Panel 3: Into the Glass Darkly: Future Directions in the 21st Century

Journal of Library Administration, 49:281–301, 2009Copyright © Taylor & Francis Group, LLCISSN: 0193-0826 print / 1540-3564 onlineDOI: 10.1080/01930820902785397

Panel 3: Into the Glass Darkly:Future Directions in the 21st Century

KAREN HUNTERElsevier

DONALD WATERSScholarly Communications, Andrew W. Mellon Foundation

LIZABETH WILSONUniversity Libraries, University of Washington

INTRODUCTION BY FRED M. HEATHUniversity of Texas Libraries, University of Texas at Austin

ABSTRACT. Panelists look at publishing and scholarly commu-nications from varied perspectives and discuss possible futuresfor interaction between publishers, libraries, and the researchuniversity.

KEYWORDS publishing, scholarly communications, digital in-frastructure, libraries as change agents, collaboration

FRED M. HEATH

It’s my pleasure to continue this colloquy and to introduce my three col-leagues this afternoon. We have asked them to peer into the future and togive us their perspectives on the directions of scholarly communications inthe remainder of this century and to suggest the roles that libraries may playin shaping those vectors, or trajectories.

It’s not chance that brings these three panelists together to considerdirections in scholarship. Each is well known in this research community,and all are in high demand for the acuity of their vision and the credenceof their observations pertaining to libraries, to research, and to scholarlycommunications.

Address correspondence to Karen Hunter, Elsevier 125 Park Avenue, 360 Park AvenueSouth, New York, NY 10010, USA. E-mail: [email protected]

281

282 K. Hunter et al.

We have asked them to take a look beyond the renewal of next year’sjournal subscriptions, the implementation of the next piece of software, orthe construction of the next library or storage center. We have asked them tothink about how scholarly communications is unfolding in the technology-enabled digital age and to suggest what fundamental transformations therenowned libraries represented in this meeting might undergo and whatwe each may do to enable teaching, learning, and research over the ensuingdecades. The perspectives of our presenters will be useful to all in this room,and I would like to introduce them in the order of their presentations.

Karen Hunter, senior vice president for the publisher Elsevier, will bethe first to make her remarks. It’s appropriate that she do so. Along with theimportant roles played by scholarly societies, the rise of commercial scholarlypublication in the years following World War II and continuing to the presentenabled our research programs to accelerate and allowed scholarly colloquyto unfold in all of its richness.

For several years, Karen has concentrated on the strategic issues thatconfront Elsevier in the evolving scholarly landscape. She has been a pivotalplayer in her company’s effort to position itself for the electronic deliveryof journal information. Those of you who are librarians or scientists will re-member the early effort in network journal delivery, The University LicensingProgram (TULIP), and Elsevier’s effort in the early part of the past decade,for which she was responsible.

Karen led the design and start-up of Science Direct, Elsevier’s Web-based journal service. She is a member of the copyright committee of theAssociation of American Publishers and a member of the board of CrossRef,and she has served on the board of the International Digital Objects IdentifierFederation. She serves us well on the National Research Council’s StudyCommittee on Intellectual Property and Emerging Information Infrastructure,and she also serves on the National Commission on Library and InformationServices Working Group with regard to issues of journal pricing, publishing,and copyright.

Following Karen will be Donald Waters, program officer for scholarlycommunications at the Andrew W. Mellon Foundation. The Mellon Foun-dation has played an important role in enabling bold initiatives in highereducation. Its role in the field of research librarianship certainly has beenpivotal. In the 6 years since Don joined the Mellon Foundation, the scholarlycommunications program has awarded more than $200 million in grants,including support for products known to scholars, such as ARTstor, and ini-tiatives embracing the future of digital preservation, such as LOCKSS, Ithaka,and Portico.

Before joining the Mellon Foundation, Don engaged new directions inscholarly communications as the assistant university librarian at Yale Univer-sity and as the first director of the Digital Library Federation (DLF). He isa fellow of the American Association for the Advancement of Science and

Panel 3 283

serves on the Steering Committee of the Coalition for Networked Informa-tion. He is on the National Digital Advisory Board of the Library of Congress,as well as the Section 108 Study Group. Don is a recognized expert on digitalpreservation, digital libraries, and scholarly communications.

“Cleaning up” will be my colleague Elizabeth Wilson, director of uni-versity libraries at the University of Washington (UW). Betsy has a longrecord of distinguished service in research librarianship, having held im-portant leadership positions at both UW and the University of Illinois atUrbana-Champaign.

As a visionary and leader, she has served the Association of ResearchLibraries (ARL), the American Library Association, and the Online ComputerLibrary Center (OCLC) with true distinction. She was president of the Associ-ation of College and Research Libraries (ACRL) and currently is chair of theOCLC Board of Trustees. She also is a member of the ARL Board and serveson the DLF Executive Council. Her contributions and her vision have beenrecognized by many personal awards and by the libraries that she directs. In2004, her library system was selected as the recipient of the ACRL Excellencein Academic Libraries award.

KAREN HUNTER

As Paul Saffo of the Institute of the Future said, we are in a period ofunprecedented uncertainty. It would be presumptuous for me to say whatresearch libraries should do, and you already have a long list from the otherspeakers. Following the briefing that Fred Heath gave us as panelists, I’mgoing to try to talk more about what I see as some of the challenges inpublishing, what we are doing to try to deal with those, and how that maychange or affect our relationship with libraries or how we work with researchlibraries going forward.

First, for the benefit of the many people from the humanities who areat this conference, permit me to spend one minute just giving a very rapidintroduction to Elsevier. We have a staff of about 700 and more than 70offices in 24 countries. About 45% of our business is in the health sciences,and 55% is in the physical, life, and social sciences. We are not humanitiespublishers.

We publish about 2,200 new books a year and 1,800 journals. The 1,800journals publish about a quarter of a million articles a year, which is around1,000 articles per business day. It’s about a quarter of the scholarly literaturethat is regularly referenced.

Our online platform, Science Direct, has more than 9 million articles.In most cases, our back issues reach to volume 1, number 1 of those 1,800journals. On a 24–7 basis, we average 500 article downloads every minute.Over the past decade, we’ve invested about $400 million in Science Directand Scopus, our bibliographic database.


So, in the publishing community, we’re the 800-pound gorilla. We arethe largest of the academic publishers. However, when you compare uswith the University of Texas (UT), we have about the same budget size; so,when you look at the whole academic community, we suddenly becomevery small. If you look at the entire scholarly publishing community, therevenues in any one year are smaller than the profits of Exxon Mobil. It’s akind of strange world in many respects.

One reason for giving this quick overview is to say that, as a community,we’ve come a long way in this electronic era in the past decade. Enormousamounts of scholarly information are available immediately at the desktopsof researchers and students—in our case, at more than 16 million desktopsworldwide. I think that we should not leave this conference on the future oflibraries without feeling good about what has already been accomplished.The infrastructure and the ability that the libraries have to get this informationto researchers and students are extraordinary.

Challenges Facing Publishers

As we know, past success does not assure future success. I want to quicklylook at some of the challenges that I see publishers facing, and I’m sure thatmany of these will resonate with the libraries as well.

CHANGE IN ROLES

The first is a change in roles. The traditional cycle of a research publicationmoves through about 10 stages: the research itself; the writing of the resultsof the research; editorial review and acceptance; production, which manyequate with publication; distribution, including marketing; the purchasingprocess; storage for day-to-day use; archival, permanent storage; access andlocation tools; and then, finally, the use and incorporation into research andteaching, which completes the cycle.

In the print environment, the stakeholders were pretty linear. Scholarsdid the research and wrote; scholars performed the editorial review, whichwas enabled by a publishing infrastructure; and publishers then took overto produce, distribute, and—not incidentally—finance. Libraries purchased,stored, archived, and provided the finding tools, and scholars picked upagain to use the information and incorporate it into new research and newteaching.

Over the past decade, there’s been a significant shifting of rolesdownstream—or the potential for a shift—which has been creating uncer-tainty. Researchers on the Internet obviously can take care of production anddistribution themselves, if they choose to. Libraries also have become dis-tributors as they establish institutional repositories. Purchasing models havebroadened as a result of the call for some type of open access and other open

Panel 3 285

activities. Publishers have taken over much of the storage responsibility, andwe share archival responsibility with libraries.

At least some publishers—certainly in our case—have become informa-tion managers with a staff of highly skilled experts in searching, text mining,and the many related areas needed for today’s access to and use of products.We’ve become preoccupied with having our publications measurable andmonitored, because what is not used will not be bought. So, there’s overlapand a potential confusion in roles—what was linear is no longer linear.

LIMITED BUDGETS

The second challenge is the limitation in growth afforded by our traditionalmarket, the research library. It’s not news to anyone in this room that librarybudgets have not kept pace with the growth in research. Research has beengrowing at a steady 3% per year for most of the previous century. Forlibraries, that has meant years or even decades of very hard choices. Forpublishers, that has meant equal frustration and the need to adapt to flatmarket conditions.

SCHOLARLY COMMUNICATIONS

A third challenge is the changing nature of the scholarly community andits interactions. We’ve been hearing a lot about that during this conference.As we’ve heard and as we know from our own experience, researcherstoday are essentially in instant and constant communication. What is notclear is how these changing communication patterns among researchers willchange the formal publishing process—that is, the formal part of scholarlycommunications. As we see new collaborative tools, how are these toolsgoing to change what’s expected in formal publication?

We hear predictions of the decrease in the importance of thejournal—perhaps even its ultimate death. I was at a conference at UC Berke-ley in June that included a number of faculty members. I asked whetherthey would care about journals without the “publish or perish” pressure inacademia. At least one prominent faculty member said, “No, get rid of it.Journal articles represent a purely publish or perish kind of activity.” I won’tsay that everyone else signed on to that opinion, but it was an interestingresponse.

DISCIPLINARY DIFFERENCES

A fourth challenge is the understanding of the idiosyncrasies of differentdisciplines. Publishers have spent the past 15 years creating superior buthomogenous online solutions. We’ve worked very hard to try to get our


costs down and to get a robust infrastructure in place. In doing that, we’veerased the distinctions in research behavior between different communities.At this point, I strongly believe that we need to go back and understandmuch better what is needed to support specific disciplines. At the sametime, we need to facilitate the growth of interdisciplinary research.

Everywhere I go, I hear about interdisciplinary research and the needto make the literature, the resources, and the authorities of one disciplinemore transparent to those outside that discipline. We’re being told that, whenyou find research information on the Web that is outside your immediatediscipline, it’s very hard to know whether that information comes from a realauthority in that field or whether it is, as they say, “a dog.” You never knowon the Internet.

NONTEXTUAL MATERIAL

A fifth challenge is the increasing importance of nontextual material. CliffordLynch talked about the importance of data retention and storage. There alsohas been the mention of images, and I think that images are seriously under-used and not as accessible as they could be. I’m thinking very much of whatis within our books and within our journals. We have an enormous repertoryof images that we should be making much more accessible to people.

We’re also seeing some increase in other nontextual items, althoughthis increase has been much slower than any of us expected, simply becauseit’s still hard for scholars to get nontextual information into a usable format.Sometimes it’s just more work than it seems to be worth. Nevertheless, as weknow in the e-science area and other areas, this becomes a core of researchin many ways, and we have to find better ways of dealing with that.

RETURN ON INVESTMENT

The final challenge that I’ll mention is probably the largest problem for ascholarly publisher—at least for a commercial publisher—but it affects thenot-for-profits as well. That challenge is how to get a return on the investmentin products and services. I said that we’re challenged by flat library budgets,but the problem is far more systemic. As a culture, we’ve become used towhat I call the “PC mentality.” By “PC,” I don’t mean “politically correct” butthe example of a personal computer. We expect that every year we’re goingto get a better service with more functions and more features and morefunctionality and that it’s certainly going to be no more expensive than inthe previous year and probably cheaper.

That expectation—that you’re going to get something better all the timefor less money—is a good one, but it’s a challenge and an important thingto consider. At the same time, it makes it difficult to know where to invest

Panel 3 287

and where you’re going to get things. In the academic environment, there’s amentality among researchers—I’m not saying among libraries—of not beingused to paying for things, and the Internet reinforces that mentality. So, it’sdifficult to know where you can add value and where you can create newservices in a way that you can actually monetize and get any return.

Yesterday, I was asked by Duane Webster to talk for one moment aboutthe reaction to the call for open everything—you can fill in the blank, butlet’s say open access. I will say quite bluntly, how could I be opposed toit? If there’s a sustainable way to offer information to the user for free, thenthat would be wonderful. None of us deny, however, that there are costsinvolved; as a commercial company, we also have an obligation to providea return to our shareholders. The answer is to get the return from the value-added services. Again, I have to know what somebody is willing to pay for.We have an enormous number of ideas, but trying to translate those intosomething that someone’s really willing to pay for is not so simple.

My boss, our new vice chairman, Y. S. Chi, frequently has said, “If thecurrent business model is really unacceptable to the academic community,if it really doesn’t work, if it’s broken, buy us out.” Make us an offer that wecan’t refuse, and then take us over. You can have our 1,800 journals, andyou can run them within the university. In the absence of that, however, weall continue to try to find ways to support the current needs for services.

Responding to Challenges

If these are some of the challenges—changing roles, flat library budgets, thechanging nature of communication within the scholarly community, under-standing discipline idiosyncrasies and interdisciplinary needs, the increasingimportance of data and other nontextual material, and the absence of amarket willing to pay for innovation—what has been our response?

First, we’ve placed an even higher priority on making sure we havehigh-quality content. We’re still content providers, and if we don’t have high-quality content, it’s not going to be used and it’s not going to be purchased.So, “think quality first” is engraved on everyone’s mind.

We know that we need to get and stay closer to our customers. We re-ally are very focused now on product development in a collaborative mode,talking and testing in a rapid fashion, and trying to make changes and dealwith people. When we developed Scopus, we had about 20 developmentpartners around the world, which made an enormous difference. We’re re-alistic about the budget situation of the research library, and we actuallyare looking to other markets in which there is more of an opportunity forgrowth. In our case, that means the health sciences. We put a lot of emphasison the health sciences because it’s a different kind of market and because itdoes still have opportunities for high growth.


At the core, though, I think that we are defining—or, perhaps,redefining—our business. We’re redefining it as providing tools and servicesthat make learning, including research and teaching processes, more effi-cient. We’re not defining our business as content delivery per se. Content de-livery is a part of efficiency and productivity but not the underlying definition.

That leaves us with a new set of challenges, including the need to reallyunderstand what will make the research and educational processes moreefficient. To do that, we decided that we needed to map the entire workflowprocess, to identify needs that can be filled and, since we are a commercialcompany, opportunities that will result in revenues and profit. We recentlylaunched an extensive project to look at the roles of researchers and thetasks that researchers perform.

We’ve classified the roles and tasks into five broad groups: core research,which includes grants, writing, working with team members, and stayingup-to-date; contributive research, which is editorial work and review andconference organization and attendance; education, which is teaching andsupervision; administration, which is evaluation and research assessment;and personal career development.

We’ve seen common elements among these groups, such as searchingfor information, organizing information, analysis, collaboration, and so forth,and we’re now looking to identify needs. We’ve already generated manysubstantial ideas. I participated in a 2-day workshop that brought togethersome of these ideas and tested them, and we certainly walked out withmore than 15 ideas that we thought were really viable. We now will startthe process of going into the market, with development partners, to ask isthis nonsense or is this real? Is this something that we can do, or are we justthinking crazy? If we did do it, would you buy it?

What kinds of needs? Some examples include better filtering of theresults of the search process, better management of grants and grant appli-cations, management of international multisite collaborative research and itsdata sets, improvement of time management in general within the researchprocess, and the need to understand who’s who on the Internet. There cer-tainly are other things that we need to do internally. We need to seriouslyrethink the organization of information and make it more suitable for dataand text mining. We need to deal more with data sets, and we need to thinkabout how we license on a more granular level, to have material availablefor use in a different way.

As we look more deeply at the workflow process and scholars’ needs,we find new opportunities for publishing, so we’re optimistic that opportu-nities are there. Maybe “publisher” is not the word anymore. We kiddinglysay that, in some ways, we need to be an uberpublisher, but there are newtools and new opportunities that we can develop that will keep us in thisbusiness.

Panel 3 289

How does this affect libraries? I’m going to list a few areas in which Ithink libraries and publishers may differ. We probably have different plan-ning horizons. The planning horizons of publishers have narrowed, for betteror worse. Typically, we now talk about 2 years and avoid trying to projectfurther out than that. Other than when we know that we need to make alonger term investment, we’re usually on a 2-year planning horizon.

We probably take a different view on costs, although we’re very costconscious, and one of our strategic goals is always to keep costs down.If there are revenue opportunities that exceed cost, we’ll go for high cost.It’s not that everything has to be cheap. If we can build a product with ahigh cost but can make money on it, then we’ll do it. We probably havemore flexibility. We can move our offices, we can outsource fairly read-ily, and we can redefine our business as we need to. Perhaps the criticaldifference—although maybe not—is that we are risk takers by definition. It’sone of our core competencies. We have access to serious resources, we’remaking huge investments, and we’re willing to take those risks most of thetime.

We also have many things in common. On the challenges side, we’reboth losing our captive audiences, and we both face the potential and per-haps already actual disintegration of the structure of what we do. We bothneed expertise on our staffs that is very different from that of the past, and Iwould argue that we both need business redefinition. I’ve already said thatwe’re redefining our business, and I think that what we’ve heard during thisconference suggests that libraries certainly are as well.

On the positive—or strength—side, we both have a passion for thequality of information and services, we’re adaptive and open to innovation,we’re collaborative and used to working together, and we have a networkof established relationships and contacts within the research community thatwe can use to test ideas and refine needed content and services. I also wouldsay that we have a mission to delight our customers. We want to do thingsthat will make scholars happy and pleased.

Conclusion

In his briefing to the panel, Fred Heath said that the challenge put to theUT Libraries was to rank with those of the best universities in the world.As they move through the process of defining what that means, they mustask who those libraries are, what makes them highly ranked, and what gapsneed to be filled. I suggest that the question also should be whether this isthe best comparison. I would put high on the list the real integration of theworkflows of the university, of researchers, and of the services that makescholars and students more productive and more efficient.

I think that a heavy focus on the users and the consumer is going to bekey. If you do that, then I’m not sure with whom you should evaluate and


compare yourself. It may be corporate libraries, it may be corporations ingeneral, or it may be somebody else completely different. It may not be justother libraries.

I optimistically believe that publishers, or uberpublishers, will continueto be an important part of the research and educational community. Our roleswill broaden or change. I don’t think that you can get rid of us easily, unlessyou want to buy us out. Absent that, what I would invite is all types of collab-oration. We will come to you asking to collaborate, but please don’t hesitateto come to us. I think that, together, we can delight the scholarly community.

DONALD WATERS

Given what has already been said in this forum by so many distinguishedspeakers, the possibility of saying anything new, I think, is rapidly dimin-ishing. I am especially reluctant to pronounce on the future, and I do notpretend to offer answers here. Instead, I take this opportunity to share someof what we are learning at the Andrew W. Mellon Foundation about theprincipal themes that are shaping the interaction among libraries, publish-ers, and scholars in the field of scholarly communications and to ask somequestions that might guide strategic thinking into the future.

First, a key principle of grant making at the Mellon Foundation is that itbe evidence-based. As much as possible, our programs in scholarly commu-nications proceed from systematic consultations with our library, publishing,and scholarly constituents, or from systematic studies of parts of the schol-arly communications field. By sponsoring these consultations and studies,we seek to inform our own funding strategies. Our hope is that they alsoinform the policies and strategic choices of our constituents.

This year, several important Mellon Foundation–funded studies havebeen or are about to be released. As has been mentioned in other presen-tations at this conference, the report of the American Council of LearnedSocieties Commission on Cyberinfrastructure in the Humanities and SocialSciences is in near-final form and makes several key recommendations forleveraging digital technologies across the humanities and social sciences.

I will come back to the infrastructure topic. In the area of intellectualproperty, an extensive study by the Berkman Center at Harvard Universitywas released earlier this summer. The report addresses how copyright lawaffects the ways in which media produced primarily for commercial use, suchas movies, music, and other related materials, can and cannot be used foreducational and scholarly purposes, perhaps to the detriment of the ed-ucational mission. Another study that is nearing completion, at ColumbiaUniversity’s Kernochan Center for Law, Media, and the Arts, focuses on thelegal constructs needed to protect digital archives from demands to changeor withdraw material from online view, in ways analogous to how the publicis protected from attempts to recall material distributed in print.

Panel 3 291

I also will come back to the topic of intellectual property. First, however,I want to focus on a series of four studies, all of which suggest that the placefor universities and libraries to look for guidance in making choices aboutresources and services is, not surprisingly, in the disciplines. Although theneeds are uneven, with some fields bursting with energy and creativity whileothers operate within relatively static paradigms, the pressures to innovateand advance knowledge are greatest within the disciplines.

At the Berkeley Center for Studies in Higher Education, Diane Harleyand former University of California provost Judd King recently concluded astudy of promotion and tenure decision making. Their results show that, ingeneral, the recognition of innovative forms of scholarly publication occursslowly but that variation is greatest at the discipline level and that emergingforms of digital scholarship and digital publications are recognized whenthey make genuine contributions to fields of study. The Mellon Foundationis now considering a follow-on study that will explore the hypothesis thatinnovation tends to occur first in informal modes of communication and onlygradually shifts to more formal modes of publication.

A recently completed Mellon Foundation–funded study at the Universityof Minnesota libraries highlights the broad demand among faculty for helpin organizing personal information stores and a general interest in interdis-ciplinary activity. However, it notes that, to make sense of these needs andinterests, one really has to understand the context at a discipline level.

Focused on their particular discipline and related fields, two leading arthistorians, Mariet Westermann of New York University and Hilary Ballon ofColumbia University, recently completed a thorough and articulate study thatshows the increasing impact of digital imaging in the field of architecturalhistory and of digital modeling and reconstruction on scholarly research andthe need to adjust the publication regime so that research can be moreeffectively reported and disseminated.

In the field of classics, the final report of Project Vivarium, led byGeorgetown University provost Jim O’Donnell, is now being drafted and willmake focused recommendations on how existing bibliographic and textualresources need to be upgraded. New resources, such as a more comprehen-sive corpus of Latin texts and interoperable databases in such subfields asepigraphy, are needed to fuel future advances in this discipline.

Of course, each research library cannot respond to the specific needsof all disciplines. As in the past, future distinctions among research librariesalmost certainly will track the strategic choices each makes about how deeplyto support certain disciplines over others; how agile they are in recognizing,shaping, and responding to changing needs in these disciplines and acrossfields; and how effectively they cooperate with other libraries in a broaderdivision of labor. In making these strategic choices, research libraries alsohave an obligation to provide a more general level of support, and this bringsme back to the issue of infrastructure, or cyberinfrastructure.


In recent months, I have been talking with a number of librarians abouta particular funding request that highlights some key infrastructure issuesthat we all need to be more adept at handling. We have been approachedby an emeritus professor who is an expert in 18th-century British periodicals.He inherited the beginnings of an analytical index of these periodicals thatwas done in a very traditional way, with hand entries on index cards. He hasslowly added entries to the index and now maintains them in a computerizeddatabase. He has requested funding from the Mellon Foundation in order tocontinue the indexing process. However, there is a commercial interest indigitizing these periodicals, and it is conceivable that the entire universe,which is estimated to number about 1,000, could easily be digitized.

Still, less than a quarter of them currently are available online. Becauseof the irregular characters and typefaces, the optical character recognition(OCR) is highly imperfect. Overall, the best thinking about this case is ob-vious: Mellon Foundation money and institutional money would be betterinvested in digitizing and improving the OCR. The argument is why shouldwe help this one scholar generate an index that reflects his particular per-spective, when digitizing would allow multiple scholars to create their ownindices from multiple perspectives.

The counterargument, however, is that we are in danger of losing theperspective of this one expert, who has devoted his life to the study of thesematerials. Why give that up on the vague promise that, sometime in thefuture, we will be able to accommodate methods for a variety of scholars toindex and annotate this material? The let’s-just-digitize answer simply is notsufficient or satisfying. How do we make it possible for personal taxonomiesand annotations to be aligned against text and other materials for scholarlypurposes? This is a larger question about the infrastructure that we need forthe future.

Google, Yahoo, Microsoft, and others provide powerful mechanisms forscholars to deal with online materials. As scholars become more sophisticatedin their use of these technologies, their needs will be correspondingly morespecialized and discipline-specific in ways that are likely to be unprofitableto address by commercial companies—or at least some of them—aimed at amass market.

The sheer volume of digitized material requires implementation of muchmore sophisticated indexing, searching, and filtering techniques, includingthe broad application of computational linguistic and machine-learningtechniques and sophisticated techniques for filtering based on markupand thesauri, which would relate results to discipline-based concepts andconcerns. Searching and information retrieval is a growth industry not onlyin the general economy but also in scholarly communications. Solutions thatthe large search engines cannot supply will have to come from applicationsdeveloped within and for the academy. Finding those solutions should

Panel 3 293

be a high priority for the academy and its libraries and for publishers toaddress.

Related to the development of search-engine infrastructure is the infras-tructure needed to support the advance of new discipline-based researchmethods. The development of search technologies will drive the scholarlyuse of large quantities of digitized resources, but scholarly use also willshape and guide the development of particular technologies and applica-tions for specific disciplinary pursuits. Disciplines will need to develop newand specialized methodologies and informatics of standards and practices toidentify, mark up, and explore the large volumes of digital information withwhich they each need to work: economists with tabular data and governmentpublications; literature scholars with literary texts from various genres; socialhistorians, such as the 18th-century British periodical expert, with contem-porary accounts of various aspects of social life; ethicists with case studiesof ethical dilemmas; art historians with evidence about the context of artistsand their art, and so on.

As scholars in various fields of study develop experience with thesematerials, the disciplines and subdisciplines will need to develop and codifypractice. There is a huge opportunity for research libraries to assist in thisprocess. A key piece of the infrastructure that they can provide is the humaninfrastructure of discipline-based specialists. There are ample examples inthe studies done at Berkeley and Minnesota and in art history and classics,mentioned earlier, that indicate how these changes are taking shape. Thereare many other examples in archeology, medieval studies, musicology, his-tory, and literary studies, to focus on just the humanistic fields of study.

My last point is to return to intellectual property. The principle of open-ness is crucial in the formation of public policy for scholarship, but theadvocacy of openness for its own sake is not necessarily sufficient. I wantto give you a few examples of how our thinking about intellectual propertypolicies needs to be deepened and sharpened.

First, let me draw your attention to complex intellectual property issuesassociated with the arrangements between libraries and commercial entities,such as Google, Microsoft, ProQuest, and others. Peter Kaufman, workingwith Ithaka, made a useful attempt last year to analyze the large varietyof these types of relationships, some of which involve secret deals that donot always articulate coherent and collective educational and public interestobjectives. Additional work is especially needed on the intellectual propertyissues associated with emerging commercial services that will likely makeuse of open-access materials.

Sophisticated publishers are increasingly seeing that the availabilityof material in open-access form gives them an important new businessopportunity—that is, they can begin to incorporate and recombine materialsthat they and other publishers have produced with data and other relatedmaterials in sophisticated databases; subject them to sophisticated searches,


data mining, and semantic algorithms; and then present these as services to avariety of specialized audiences willing to pay for the added value over andabove the original content. In the end, these may be desirable outcomes andcertainly represent opportunities for useful partnerships among scholars, li-braries, and publishers. However, what is worrisome about many argumentsin favor of open access is the lack of strategic thinking about how open-access material actually will be used once it is made available and about thefaith-based assumptions that only beneficial consequences will follow fromthe provision of open access.

One worry is that open access to traditionally published monographsand serials will cannibalize sales and push smaller publishers into furtherdecline, making it difficult for them to invest in ways to help scholars se-lect, edit, market, evaluate, and sustain the new products of scholarshiprepresented by digital resources and databases. A bigger worry, which ishardly recognized and much less discussed in open-access circles, is that thelarge, heavily capitalized publishing firms will exploit open-access reposi-tories, cherry-picking the most valuable open-access products, combiningthem with the most valuable new databases and resources, and selling ser-vices back to the academy at a significant profit, while chasing out sourcesof capital from within the academic community that are desperately neededin order to advance study in the sciences, humanities, and social sciences.

One concrete step that the organizations represented in this assemblycan take is to begin to engage the open-access advocates in this criticalstrategic and policy debate about the full life cycle of scholarly communi-cations, not just the trendy, glitzy rhetoric about the initial step of makingmaterials freely available. The following questions have to be asked: openaccess to what and for whom and by what mechanism? How can we ensurethat there is sufficient capital for investment in the dissemination of new andemerging forms of scholarly output?

In the software arena, a variety of alternatives has been explored andarticulated in the form of open-source licenses, some of which facilitatedesirable downstream activities and others of which do not. For content,options like the Creative Commons license are important to consider, but,because there is so little experience in this area of use, it is doubtful thatsuch a license represents a sufficient answer.

I am tempted to close by calling for us to lock arms in a spirit ofcooperation and good feeling as we march forward into this uncertain butpromising future. Instead, I suggest a more complex approach. There arecertain areas, such as the development of key elements of infrastructure,in which cooperation is absolutely necessary in order to achieve scale andother benefits. We need some careful discussions about what those elementsmight be. However, some good, old-fashioned, aggressive entrepreneurshipand some healthy doses of risk-taking competition among research librariesare also sorely needed to help advance scholarship in various disciplines.

Panel 3 295

This entrepreneurship and competition is necessary in order to ensure thediversity in our libraries and universities that Jim Duderstadt observed is oneof the strengths of our system of higher education.

LIZABETH WILSON

Early in the 20th century, Henry Suzzallo was the president of UW, a fledglinginstitution way out in a rainy wilderness called Seattle. President Suzzallo’svision was to build a “university of a thousand years.” He knew that all greatuniversities had great libraries. So, his first action was to create a library torival those in Europe. He called it a “cathedral of books.”

Up from the empty land rose a grand Gothic structure with the OlympicMountains and the Pacific Ocean in the distance. Suzzallo’s university of athousand years had its cathedral. Since then, the Suzzallo Library has becomeknown as the soul of the university and as a beloved symbol for Huskiesaround the world.

I should let you know that, ultimately, Suzzallo’s cathedral of bookswould get him fired for having aspirations that the governor of Washing-ton viewed as foolish and extravagant, but President Suzzallo knew whatthe 20th-century library should be—a magnificent building of inspirationalarchitecture, filled with the finest books from around the world. It was sosimple then. Suzzallo had a clear vision. He did not have to look through aglass darkly.

Fast-forward to today, and one thing remains the same. The futureof the university is inseparable from the future of the library. Or, as JimDuderstadt said, the library of the future indeed may predict the future of theuniversity. The networked environment and the accelerated pace of changehave transformed our libraries and higher education. The rise of easy-to-usesearch engines providing access to a vast array of content has changed all ourdaily lives. Wonderful opportunities now exist to create digital content fromthe library’s own stacks and make widely available those hidden treasures.Libraries have reshaped their spaces into flexible learning spaces to meeta variety of user needs—collaborative and individual study, high-tech andhigh-touch instruction, and caffeine and chatter. If President Duderstadt isright, Starbucks may have had more impact on reshaping libraries than hasthe Web. Howard Schultz, our hometown boy in Seattle, will be glad to hearthat.

Throughout this period of transformation, libraries and librarians havebeen agents of change. They’ve been innovative and creative and at leasthave tried to bring technology into the service of learning and research. Inmany ways, we have put the 20th-century library of Henry Suzzallo out ofbusiness. At the same time, we can’t fully articulate the shape of the 21st-century library. However, we all know that it will not be a cathedral of books.


Our future, I believe, will be determined in large part by how we collec-tively respond to the networked world and “anytime, anyplace” expectationsand realities. Our previous speakers reminded us that education and researchin this century will demand a complex, integrated, and increasingly globalinformation infrastructure. Universities will be measured on how well theymanage and disseminate knowledge. Universities will need to find new waysto share intellectual effort in order to advance discovery and educate studentsfor a future we cannot even begin to imagine. Yet, during this transforma-tion, the mission of the library has remained relatively constant: to meetthe information needs of its community through the gathering, organization,preservation, creation, and dissemination of knowledge.

We all know that the tactics and the strategies have changed. Likemy fellow library directors, I grapple with the shape and the form of theemerging library every day. I ask what do and what will our faculty andstudents value? How can we support the expanding university vision that isincreasingly focused on solving big, transdisciplinary, global problems? Whatare the possibilities? What are the costs? What will we trade off?

Where should we invest, when we have limited resources, conflictingpriorities, and competing and often contentious clientele? When it comesto making predictions about the future, my mother always warned me thatthose who make crystal ball predictions often end up eating glass. So, atthe risk of getting some shards in my teeth, I’m going to look through theglass darkly or, to paraphrase Karen Hunter from an especially clairvoyant1992 article on electronic publishing, through the kaleidoscope darkly. I’mgoing to speculate a little bit about the research library of the 21st centuryand pose some questions that build on others that have been raised over thepast 2 days. So, what do I see through my glass or my kaleidoscope?

Well, I see a preferred future. In this century, I see a future in whichscholars, faculty, students, and researchers will be able to access and use theinformation that they require when and where they want it and in whateverform most appropriate to their needs. Better yet, make that a future in whichlibraries anticipate those needs, which then are woven into the fabric of thesearch for knowledge. I hope—I anticipate—that information will remainavailable for generations to come, whether it’s a 19th-century book, a 20th-century Maria Callas recording, or a 21st-century political Web site. I envisiona future in which our physical and virtual libraries are trusted and robust andfacilitate collaboration. I envision a transformed scholarly communicationsystem that is both accessible and affordable. I envision digital libraries thathave reached their potential to improve research and facilitate learning. AndI envision a future in which our students will be as information fluent asthey are reading and writing literate and technology competent.

We’re trying to work on this future at UW. In fact, the UW Library soonwill launch Vision 2010, our new strategic plan based on a lot of strategicthinking. This plan is steeped in the local, with a reach to the global. In

Panel 3 297

it, we commit to the vision of being an international leader in imagining,creating, and realizing the promise of the 21st-century library. Granted, it iseasy to spew vision, but it’s much harder to turn vision into reality. As KevinGuthrie said earlier, it’s always harder to change something that exists thanto create something new. So, how do we get there? Where will we at UWinvest?

I can tell you that we will continue to invest in assessing the landscape,listening to our users, tracking patterns, and looking for places where we canmake a difference in connecting people with knowledge. As an example,we recently focused efforts on understanding the information needs of ourburgeoning number of bioscientists in figuring out how to better serve themand better work with them. I thought that I’d take a few minutes to sharewhat we learned about this particular group.

Well, no surprise. We learned that everybody wants more electronicaccess. Clifford Lynch is right. For those folks, the library is seen primarilyas an e-journal provider with a big checkbook. We learned that bioscienceresearchers who work at the molecular or smaller level don’t use books.Bioscience researchers who work at the systems or ecology level use books.Most faculty researchers don’t come to the physical library. In fact, theyequate fewer visits to the library with higher productivity. We’re glad aboutthat. That was one of our strategic directions—the anytime, anyplace library.Most graduate students and undergraduates, however, come to the physicallibrary. Article databases are greatly underused and declining.

Researchers are generating vast amounts of data and are having diffi-culty managing it, and they are expecting the library to step in. There isa great need for personal information management. In fact, many of thebioscientists thought that we were coming to talk to them about scanningtheir file cabinets. With grant support, most buy the books that they needfrom Amazon. The transaction cost from discovery to delivery is too highin time and attention. There is a need to integrate fragmented systems andprocesses. Researchers are suffering—truly suffering—from an overwhelm-ing amount of information, the demands of immediacy, and the managementof expectations.

Bioscience researchers are multidisciplinary and multi-institutional col-laborators. They work with people at the university, across the nation, andaround the globe. One typical researcher named at least five different coun-tries and more than 10 institutions with which he collaborates every day. Thisis Clifford’s virtual community. They are everywhere, in scattered locations.The department is simply a placeholder where they pick up their check.Bioscience researchers are independent and self-sufficient scholars who relyon external funding for their existence. They are free agents.

These findings have huge implications for our strategies at UW libraries,and this is just one segment of our diverse community. We also will investin people—new kinds of people with new kinds of diversified skills. For


example, we brought in a bioinformatic scientist, thanks to a Howard Hughesgrant, to build a biocommons. Ten years ago, only librarians and librarytechnicians worked in our library. Now, we have professionals in computing,fund-raising, grants administration, publications, communications, graphicsdesign, human resources, organizational development, assessment, diversity,financial investment, and usability.

Most of all, we will invest in collaboration and collective action. Our vi-sion of the 21st-century library is only possible through collaboration—deep,true collaboration. Collaboration, I believe, will be the defining character-istics of the library of this century. We can no longer feel complacent orcomfortable about the artificial boundaries between libraries. Libraries havea long tradition of cooperation. We have operated for at least a century ina circle of gifts, but libraries will be even more interdependent and inter-twined than ever before, not just with each other but also with stakeholders,information providers, information creators, and users.

The strategic repositioning of the ARL; the wise investments of the Mel-lon Foundation, the Institute of Museum and Library Services, and others; ourpartners in the publishing world; our commercial alliances; and the comingtogether of the OCLC and the Research Libraries Group with organizationslike the DLF and others and with our colleagues in information schools allhold great promise for building a robust social and technical platform tofuel this interdependence. Somebody has suggested that we might even begetting close to having some planetary alignment. As Lorcan Dempsey sug-gested, we must move what we can to the network level. We must do at thelocal level only what can’t be done collectively or what doesn’t make senseto be done collectively.

Wring out the unnecessary redundancy. What we’re really talking aboutis an ecosystem. If I remember high school biology—and I didn’t pay re-ally close attention—when there is a weakness in the ecosystem, the entiresystem suffers, shifts, and ultimately adapts. My colleague from across LakeWashington, Randy Hinrichs of Microsoft, and I have been musing about theusefulness of thinking about all this as an ecosystem. We both think that itmight help us ask questions that we seem to routinely avoid—questions suchas what are the business models for reintermediation, making the ecosystemwhole, or making a new ecosystem. Donald Waters has asked us to considerthe full life cycle of open access. What about transactional systems? Howabout derivatives? How do we overcome the constraints of competition? Wealso should be asking, what role should and can multinational corporationsand foundations play in this ecosystem?

Alice Prochaska read our minds. We need to be asking questions aboutappropriate outsourcing, sharing, and other services and about the varietyof learning and research communities, accreditation interests, faculty repu-tations, student experience, and lifelong learning. Finally—and close to myheart—what are the new and critical roles for libraries? If we can bring all

Panel 3 299

the pieces together and the stakeholders of the ecosystem and ask the rightquestions and come to some meaningful consensus, we think that we canbegin to collectively drive forward tools, policies, standards, and research ona global scale. To echo Kevin Guthrie’s admonition, we need to move fromad hoc decisions to purposeful decisions, and then maybe we can better de-fine a pathway for the 21st-century library and university. If you’re interestedin joining us in this conversation, let us know. We’ll buy the drinks.

I still have many questions about the 21st-century library. Judging fromwhat I’ve heard over the past couple of days, I am confident that it willbe virtual and real, flexible and networked, global and local, clear, depend-able and comprehensible, multidimensional and integrated, and part of anecosystem sustained through collective action and new modes of trustfuloperation. Whether you are a domain expert, an authored faculty memberwho’s interested in sharing and promulgating your ideas, a publisher dissem-inating research, an educator concerned about your students’ informationalliteracy, a technologist designing information systems, or a funder choosingwhere to make investments, each of us has a very important role to play inrealizing the library and, thus, the university of the 21st century.

For those of you who are interested, I brought a copy of UW’s newVision 2010, not because it really speaks to your institutions but so that youcould see the framework of the Suzzallo Library from 1920. We chose to usethe photo of this framework, Suzzallo’s cathedral of books, to symbolize thework that’s ahead for us in reframing or rebuilding the 21st-century library.

So, I’ve done my crystal ball gazing and no doubt have a few shards ofglass in my teeth. The possibilities are enormous. The uncertainties are evenmore so. I’d like to thank UT for providing this amazing, great service bybringing us together to begin imagining the 21st-century library. I hope thatin the concluding session, which will be moderated by Duane Webster, wewill find ways to move to action this conversation started here in Austin.

QUESTIONS AND ANSWERS FOLLOWING PANEL 3

Commercial Publishing and the Academic Sector

From the audience, one participant asked Karen Hunter the proportion ofjournal revenue derived from academic libraries.

KAREN HUNTER: In the physical, life, and social sciences, I wouldestimate about 90%. It’s very much academic, government research, andprivate institutions, but it’s library based. In the health sciences, it’s muchsmaller. I don’t know the exact number, but it’s probably less than 50%,because there are individual subscribers. You also have advertising revenue.You have corporate pharmaceutical companies. You have a whole differentset of revenue streams. That’s one of the reasons that subscription prices forclinical medical journals are significantly lower.


There are so many other sources of revenue. Overall, you may be happyto know that journals actually are less than 50% of our revenue these days.We’ve finally gotten that number pushed down. We’re trying. We’re tryingnot to lean on you for growth.

Developing Services for Different Disciplines or Markets

One library director in the audience raised the question of the size and attrac-tiveness of different academic markets and their varying capacity to attractinvestment. When the market is big enough, she observed, a commercialenterprise may feel comfortable coming into it to design and deliver a set ofproducts and to build services around it. When the market is smaller, morespecialized, or more focused, it can be harder to define the audience and,hence, to attract investment. She asked the panelists to speak about howlibraries should think about the investments that they make in providing anddeveloping services for different markets.

DONALD WATERS: I’ll take a small stab at that. Part of the focus that Iwas raising about disciplines is that markets are pretty diffuse. The activity isall over the map. There are opportunities within the disciplines. The studyof classics—this study in art history, for example—has helped focus theenergies within that discipline on what’s needed next.

The more that we can do that kind of strategic thinking—such as whenthere are workshops held in the sciences to focus activity, so that an um-brella theme is developed and everybody working in a field connects to thattheme—helps focus activity. My hypothesis is that similar kinds of activitiesin the humanities would have similar effects.

LIZABETH WILSON: Your question made me think about strategiesthat we have been using and can use even more now in a networkedenvironment, where we have one scholar at our institution who’s workingon something, one over here, and another one over there. How can we buildenvironments or distributed collections or collections on the fly to supportthose three scholars, which we weren’t able to do as easily in the past?

I think of the fine work that’s been going on with the global networkof ARL and the Center for Research Libraries. Those are, I think, really goodexamples of sharing the community or of making the community largerand then supporting it through these different connections. It was Clifford,I think, who suggested that maybe the way of building alliances amonglibraries that used to be geographic is through disciplinary alliances that maycome and go and change according to what the community served is workingon.

The question concluded with a request from one of the prior presentersthat the panelists revisit the kinds of cyberinfrastructure required of thevarious disciplines.

Panel 3 301

DONALD WATERS: I’ll refer back to John Unsworth’s presentation as apretty interesting and clear example of the kinds of infrastructure needed ina particular field. The ability to combine text for certain kinds of data-miningactivity is extremely important. That kind of activity has to be undertakenin a collaborative way. It’s highly inefficient for John to try to collect thosetexts. In order to demonstrate what’s needed, he’s mashed them togetherand done the reformatting and so on. We need to raise that to a level ofstandard activities, so that the bringing together of texts for various kinds ofanalysis is much easier to do. We can go field by field, I think, with thosekinds of analysis.

That’s really what I meant—that the collaborations aren’t necessarilygoing to be universal and common. There need to be strategic partnershipsamong those institutions that have an interest to create that common infras-tructure, so that they then can continue with their business. That’s reallywhat I was calling for—not that we don’t know where those infrastructuresare needed. I think that we still need to know more about some areas. But,there’s a lot that we do know, and we need to act on it.

KAREN HUNTER: I’d like to add one footnote to that—not exactly re-spond to the same question. You reminded me of something else that JohnUnsworth said. It’s just an example of what I was trying to say. One thingthat we’ve heard would be helpful for all publishers to do is to attach meta-data to downloaded PDFs, so that people could really have data that wouldmanage those downloads. That’s the kind of thing that we can try to pushand push within the industry and get done. It’s not the kind of thing that wecan charge anything for. It becomes another service, and that’s fine, becauseit makes the information more useful and usable. But, it’s a dilemma thatpublishers are facing. There are a lot of things that we know we can andshould do, but we probably can’t recoup anything on them.

REFERENCE

Hunter, Karen. (1992). Through a kaleidoscope darkly. Science & Technology Li-braries, 12(4), 91–98.

Documents

Panel 3: Into the Glass Darkly: Future Directions in the 21st Century