrating from print to electronic journal collections both in processing the material and in man- aging the collections. Schonfeld, King, Okerson, and Fenton have documented life cycle savings of from 20 percent to 60 percent. 7 Academic e-books first became avail- able in the late 1990s when netLibrary introduced its first collections. A er a bumpy start, netLibrary and other e-book providers became established in the mar- ket. Readex and others have introduced large retrospective e-book collections. Project Gutenberg has been digitizing and making freely available out-of-copyright titles since the early 1970s, though this effort has had li le impact on library col- lecting. In late 2004, Google created a stir by announcing its partnership with five major research libraries in the Google Print Library Project. The project intended to digitize and make electronically avail- able millions of volumes including the complete collection of the University of Michigan. Shortly therea er, the Internet Archive launched a competing project, the Open Content Alliance, focusing on out-of-copyright titles. To date there has been much talk about e-books, but li le evidence exists to prove that e-books are a suitable substitute for printed books. As a result, there has been li le change in library practice. However, it does not seem unreasonable to suggest that this may change in the near future. Federal documents pose an interesting parallel to e-books. By 2005, 92 percent of all documents distributed to deposi- tory libraries were available in electronic form.8 The University of Arizona, in a pilot program with the Government Print- ing Office (GPO) reduced the number of titles received in dual form to 25 titles.9 Despite the difficult and time-consuming nature of processing federal documents, most depository libraries have been slow to modify their collecting practice, but when they do there should be significant savings of processing costs. It is clear, at least for most reference materials, nearly all journals, and for federal documents, if the University of Arizona experience is generalizable, that electronic versions are at least acceptable substitutes for their paper equivalents. What is less clear is the extent to which libraries have abandoned their print versions and reinvested those resources in other areas. There are clear savings as fewer paper items are processed, as reshelving declines, and as fewer volumes are bound. But I suspect that few librar- ies have clear strategies to manage this migration and how and when they will reclaim resources. Nor do many libraries seem to be in a hurry to move assertively in this direction. An additional area of potential savings is available in the selection of materials. Electronic resources are o en packaged in larger bundles than their printed equiva- lents. In some cases, this bundling is a dis- advantage—for example, when done by the large commercial journal publishers— but in other cases the savings in selection time might be significant. For example, 422 College & Research Libraries September 2007 subscribing to ebrary provides access to tens of thousands of e-books with only one decision rather than the many hours of librarian time that would otherwise be spent on this selection task. User-driven purchase models, like netLibrary’s PDA model, passes the selection task to library users and, at least in some cases, can be markedly more effective than traditional selection.10 However, since models like these threaten to displace the traditional roles of librarians, it is likely that there will be resistance to this change. I believe libraries need to move assert- ively from print to electronic materials and, of equal importance, work diligently to capture the savings this move makes possible. Unless we do the la er, we will lose much of the benefit of the former. Doing both will require discipline, as well as rigorous and continuous assessment of practice. Part Two: Retire Legacy Print Collections As libraries move from print to electronic collections, our legacy print collections will serve a different purpose and we will need to manage them differently. While some print materials will remain important, particularly monographs in the humanities and social sciences, in general, print materials will cease to be the primary part of working collections. Significant efficiencies can be achieved with this shi , particularly in the use of space. In addition, new strategies and funding models will be required for the long-term preservation of and access to this material. If we do not develop clear strategies, our ability to repurpose space will be limited. But we need to keep in mind that sooner rather than later it will become clear to every academic adminis- trator that using prime campus real estate to house li le-used books and journal volumes is unacceptable. Fortunately, the underlying infrastruc- ture upon which this strategy can be built is well established. Many large research libraries and some consortia have con- structed high-density off-site storage facilities and have developed good prac- tice for the management of collections in these facilities and for providing access to them.11 Proposals for regional collection man- agement have been made, and Connaway, O’Neill, and Prabha have shown that OCLC’s WorldCat has the capability to identify unique materials to implement such programs.12 Whether it will be pos- sible to build a national consensus and to implement a concerted program of action or whether a laissez-faire approach will be adequate is unclear. Until one approach or the other is proven to work, individual li- braries will either have to delay decisions or make them on faith. Neither choice will be attractive to tradition-minded librarians who do not wish to antagonize faculty who value proximity to “their” books. An easy exception to this might be the JSTOR journal collection. Many librar- ies may be able to discard these volumes. This was, a er all, the intent of the JSTOR project from its inception.13 If the library community can establish regional or national strategies for the storage and long-term preservation of print collections, then individual libraries can confidently retire, or discard, their legacy print collections, especially those that are available in digital formats, and ultimately move to repurpose high-value campus space. An example of how this might work is being implemented in Indiana for federal documents. Indiana University Bloomington, Purdue University, and the University of Notre Dame have agreed to create a second comprehensive federal documents collection in Indiana. The first is in the Indiana State Library, which is the regional depository library. Much of the current combined collection is housed in the Indiana University Bloomington high- density storage facility, and the plan is to eventually house all of it there. The three universities have agreed to divide the collecting and retention responsibility for the full output of the GPO. Because good http:inception.13 http:programs.12 http:selection.10 A Strategy for Academic Libraries in the First Quarter of the 21st Century 423 bibliographic records are available for post-1976 titles, the comprehensiveness of the collection can be verified. Thus, a complete “light archive” collection of federal documents will be created for the state. A er this agreement was finalized, other depositories in the state were given permission to withdraw post-1976 docu- ments without listing them. This system will provide Indiana depository librar- ies the ability to confidently and easily withdraw unneeded materials from their federal documents collections.14 Part Three: Redevelop the Library as an Informal Learning Space Until very recently, the study space in most libraries was a mix of carrels, tables, and some so seating that was designed to serve individual users. Beginning in the mid-1990s, substantial numbers of public computers were deployed, though most o en these computers were configured and managed as public computer labs that just happened to be located in the library. In the past several years, there has been a concerted effort in many libraries to rethink and redevelop study spaces to create what are generally referred to as the “Information” or “Academic” Commons. The first commons projects generally focused on bringing technology into the library and o en involved collaborations with campus technology organizations. More recent projects have focused on developing spaces that are conducive to group work and involve partnerships with writing centers and other campus groups focused on student success. Multimedia centers and presentation re- hearsal rooms are not uncommon, nor are collaborations with centers that provide technological and pedagogical support to faculty. Following the Barnes and Noble model, coffee shops are becoming the norm. The aim is to create comfortable, lively, and active spaces where students can interact with each other, with infor- mation and with technology and where support for the use of library resources and technology can be found. Increas- ingly, these spaces are being thought of as places to create, as well as to access, knowledge.15 At the same time, libraries are con- fronting the disruption brought on by cell phones and laptop computers, which, because of wireless networks, can be used anywhere in the building, by creating “quiet” study areas. What has become clear is that the relatively homogenous and open space that had been the norm in most library construc- tion since the 1950s no longer works. As Steven M. Foote, an architect involved in library projects, puts it, “As we trace the history of how to accommodate readers in libraries, we are struck by the new paradigms that apply. In every instance—from freshman orientation at liberal arts undergraduate institutions to the most sophisticated post-graduate research—it is apparent that changes are upon us, and that the old programmatic models are no longer adequate.”16 What is needed is a new mix of different kinds of spaces and work environments that can accommodate different uses and possess different ambiances. Library space will need to be shared with a va- riety of partners, and it is likely that the distinction between the library and other informal campus space will blur. With the retirement of paper collec- tions, space should be available to be redeveloped, but in most cases the costs of this redevelopment will be significant. Campus conversations will be required to forge a consensus on the form and function of future library space. The rede- velopment of library space should be an a ractive philanthropic opportunity and will likely be funded in large part with external funds. In the longer term, it may be possible for some space to be returned to the campus for nonlibrary uses. Part Four: Reposition Library Tools, Resources, and Expertise As we think about the future of library services, it is useful to consider OCLC’s http:knowledge.15 http:collections.14 424 College & Research Libraries September 2007 College Students’ Perceptions of Libraries and Information Resources. The conclusion stated: There is widespread high use of general Internet information re- sources among college students. They regularly use search engines, e-mail and instant messaging to obtain and share information. The library is not the first or only stop for these information seekers. Search engines are the favorite place to begin a search and respondents indicate that Google is the search engine most recently used to begin their searches. Among students who have started a search using a search engine, 48 percent ended up at a library Web site. Forty-one percent went on to use the library Web site, but only 10 percent agreed the library Web site fulfilled their information needs. Twenty-seven percent indicated they also had to use other resources. The results of this survey confirm that libraries are not seen as the top choice for access to electronic resources, even among college students who have the highest level of awareness of those resources.17 Undergraduates live on the Web. They begin, and o en finish, their research with Google, and mostly use the library as a place to study. This is a sadly accepted truth among librarians, but we all like to think that faculty and graduate students are different. This might be true for now, but it is quickly changing. A recent University of Minnesota study of faculty and graduate students in the humanities and social sciences concludes by chart- ing a new direction for library services for scholars: Our proposed Scholar’s Collective would address the dual challenge of creating useful tools for humanities scholarship, while simultaneously creating capacity for collaboration… The scope of the Scholar’s Collective addresses two significant cultural shifts in humanities scholarship. The first is the research practices of scholars who depend on electronic media and tools for individual and collaborative work but whose research methods have not yet suc- cessfully incorporated techniques to manage a hybrid information environment. The second is the in- creasingly social dimension of new online environments. By building a comprehensive research environ- ment for humanists that leverages scholars’ expertise and specialized knowledge and that offers personal- ized and customized resources and support for individual and collab- orative research.18 The Scholar’s Collective is not a place; rather, it is a set of tools for the discov- ery, gathering, creating, and sharing of information. It will be Web based, and, while it will have some traditional library functions built into it, it will not be the library. What is most important about both of these studies is that they clearly show that, if the library chooses to stand alone, it will be bypassed. Alternative informa- tion sources may not be as extensive or as authoritative as those housed in or subscribed to by the library, but they are good enough and they fit easily and seamlessly into the lives that our students, and increasingly our faculty, live. For students the primary digital space they will do their academic work in will be the campuses’ course management systems. For faculty, institution-based systems, like Minnesota’s Scholars Collective, may work, but, given the importance of cross-institutional collaboration among scholars, national or international disci- plinary systems might prove to be more effective. Both students and faculty will use the general Web search engines as their http:research.18 http:resources.17 A Strategy for Academic Libraries in the First Quarter of the 21st Century 425 primary discovery tools. Library tools, resources, and expertise need to be where the users are. The simple truth is: if you can’t get to the library from Google, you won’t go there. Libraries need to use link- ing strategies to make this simple and easy. It should also be transparent. There are two strategies that need to be deployed: 1. Libraries need to embed their re- sources and expertise into the systems and tools students and faculty use in their daily lives. We should resist inventing new systems unless there is absolutely no alternative. OCLC’s Open WorldCat and its linking to the Google Book project is a good example of the right approach. Integrating library tools into course man- agement systems should be an obvious priority. 2. Libraries must reposition in-per- son interactions so that they are used to responding to the most complex and difficult problems. The aim should not be to replace in-person interactions and the relationships that are built through them, but rather to find ways to enhance them and to build stronger relationships. Traditional reference desks, even when extended with chat and e-mail, are prob- ably not the best strategy, though it is unclear to me what alternatives will work best. It is also unclear what the best ap- proach to instruction will be, but I suspect a new mix of tutorials, learning tools, and in-person classroom involvement will need to be developed. While the tool- based approach of much of the traditional library instruction activities will probably become less important, new topics such as evaluating the authority of resources, academic integrity, and intellectual prop- erty have entered the library’s domain. There are obvious opportunities to place librarians in centers for teaching and learning and to involve them formally in undergraduate research programs. Blogs aimed at individual courses or depart- mental audiences should be explored, as should a library presence in social spaces like MySpace or Facebook. In larger institutions there may be op- portunities to create systems and services like those envisioned in the University of Minnesota study that could provide data as grist for analysis of the scholarly process. As Rick Luce states, “I believe there is great potential value, for bona fide research institutions, in mining the knowledge space/transaction space rela- tionships in predictive ways that could be every bit as valuable as the publications and research reports that our institutions produce today—and that value has real economic and financial consequences.”19 Part Five: Migrate from Purchasing Materials to Curating Content The transition of information from print to electronic format is clear and its impact is obvious. But there is a second, equally important transition whose impact has not been fully recognized—the transition from purchased to open access content. This second transition will do more to reshape what libraries will be and do in the future than the first, but this has not yet been carefully considered or broadly discussed. The number of open-access journals has steadily increased.20 There is also a growing body of evidence that authors increase the impact of their articles when they are available through an open-access mechanism.21 It has also been argued that open access accelerates the pace at which science develops.22 But between the squabbling of open-access proponents and the o en-misleading rhetoric of com- mercial publishers trying to guard their markets, it is easy to miss the fundamental transformation that is taking place. Peter Suber defines open access as follows: “Open-access (OA) literature is digital, online, free of charge, and free of most copyright and licensing restrictions. OA removes price barriers (subscriptions, licensing fees, pay-per-view fees) and permission barriers (most copyright and licensing restrictions).”23 Open-access literature is free to the user, but it is not produced without cost. Like all informa- http:develops.22 http:mechanism.21 http:increased.20 426 College & Research Libraries September 2007 tion products, open-access literature has an initial fixed cost, what in the print world was called the first-copy cost. This cost has to be covered just as it was in the print world. In the print world the marginal cost of producing each copy of a book or journal was nontrivial and needed to be covered. The only way to do so was to sell the book or journal and pass the first-copy costs and marginal cost of the book or journal on to the reader (or library). Increasing returns to scale are achieved with all information products and sizeable profits can be made once first-copy costs are covered. This happens with best sellers, which is why block- buster authors get large advances. But in the realm of scholarly publishing, this happens less o en and first-copy costs are o en subsidized. Two things have changed with the development of the Internet. First, production or first-copy costs have declined, o en dramatically.24 Second, the marginal cost of distribution of the information product has dropped, for all practical purposes, to zero. For pub- lications that are electronically produced and delivered, fixed costs are lowered and marginal costs disappear. Thus if the low first-copy cost can be covered, the item can be made available at no cost to the user. What can be lost in this analysis is that, while the increasing returns to scale still exist, what is returned is not money, but impact and reputation. Most of the conversation about open access has focused on the scholarly jour- nal literature, but it is more appropriately regarded as any information product where the first-copy cost is subsidized and the product is freely available to the user. This includes a lot of things. Impor- tantly, it includes a lot of what libraries do, including most digital library projects. From the perspective of students and faculty, the growth of open access means that more high-quality scholarly material is freely available (and most easily found with Google or Google Scholar). This frees them from reliance on their campus’s library as the sole source for scholarly materials. Over time, this will mean that the library’s collection of purchased mate- rials, in both print and electronic formats, will be less important. The good news is that as this happens, libraries will be required to purchase less. The especially good news is that this should happen first in the area of science and technology journals where the cost of materials has increased at double-digit rates for several decades. The bad news is that much of what libraries have done in the past is make available purchased collections, and, as this role declines, so may we. It will be critical for libraries to articu- late a change in the role of their collec- tions if they are to remain vital. To do so, I think it is important to recall that most academic libraries have always done two things: 1. They have purchased collections to support their local communities or organizations. 2. They have curated special collec- tions of unique or valuable items for the world. In the past, the first role was dominant. In the future, it will be the second that will become most important. In the past, the collections that were curated were primarily manuscripts and rare books. In the future, the bulk of what is curated will be digital. A part will be digital ver- sions of traditional special collections, but, increasingly, it will be born-digital documents and digital outputs of the research enterprise. Managing the former is reasonably well understood; managing the la er will be a challenge, especially as large data sets become common with real-time ubiquitous data collection in many areas of science (o en referred to as e-science) and the social sciences. Figure 1 is a graphic representation of how we might view this change. Assume that at the present time 80 percent to 90 percent of a library’s resources devoted to collections go into purchasing materi- als. This includes not simply the cost of the materials themselves but also the cost of selecting, processing, and man- http:dramatically.24 A Strategy for Academic Libraries in the First Quarter of the 21st Century 427 FIGURE 1 Transition from Purchased Materials to Curated Content Curated Digital Content Purchased Materials Traditional Special Collections Percent of Library Resources Allocated to Collections 2005 2015 aging these collections. It should also be noted that this includes the purchase of both print and electronic materials. The remaining 10 percent to 20 percent of a typical academic library’s collection resources goes into traditional special collections. I am prepared to predict that, in the next 20 years, less than 50 percent of a library’s collection-related investments will go into purchasing col- lections and over 50 percent will go into curating digital content. Investments in traditional special collections will remain at approximately the current level. I believe it is important to recognize the inevitability of this transition and to embrace it. There are a number of challenges that we will need to meet in this transition. First, libraries need to develop the skills and infrastructures to manage collections of content with which we are not familiar. We also need to develop technologies and strategies for the long-term preservation of digital information. I am reasonably confident that we now know how to preserve bits for decades, but we need to be able to do so for centuries. Among the important skills required will be the ability to assist and support faculty in the creation and collection of this content. I believe this will be an important new role for subject librarians. 60% 10% 2025 A second chal- lenge will be to develop the level of support for this activity to match the level of sup- port that currently exists for purchas- ing content. The v a l u e o f p u r - chased content is clear, especially to the users. Curat- ing content has a similar value, but this value is not always as clear.25 There will be a temptation to be a free rider. Since open access provides information at no cost to the user, why should my campus in- vest in being an information provider? I am convinced that most campuses will quickly come to see the value in curation as it provides researchers with a mecha- nism to share their results and, in the process, brings the researchers and the campus recognition and prestige. The recent recommendations of the Modern Language Association on promotion and tenure that call for a broader range of acceptable venues for scholarly con- tribution, including portfolios and new media, may be an indication of changing views.26 NIH and NSF mandates on data management will also move researchers to value this service. A third challenge will be to balance the benefits of the curation program across the various academic units on campus. Libraries will curate different things for historians than they do for biologists, but they need to be doing something for everyone. The final challenge will be to remain disciplined in making the transition. We cannot build a curation program unless we repurpose resources that are now used to purchase materials. We can expect publishers to make this as difficult as they can, and many librarians will be resistant http:views.26 http:clear.25 428 College & Research Libraries September 2007 to the change, as it will threaten their traditional roles. In my view, an explicit strategy ve ed by the campus will be required. Dri ing and incremental de- velopment will not be successful. Such a strategy will be difficult, because in effect it will require canceling or not purchasing published materials on the assumption that the content will be available in an open-access format. From a broad long- term perspective this might be true, but on an item-by-item day-to-day basis there will rarely be perfect or o en even approximate substitutability. The easiest way to manage might be to constrain the budget for purchased content and require selectors—both librarians and faculty—to live within the budget. It is also likely that a robust document delivery system and/ or a mechanism for the rapid purchase of individual items will be required during the transition period. Putting the Parts Together Three of the parts of the model—migrat- ing from print to electronic collections, retiring legacy print collections, and moving from purchasing to curating collections—represent a change in how the collecting activity is conceived. The third part is a new way of thinking about space. The fourth modifies the way librarians employ their expertise. In all cases, there is a blurring of the boundaries that se p a r a t e t h e l i- brary from the rest of the campus and the external infor- mation environ- ment. The library becomes less of a distinct place. While the dif- f e r e n t p a r t s o f the model can be pursued indepen- dently, there are interdependencies between them. They are shown in figure 2 and described below. 1. The transition from print to elec- tronic resources should provide staff savings as the number of individual print items selected, processed, and managed decreases and more comprehensive electronic resources are acquired. These savings should be both professional and clerical. It will be important to capture and redeploy these resources. In addi- tion, there should be savings in the costs of binding, postage, and cataloging fees. 2. In the short term, the retiring of the legacy paper collection will require additional staff. It should be possible to use staff freed as part of the migration from print to electronic for this purpose as skill sets should be comparable. When the retirement of the print collection is complete, there should be staff savings that should be able to be captured, es- pecially in libraries that do not manage their own high-density storage facilities. The most important resource that will be created with the retirement of the paper collection is space. In the short term, the library will want to retain most of the space to be redeveloped, and any space that is given up should be traded to de- velop relationships with other campus FIGURE 2 A Model for Academic Libraries 2005 to 2025 Complete Migration from Print to Electronic Retire Legacy Print Collection Redevelop Library Space Reposition Library, Tools, Resources, and Expertise Transition from Purchasing to Curating Space $Staff External $ $ $ To Campus A Strategy for Academic Libraries in the First Quarter of the 21st Century 429 organizations that enhance the library’s capacities and mission, such as teaching and learning centers or writing centers. In the longer term, however, it seems likely that some space could be returned to the campus to be repurposed for uses that are not directly related to the library. 3. The redevelopment of library space will require financial resources beyond what can be expected to be recouped from the transition from print to elec- tronic resources. It will have to come from external sources. It seems that this could be a good candidate for philanthropic support, as there should be many naming opportunities. It may also be the case that a “contemporary” library will become a requirement to a ract students much as recreation centers have been in the recent past. It may also be that the promise of space for other purposes will convince some campuses to make investments in the redevelopment of library space. 4. The repositioning of library tools, resources, and expertise will require staff resources and some new investments. Most libraries will require technology skill sets that are not possessed by cur- rent staff. It is not clear if the best strat- egy will be to train existing staff, hire librarians who have the required skills, or hire technologists and instructional designers. I suspect some combination of the la er two approaches will be most successful. It is possible, but not certain, that ultimately this transition will result in net savings in staff resources. While new investments in hardware and so - ware will be required, in the long term it is likely that there will be savings in systems costs as libraries increasingly embed the resources in systems man- aged by others rather than maintaining their own proprietary infrastructure. For example, it is easy to imagine that some combination of WorldCat and Google Books could replace the library’s catalog. Adam Smith, group business-product manager for the Google Book Search and Google Scholar programs, has said about Google’s ambitions, “One of the key a ri- butes of Google Book Search is going to be comprehensiveness. For it to really be a powerful tool, we need to ensure that you can search all the world’s books… what we are really doing is making a discovery tool for books.”27 Again, net savings are possible, but not certain. 5. For most libraries, the migration from purchased resources to curated content will require an input of staff. For- tunately, many of the skills required exist in current employees. For example, mov- ing from cataloging to metadata creation should be straightforward, and subject librarians should be able to be able to support faculty in archiving their research output and developing other digital library collections. New investments in hardware and so ware may sometimes be funded with external support, primarily grants or contracts, but it will be critical for the library to convince the campus of the necessity of moving some funds from collection building through the purchase of materials to the curation function if this transition is to be successful. While it is difficult to predict, I do not think that it is unreasonable to anticipate that the cost of the model described will not be greater than the current cost of operating most academic libraries. Large research libraries that take responsibility for large collections of special or unique materials (for example, area studies col- lections) may require increased resources, and some libraries that can ride free on the increase in open-access materials may require less. On balance, though, it is not unreasonable to expect most libraries to manage without increases in funding beyond the general rate of inflation. Organizational Issues The changes that are necessary will require libraries to be managed in different ways than has been the practice over the past 50 years. The culture in libraries, which dates from the 19th century, is based on care- fully managed and controlled procedures and a conservative approach to change. This made a great deal of sense. We need 430 College & Research Libraries September 2007 to remember that, in the paper world, the most important thing that libraries, particularly large academic libraries, did was to keep millions and millions of small pieces of paper in the correct order. They did other things, of course; but if the small pieces of paper were not in the correct order nothing else ma ered. The current challenges require different approaches and a different culture. Beyond this, there are a number of organizational issues that will need to be managed. Library Staff Composition. As we look out a decade or two and if we assume developments similar to those I have proposed, I think we can make several assumptions about changes in the com- position of library staffing. 1. There will be a reduction in the number of clerical positions. This will also include a reduction in hourly stu- dent positions that do clerical work. This will not begin immediately, as the task of retiring legacy collections will replace the labor that is saved between the reduc- tion in acquiring print and in managing print collections, but within the next two decades we will see reductions in clerical positions. I estimate that this will be in the range a 25 percent to 30 percent decline over the next 20 years. This would take the ratio of clerical staff to librarians from 2:1 to close to 1:1. 2. There will be a continuing increase in the number of technologists. I would anticipate they will represent 25 percent to 40 percent of professional staff by 2025. 3. The number of librarians will re- main roughly constant, but the roles they play will change. Fewer librarians will be involved in the traditional library roles of selecting, processing, and managing purchased collections and in providing their expertise in person, either through reference work or classroom instruction. Librarians will be increasingly involved in new roles of curating collections and providing their expertise in ways that embed it in systems and in other environ- ments. In cases where librarians cannot be found with the skill sets for these roles, libraries will look to staff without library credentials. James G. Neal argues that an influx of non-MLS professionals could create a new vitality in academic librar- ies.28 In my view, the extent to which this takes place will likely depend on the size and specialization of the library. Larger institutions that are developing their own tools will likely require higher levels of specialized skills that few librarians will possess. Smaller libraries that rely on tools provided by others may be be er served by the broader general skills of librarians. The increase in librarian retirements will provide the necessary flexibility, but there will undoubtedly be many challenges.29 4. The net effect of these changes will total compensation levels (in constant dollars) that are approximately what they are today. Flexible Staffing and Flexible Staff. The next several decades will be full of change. The adaptability of staff and the ability of the library to have staff with the required skill sets to try and succeed at new things will be critical. This will be a complex challenge, and it will require at least the following: 1. An organizational culture that val- ues learning and is willing to experiment even when success is not assured. 2. An explicit strategy for hiring and retaining staff with the skills, abilities, and characteristics the organization re- quires. 3. A willingness to invest in staff de- velopment. 4. A commitment to organizational development. Library staff will need to recognize that they are unlikely to be doing, ten or even five years hence, the same things they are doing now. They also need to prepare themselves to acquire the skills needed to play the new roles that will be required. The Principles of Disruptive Innovation. Clayton Christensen and his colleagues have developed strategies designed to http:challenges.29 A Strategy for Academic Libraries in the First Quarter of the 21st Century 431 create success when introducing inno- vative or disruptive programs or tech- nologies.30 Among others, the strategies libraries need to pursue include: 1. Make products and services more reliable, more convenient, and cheaper (as measured in the user’s time if not in dollars). This should be a constant and never-ending quest. 2. Use exploratory project develop- ment strategies that ensure learning rather than success and that preserve re- sources for the second and third a empts at ge ing it right. 3. Be impatient for success with small projects, but don’t be in a hurry to grow the project to full scale. This will push the exploration of new ideas but avoid risking immature developing projects by banking on them too heavily. Done correctly, this will drive innovation. 4. Begin with simple projects that meet the needs of undemanding users and then move up market to provide services to more demanding users. In practice, this means beginning with services to students and only moving to faculty services when some expertise has been developed. This is contrary to the approach academic libraries usually employ. 5. Don’t ask users what they want; rather, watch what they do with the tools you provide. Our users cannot anticipate how the new technologies will solve their problems any more than we can. Especially watch new users who are unencumbered by old systems and practices. 6. We should encourage standards that allow for modularization of the scholarly information value chain. This will make it more difficult for for-profit or other large enterprises to gain monopoly control of pieces of the value chain and thus extract unreasonable income from that control. This is what has happened with scholarly journals over the past three decades, and we should work to keep this from happening in the developing information ecology. 7. Add value where things are “not good enough.” Studies like the one conducted by University of Minnesota Libraries show where the possibilities lie. In general, what libraries have done in the past works “well enough” and is not where we should look for future op- portunities. 8. We should use technology to create new approaches that are scalable and save time for both the user and the library. We should look outside the library world for trends and inspiration. As Christensen, Baumann, Ruggles, and Sadtler put it: What accounts for this poor showing [of U.S. health care and education]? It’s not a lack of solutions but rather misdirected investment. Too much of the money available to address social needs is used to maintain the status quo, because it is given to or- ganizations that are wedded to their current solutions, delivery models, and recipients… What’s required is expanded support for organizations that are approaching social-sector problems in a fundamentally new way and creating scalable, sustain- able, systems-changing solutions.31 We cannot continue to operate as we have and must look outside our world for new ideas and solutions. Conclusion Seven years into the new millennium, academic libraries are facing a great deal of uncertainty. The structure of scholarly communication is changing, as is the role academic libraries will play in it. But it still seems to me that the way forward is really not that difficult to envision, at least in its broad outlines. The challenges we face are complex in detail; and some, most notably the long-term preservation of dig- ital objects, will take both inspiration and hard work. But none of what needs doing is beyond our capabilities. Moreover, the work that needs to be done is at the core http:solutions.31 http:nologies.30 432 College & Research Libraries September 2007 of what libraries have always done—mak- ing knowledge available in communities and organizations. We will use new and different techniques for doing so and we will undoubtedly define community somewhat differently—more o en as the world and less o en as the campus. But our underlying values need not change. As individuals, we will need to be ready to invest in ourselves by acquiring new skills and looking at problems in new ways, but the work will serve the same end and will probably have many of the same frustrations and rewards. Notes 1. Yogi Berra, The Yogi Book (New York: Workman Publishing, 1998), 102. 2. John Lombardi, “Library Performance Measures That Ma er,” Library Assessment Confer- ence: Building Effective, Sustainable, Practical Assessment, Charlo esville, Virginia, September 25, 2006. 3. Jerry D. Campbell, “Changing a Cultural Icon: The Academic Library as a Virtual Destina- tion,” EDUCAUSE Review 41 (Jan./Feb. 2006): 30. Available online at www.educause.edu/apps/ er/erm06/erm061.asp. [Accessed 22 April 2007]. 4. “Changing Roles of Academic and Research Libraries,” Roundtable on Technology and Change in Academic Libraries, convened by the Association of College and Research Libraries (ACRL) on November 2–3, 2006, in Chicago, Illinois. Available online at www.ala.org/ala/acrl/ acrlissues/future/changingroles.htm. [Accessed 22 April 2007]. 5. See David W. Lewis, “What if Libraries Are Artifact-Bound Institutions?” Information Technology and Libraries 17 (Dec. 1998):191–97. Available online at h ps://idea.iupui.edu/dspace/ handle/1805/434. [Accessed 22 April 2007]. 6. See David W. Lewis. “The Innovator’s Dilemma: Disruptive Change and Academic Librar- ies,” Library Administration & Management 18 (Spring 2004): 68–74. Available online at h ps://idea. iupui.edu/dspace/handle/1805/173. [Accessed 22 April 2007]. 7. Roger C. Schonfeld, Donald W. King, Ann Okerson, and Eileen Gifford Fenton, The Non- subscription Side of Periodicals: Changes in Library Operations and Costs between Print and Electronic Formats, Council on Library and Information Resources Report 127, June 2004. Available online at www.clir.org/pubs/abstract/pub127abst.html. [Accessed 22 April 2007]. A summary of the results is reported in: Roger C. Schonfeld and Eileen Gifford Fenton, “Digital Savings,” Library Journal 130 (Mar. 1, 2005):50–51. 8. See: Judith C. Russell, “Remarks of Superintendent of Documents Judith C. Russell,” De- pository Library Council Meeting, Albuquerque, New Mexico, April 17, 2005, 7. Available online at www.gpoaccess.gov/about/speeches/jrussell_gpo_update.pdf. [Accessed 22 April 2007]. 9. See Atifa Rawan and Cheryl Kno Malone, “A Virtual Depository: The Arizona Project,” The Reference Librarian 94 (2006): 5-18. 10. See Lynn Su on, “Collaborating with Our Patrons: Le ing the Users Select,” ACRL Eleventh National Conference April 10–13, 2003, Charlo e, North Carolina. Available online at www.ala. org/ala/acrl/acrlevents/lsu on.pdf. [Accessed 22 April 2007]. Similar conclusions can be reached by reworking the data presented in Justin Li man and Lynn Silipigni Connaway, “A Circulation Analysis of Print Books and E-Books in an Academic Research Library,” Library Resources & Technical Services 48 (Oct. 2004): 256–62. 11. See for example: Willis E. Bridegam, A Collaborative Approach to Collection Storage: The Five- College Library Depository, Council on Library and Information Resources Report 97, June 2001. Available online at www.clir.org/pubs/abstract/pub97abst.html. [Accessed 22 April 2007]; Bernard F. Reilly, Jr., and Barbara DesRosiers, Developing Print Repositories: Models for Shared Preservation and Access, Council on Library and Information Resources Report 117, June 2003. Available online at www.clir.org/pubs/abstract/pub117abst.html. [Accessed 22 April 2007]; Sco Seaman, “High- Density Off-Site Storage: Document Delivery and Academic Library Research Collections,” Journal of Interlibrary Loan, Document Delivery & Information Supply 13 (2003): 91–103. Available online at www.haworthpress.com/store/E-Text/View_EText.asp?a=3&fn=J110v13n03_05&i=3&s=J110&v=13. [Accessed 22 April 2007]; and Ann MacKay Snowman, “The Penn State Annex: The Life and Times of an Off-Site Storage Facility,” Collection Management 30 (2005): 45–53. 12. See John Burger, Paul M. Gherman, and Flo Wilson, “ASERL’s Virtual Storage/Preserva- tion Concept,” ACRL Twel h National Conference, Minneapolis, Minnesota, April 7–10, 2005. Available online at www.ala.org/ala/acrl/acrlevents/burger-etal05.pdf. [Accessed 22 April 2007]; www.ala.org/ala/acrl/acrlevents/burger-etal05.pdf www.haworthpress.com/store/E-Text/View_EText.asp?a=3&fn=J110v13n03_05&i=3&s=J110&v=13 www.clir.org/pubs/abstract/pub117abst.html www.clir.org/pubs/abstract/pub97abst.html www.gpoaccess.gov/about/speeches/jrussell_gpo_update.pdf www.clir.org/pubs/abstract/pub127abst.html www.ala.org/ala/acrl www.educause.edu/apps A Strategy for Academic Libraries in the First Quarter of the 21st Century 433 and Lynn Silipigni Connaway, Edward T. O’Neill, and Chandra Prabha, “Last Copies: What’s at Risk?” College & Research Libraries 67 (July 2006): 370–79,. A preprint version of the paper is avail- able online at www.oclc.org/research/publications/archive/2006/connaway-crl07.pdf. [Accessed 22 April 2007]. 13. “Originally conceived by William G. Bowen, President of The Andrew W. Mellon Founda- tion, JSTOR began as an effort to ease the increasing problems faced by libraries seeking to provide adequate stack space for the long runs of backfiles of scholarly journals.” From the “History of JSTOR” page on the JSTOR Web site; www.jstor.org/about/background.html. [Accessed 22 April 2007]. 14. An outline of the program can be found at the Academic Libraries of Indiana Web site at h p://ali.bsu.edu/. [Accessed 22 April 2007]. 15. There is a large quantity of information on Information Commons available. A Google search on “library information commons” is a good start. See also Joan K. Lippinco , “Linking the Information Commons to Learning,” in Learning Spaces: An EDUCAUSE e-Book, ed. Diane G. Oblinger (Washington, D.C.: EDUCAUSE, 2006). Available online at www.educause.edu/ LearningSpaces/10569. [Accessed 22 April 2007]. The papers from the “Information Commons: Learning Space Beyond the Classroom: Leavey Library 2004 Conference,” September 16 and 17, 2004. Available online at www.usc.edu/libraries/locations/leavey/news/conference/presenta- tions/. [Accessed 22 April 2007]. “Collaborative Facilities,” available online at www.dartmouth. edu/%7Ecollab/index.html. [Accessed 22 April 2007]. For a good summary of recent academic library building trends, see Harold B. Shill and Shawn Tonner, “Creating a Be er Place: Physical Improvements in Academic Libraries, 1995–2002,” College & Research Libraries 64 (Nov. 2003): 431–66; and Harold B. Shill and Shawn Tonner, “Does the Building Still Ma er? Usage Pa erns in New, Expanded, and Renovated Libraries, 1995–2002,” College & Research Libraries 65 (Mar. 2004): 123–50. 16. Steven M. Foote, “Changes in Library Design: An Architect’s Perspective,” portal: Libraries and the Academy 4.1 (Jan. 2004): 42. 17. Cathy De Rosa, College Students’ Perceptions of Libraries and Information Resources: A Report to the OCLC Membership A Companion Piece to Perceptions of Libraries and Information Resources, Dublin, Ohio: OCLC Online Computer Library Center, Inc., 2006. Available online at www.oclc. org/reports/perceptionscollege.htm. [Accessed 22 April 2007]. 18. A Multi-Dimensional Framework for Academic Support: A Final Report, Submi ed to the Andrew W. Mellon Foundation from the University of Minnesota Libraries, June 2006. Available online at www.lib.umn.edu/about/mellon/docs.phtml. [Accessed 22 April 2007]. 19. Rick Luce, personal e-mail communication (Jan. 3, 2007). 20. The number of titles included in the DOAJ: Directory of Open Access Journals has increased by about 500 per year since its founding in May of 2003, with 350 titles until December 2006 when it topped 2,500 titles. See DOAJ news reports online at www.doaj.org/doaj?func=loadTempl&te mpl=news. [Accessed 22 April 2007]. 21. The Open Citation Project maintains a good current bibliography of this work. It can be found online at h p://opcit.eprints.org/oacitation-biblio.html. [Accessed 22 April 2007]. As the introduction to the bibliography states, “Why might open access be of benefit to authors? One universally important factor for all authors is impact, typically measured by the number of times a paper is cited (some older studies have estimated monetary returns to authors from article publication via the role citations play in determining salaries). Recent studies have begun to show that open access increases impact.” 22. Alma Swan, “Open Access and the Progress of Science: The Power to Transform Research Communication May Be at Each Scientist’s Fingertips,” American Scientist 95 (May–June 2007]. Available online at www.americanscientist.org/template/AssetDetail/assetid/55131. [Accessed 22 April 2007]. 23. Peter Suber, “Open Access Overview.” Available online at www.earlham.edu/~peters/fos/ overview.htm. [Accessed 22 April 2007]. 24. A good analysis of this for journals is provided in Andrew Odlyzko, “The Economics of Electronic Journals,” First Monday 2 (Aug. 1997). Available online at h p://firstmonday.org/is- sues/issue2_8/odlyzko/index.html. [Accessed 22 April 2007]. 25. At this time it does not seem that many librarians are prepared to invest in the curation function. A recent survey by Palmer and Dill indicates that only 21 percent of librarians strongly agree that providing financial support for open access should be a library priority, and only 13 percent strongly agree that existing resources should be reallocated to open access projects. See Kristi L. Palmer and Emily Anne Dill, “Preaching to the Choir? How Academic Librarians Really Feel About Open Access,” Electronic Resources & Libraries 2nd Annual Conference, Atlanta, Georgia, February 23, 2007. Available online at h p://hdl.handle.net/1805/706. [Accessed 22 April 2007]. 26. MLA Task Force on Evaluating Scholarship for Tenure and Promotion (2006). Available online www.earlham.edu/~peters/fos www.americanscientist.org/template/AssetDetail/assetid/55131 www.doaj.org/doaj?func=loadTempl&te www.lib.umn.edu/about/mellon/docs.phtml www.oclc www.dartmouth www.usc.edu/libraries/locations/leavey/news/conference/presenta www.educause.edu http:h�p://ali.bsu.edu www.jstor.org/about/background.html www.oclc.org/research/publications/archive/2006/connaway-crl07.pdf 434 College & Research Libraries September 2007 at www.mla.org/resources/documents/ten- ure_promotion. [Accessed 22 April 2007]. 27. Daniel Greenstein, Adam Smith, and Danielle Tiedt, “The Library as Search En- gine,” Chronicle of Higher Education 53 (Jan. 5, 2007): B24. 28. James G. Neal, “Raised by Wolves: Integrating the New Generation of Feral Professionals into the Academic Library,” Library Journal 131 (Feb. 15, 2006): 41–44. Available online at www.libraryjournal. com/article/CA6304405.html. [Accessed 22 April 2007]. 29. Denise M. Davis, “Library Retire- ments: What We Can Expect,” American Libraries 36 (Sept. 2005): 16. 30. See Clayton M. Christensen and Michael E. Raynor, The Innovator’s Solution: Creating and Sustaining Successful Growth (Boston, Mass.: Harvard Business School Press, 2003); and Clayton M. Christensen, Sco D. Anthony, and Erik A. Roth, Seeing What’s Next: Using the Theories of Innovation to Predict Industry Change (Boston, Mass.: Harvard Business School Press, 2004). 31. Clayton M. Christensen, Heiner Baumann, Rudy Ruggles, and Thomas M. Sadtler, “Disruptive Innovation for Social Change,” Harvard Business Review 84 (Dec. 2006): 94–96. www.libraryjournal www.mla.org/resources/documents/ten