7

COOPERATIVE COLLECTION DEVELOPMENT

The Manhattan Research Library Initiative, Electronic Books, and the Scholarly Monograph at Risk

Angela Carreño and William Maltarich

THE MARLI COLLABORATIVE COLLECTION VISION

Columbia University, the New York Public Library (NYPL), and New York University (NYU) established the Manhattan Research Library Initiative (MaRLI) in March 2011. MaRLI aims to reassess collection development activities and take full account of the opportunities and risks presented by expanding electronic collections.1 NYU Dean of Libraries Carol Mandel summarized the rationale, writing, “There is so much content that our scholars need. With MaRLI, our combined collecting power will enable us to create collections more wisely and make more content available to more people. Our shared collection will be a research resource greater than the sum of its parts.”2

The MaRLI Collection Development Committee, consisting of collection development officers from Columbia, NYPL, and NYU, has met since the beginnings of the consortium.3 Its goal has been to discover spaces where the consortium can cultivate transformative methods of collaborative and distributed collection development within the landscape of shifting publishing models, proliferating publishing formats, and enduring institutional priorities. MaRLI was spurred on by concerns about shrinking library acquisitions budgets, duplicative purchasing among the libraries, and the cost of onsite and offsite print storage. In addition, MaRLI seeks to increase the visibility of low-use materials and to support their continued publication and distribution, and strives to balance each institution’s specific goals and strategic priorities with consortial aims and a desire to decrease overlapping efforts.

Given these trends, the scholarly monograph stood out as a crucial focus. It is central to each library’s strategy; it faces uncertainty, crisis, and opportunity as publishers shift toward producing more electronic publications; and some of the scholarship most important to the libraries seemed to be at risk given trends in serials pricing, library budgets, and the economic climate during and after the recession of 2008. The MaRLI libraries determined that their programs should seek to take advantage of the availability of electronic monographs, address preservation and long-term access to research, and aim to increase value by expanding the resources available to each institution through collaboration rather than focus on cost savings. MaRLI concentrated on developing a shared collection approach for three categories of monograph because of the challenges they presented and threats they face: the university press monograph, the foreign language scholarly monograph, and the open access monograph. These categories are similar in that the economics of publishing low-use scholarly monographs (and therefore its sustainability) are hardly clear, especially as many libraries shift toward collection development decisions based on usage and just-in-time collecting. Neither the majority of scholarly monographs published by university presses nor scholarly publishing in languages other than English is expected to be heavily used, and open access monographs continue to struggle to find a sustainable model.4

SCHOLARLY MONOGRAPHS FROM UNIVERSITY PRESSES

Around the time MaRLI was forming, fiscal crises drove libraries to purchase fewer scholarly monographs, particularly from university presses. This change, compounded by the impact of patron-driven acquisitions models of ebook purchasing, which emphasize access over ownership, did not escape the notice of MaRLI. Elisabeth Jones and Paul Courant saw this as well, and their article “Monographic Purchasing Trends in Academic Libraries: Did the ‘Serials Crisis’ Really Destroy the University Press?” quantifies this trend.5 The threat this decline in purchasing posed to the presses themselves and, therefore, to the scholarly communications ecosystem, made a focus on university press monographs a natural first step for collective acquisition of content in print and electronic formats.

Recent publications like the Association of American Universities (AAU) and the Association of Research Libraries (ARL) joint Prospectus for an Institutionally Funded First-Book Subvention demonstrate that as time has passed this crisis has become more acute. The needs addressed by the prospectus arise from the conflicting functions of publication for tenure and publication for sales, and the report bluntly points to “the inability of a market model to adequately support research monograph publication based primarily on scholarly merit.”6

University presses focus on publishing books without a clear market (i.e., specialized, niche, and first books). MaRLI recognizes the specialized monograph as an important format for scholarly communication in the humanities and humanistic social sciences and made a concerted effort to support the output of the university presses. This includes the sometimes esoteric research in fields that rely on book-length thought and contrasting analyses of multiple sources. This support, it was hoped, might contribute to the functioning of the university presses and the sustainability of the specialized monograph. To this end, MaRLI libraries did not seek to spend less on University Press books, but to spend more wisely, more consistently, and as partners in tandem. In addition, the consortium wanted to transition aggressively to the electronic version of the scholarly monograph but without the risk of giving up print. Therefore, MaRLI sought a partner who could supply a large range of electronic books from presses whose output libraries had collected heavily via print approval plans, whose platform was familiar and functionally acceptable, and who could include print books in existing arrangements. MaRLI found a partner in Oxford University Press, because of its fortuitous expansion into University Press Scholarship Online (UPSO), at the time of these initial discussions.

THE UPSO COLLABORATION

At the Oxford University Press advisory board meeting in the fall of 2011, the NYU MaRLI representative presented “Collaborative Collection Development of Ebooks (MaRLI),” a proposal for an ideal pilot project. That proposal asked for a package price to include access to the electronic version at all three MaRLI institutions and a single print copy of each book. Because the goal was to maintain traditional expenditures on the books included, NYPL agreed to serve as print repository. Oxford requires NYPL to continue buying print based on past purchasing patterns. Although this meant that in some subject areas—the hard sciences and business—MaRLI would lack an automatic print copy, the project’s and presses’ focus on the humanities made this a minimal concern. Given the belief in the importance of print for the scholarly monograph in the humanities and social sciences areas covered heavily by Oxford and its partner presses, MaRLI requested a discount on the price of additional print copies of books in the program. Because NYPL and Columbia share an offsite repository, and other MaRLI arrangements allowed for circulation among MaRLI libraries, this single copy of the print would be considered consortial. A consortial approach to technical services was considered, but was not implemented because each library’s local needs were incompatible. MaRLI did ask for flexibility on Oxford’s part in respect to invoicing and the receipt of MARC records.7 In November 2011, Oxford responded favorably to MaRLI’s request. The MaRLI/UPSO arrangement is now in its fourth year. Over those years the model has remained essentially the same but there have been some modifications, some the result of technical restrictions, some due to business concerns.

First, although the project was designed to focus on ebook frontlist titles, backlist content became a part of the pilot almost immediately. This backlist content, which for Oxford’s university press partners was not available independent of their frontlist offerings, was of differing interest to each of the MaRLI partners. NYU was interested in delivering content to its overseas campuses in Shanghai and Abu Dhabi because of its commitment to providing its entire collection to NYU researchers, sometimes by shipping print abroad but more often by seeking out electronic journal backfiles and book backlists (even backlists already owned in print). Columbia and NYPL were less interested in backlist electronic content. Nonetheless, backlist pricing that reflected a lesser need for this content and balanced consortial arrangements for backlist cost-sharing made the backlist purchase possible across MaRLI.

Second, although this model has made it possible for each of the MaRLI libraries to scale back on print purchasing, none of the libraries has, in fact, made meaningful changes to their print approval process. This is, of course, the case at NYPL because its print ordering serves as the mechanism for obtaining a preservation print copy. NYU and Columbia have continued purchasing varying amounts of UPSO print because of their belief that researchers use print books in ways that differ from their use of electronic books, and that those uses are most relevant for areas heavily represented in UPSO. Another, perhaps more important, reason for continuing print purchases has to do with the output of these presses in print as it relates to their electronically available output. It is nearly impossible to know in a timely way whether a new UPSO press’s print book will be available on the UPSO platform. MaRLI academic library participants were not prepared to forgo the approval process, because it provides timely access to titles from these publishers and ensures acquisitions of a copy of all the desired print books. Therefore, print purchasing from these presses continues unabated.

Similarly, it has proven difficult to manage the discount for second copies of print books ordered by MaRLI libraries. In order to have those print books flow through acquisitions and cataloging in a standard way, these titles are purchased through YBP, the library’s book vendor. Managing the print discount involves a three-stage process: library payment of the full value of the print, the vendor’s matching of the respective library’s print purchases with titles on the UPSO platform, and the application of a credit to library accounts.8 This process has proven unwieldy and labor intensive.

The online publishing frequency, especially of the partner presses, has made the electronic environment difficult. Ebooks in UPSO were initially released in quarterly batches. This meant that an ebook title destined for the platform could sit in limbo for up to three months before appearing online. At the same time, it is impossible to predict and nearly impossible to know in what format(s) any given title will be published. Some titles may not be available as ebooks, some may be available only through aggregators, some may be available through aggregators and on the UPSO platform, and others may be available from aggregators, on the UPSO platform, and in other university press ebook packages (e.g., Project MUSE, JSTOR, ACLS Humanities E-book).

The MaRLI consortium continues the collaborative purchasing program with Oxford University Press because the program has met its fundamental aims. The advantages are many. The discounted print price has enabled MaRLI libraries to move to more comprehensive collecting of ebooks without the risks and ruptures associated with the absence of print. The collaborative program provides a path for moving to a model that prefers collecting ebooks because print is optional and print purchasing criteria can be refined over time. Ebook purchasing has been integrated into established workflows, providing a model applicable to other ebook packages. The consortium acquires the complete electronic content available from UPSO at attractive pricing. The program addresses preservation concerns both with the consortial print copy housed at NYPL and by the involvement of OUP and several of the partner presses in Portico’s Ebook Preservation Service (www.portico.org/​digital-preservation/​services/​e-book-preservation-service). MaRLI libraries’ combined purchasing power expands the range of content acquired. MaRLI also believes that by committing to this content, it is supporting the publication of the endangered yet essential specialized scholarly monograph.

The means by which MaRLI sought to achieve these ends are now in place and are being refined continually. The next step will be to assess circulation and usage statistics to determine whether the goal of increasing the use of this content has been met. The initial success of the program with Oxford University Press has encouraged MaRLI to address other low-use content, as discussed below. In addition, MaRLI expects to expand the program to other sources of university press ebooks.

FOREIGN LANGUAGE MONOGRAPHS

MaRLI subsequently focused on academic monographs published outside the United States in languages other than English. In Provocations and Irritations for the Globalized Research Library, Dan Hazen eloquently points out many of the challenges facing libraries that try to collect these books. Among these, Hazen notes that the priorities that inform this sort of collecting, the tools used to assess the value of these collections, and the infrastructure supporting the production of this material are each in flux, and he worries about the sustainability of the scholarly system within which these materials circulate.9 Hazen points to cooperative collection development, open access, and broad collective discovery as potentially helpful courses to take, but he also sees their potential to disrupt the market and system they attempt to reform. He calls for broad digital access to foreign materials, internationalized services and perspectives, and international collaboration in order to make this type of collecting succeed. MaRLI is taking first steps in these directions. It sees this project as a practical application of cooperative collection development through its emphasis on the collective acquisition of content in print and electronic formats, the expansion of access to materials, and cooperation with partners and publishers to continue making collecting efforts economically feasible.

Like the English language university press collective acquisition program, the MaRLI collaborative collection program for foreign language publications pursues electronic access for the three MaRLI institutions plus a preservation print copy of the titles acquired. Here too, the NYPL serves as the print repository library with the print books going offsite to a facility shared with Columbia and others. Participants seek to expand ebook access while maintaining their historical expenditure levels for the materials from target publishers. If efforts with English language presses edged MaRLI toward the involvement of third-party vendors in collaborative collection building, the foreign language publications demanded it.

Without a vendor’s local expertise, relationships with publishers, and the ability to represent library needs to multiple publishers, the MaRLI libraries never could have attempted to collect with a breadth that would make this program meaningful. In order to partner with a vendor, there must be trust in its knowledge of the country-specific publishing industry and book trade in addition to demonstrated fluency with US library service needs. On top of this—and crucially—the vendor must be able to supply ebooks under the MaRLI model.10 Prior to exploring MaRLI cooperation, MaRLI libraries were already working—each in its own way—with Digitalia in Spain, Casalini Libri in Italy, and Garcia Cambeiro in Argentina.

For Spanish ebooks, NYU and Columbia had been using Digitalia as a vendor. NYPL had not used Digitalia for ebooks but, like NYU and Columbia, had been receiving print titles from many Spanish publishers through another vendor, Puvill. Because Puvill supplies shelf-ready books to NYPL, MaRLI decided to continue receiving preservation print copies at NYPL using them as vendor. On the ebook front, it was a requirement of the program that the three libraries establish a baseline ebook collection by bringing their collections into sync. Columbia and NYU each had purchased some unique content via Digitalia and each had to catch up with the other partner’s purchasing. Once this synchronization was achieved, the libraries were able to move forward with the program of acquiring three electronic copies of purchased Spanish print books. Although made through a separate print vendor, the NYPL print purchase was considered a part of the shared payment arrangement. NYU and Columbia pay a higher percentage of the electronic payment to Digitalia, and NYPL pays Puvill. The payments for the electronic and print versions of the titles are treated nonetheless as a whole.

Unfortunately, hesitancy on the part of most Spanish publishers has meant that this program must focus on older titles for perpetual purchase. In addition to perpetual purchases, however, the arrangements with Digitalia include access to all electronic books on its platform in exchange for the promise of an annual expenditure. This is, in effect, a variant of evidence-based acquisitions (EBA), where libraries commit to expend a certain amount for perpetual access to specific titles, following one year’s unlimited usage of a much larger collection. At first, the libraries made a payment at year’s end, and the prior year’s payment was applied toward perpetual access to a set of ebooks chosen by the libraries based on whatever criteria they chose. Taking usage data into account, MaRLI purchased selected ebooks in groups based on publisher, and chose publishers based on the quality and availability of the content at the end of the year. Quickly the model changed because of increased comfort among the MaRLi libraries and the presses working with Digitalia. MaRLI libraries determine at the beginning of the year which press’s titles to purchase. Print books continue to come to NYPL, as in the university press example above, as well as to Columbia and NYU, if they wish to order them. There is no discount for print versions of titles also available in electronic format. As indicated above, Spanish publishers have preferred to embargo their frontlist books (only 2 percent of the books made available in 2014 were published in 2014). Consequently, MaRLI vendor purchasing has focused more on backlist content than the university press purchasing has. MaRLI has purchased about 500 books per year in order to maintain access to the full range of books offered through Digitalia.

In fall 2014 MaRLI expanded the Digitalia program to include Catalan books offered through a separate but functionally equivalent platform. The payment arrangements, too, are identical to those outlined above for Spanish books. This collection was of particular interest to the libraries because many of the titles are sparsely held in North America (if held at all). In this sense, the Catalan effort meets the MaRLI goal of expanding coverage through cooperation and looks especially promising. NYU will collect these titles in electronic format only, while NYPL and Columbia will continue to seek print copies of some of these titles.

MaRLI’s efforts to expand access to Italian books relied on the assistance of Casalini Libri, an established book vendor, and differ from the Spanish model in important ways. Casalini Libri offers a perpetual access ebook collection for an annual fee. Of the MaRLI partners, only Columbia had purchased such books prior to this project. One necessary step in moving forward was for the other two MaRLI partners to license this database. NYU completed the purchase in 2012 but NYPL has yet to do so. Even so, MaRLI has been able to move forward with perpetual ebook purchases from additional important publisher partners willing to meet MaRLI requirements. After some searching and effort, in 2012 Casalini was able to propose the MaRLI purchase of ebooks from the Leo S. Olschki collection, which had been available only through subscription. That 2012 collection, like most of the Spanish ebooks, consisted of backlist titles: 1,011 books published in print between 2000 and 2010. The publisher, taking into account the print purchases MaRLI libraries had made in the past, offered these titles at a discount. Later MaRLI expanded its purchasing to include 398 ebooks—primarily art books—from L’Erma di Bretschneider. The hope is to continue to acquire Italian monographs from Casalini’s pre-established collections as well as from other Italian publishers according to this precedent.

In Argentina, MaRLI has turned to another familiar vendor, Garcia Cambeiro, for assistance. MaRLI priorities remain constant: keeping the overall libraries’ budget the same, decreasing print duplication, expanding coverage, and starting a concerted transition to ebooks in current acquisitions. Working with Garcia Cambeiro requires first that the libraries consolidate their print purchasing—no small decision, and one that will take time. In the meantime, NYU has moved forward on its own with an e-only approval profile.

With Garcia Cambeiro, the MaRLI process has been gradual and considered, and MaRLI has yet to meet its goals. At the same time, MaRLI’s work with this vendor illustrates an incremental approach toward consortial purchasing. MaRLI will carefully evaluate Garcia Cambeiro’s ability to meet MaRLI criteria for content quality, standards for metadata, and needs for a mature and relatively touchless workflow. The hope is that this pilot with NYU will help ease the way into a consortial project—not only aiding vendor development, but also acclimating publishers and demonstrating potential to MaRLI. MaRLI has shown its commitment to the project by changing print vendors and the single-institution approval plan serves as a proof of concept. All sides certainly hope to move forward with a consortial approach to commercial content using this program.

OPEN ACCESS SCHOLARLY MONOGRAPHS

Conversations with Garcia Cambeiro regarding scholarly monographs in Argentina quickly brought to light another shared collection development challenge that MaRLI aims to confront: the open access (OA) scholarly monograph. MaRLI recognized many publishers of Argentine scholarly monographs produced output captured in traditional approval plans but also output that escaped print and electronic approval plans for the simple reason that it is not for sale. Although the publishing and dissemination of OA monographs align with the fundamental values of librarianship and have arisen with support from our ranks, when looking at Argentine books MaRLI was ill-prepared to manage OA monographs thoroughly and according to traditional requirements. Because all three MaRLI institutions have strong interest in collecting in Latin America, Argentine books stood out as an important OA pilot.11

Although OA scholarly monographs present a unique set of challenges to academic libraries, MaRLI proposes that its experiences with OA ejournals, websites on the open web, purchased and gift print books, and leased and purchased access to ebooks should inform plans. Like OA monographs, OA journals come from a multitude of sources, vary widely in quality, and present issues of preservation, link stability, and collection development. Importantly, MaRLI has solved many of those problems by turning to centralized vendors that work in cooperation with publishers and organizations and to which MaRLI libraries pay for services. MaRLI’s ejournal experience shows that a centralized form of management and collection building of OA titles will be a necessity.

Since many OA books reside on unvetted servers in a state of tenuous availability, MaRLI looked to library efforts to collect and preserve websites as an analogous undertaking. Investigations of and policymaking for web archiving address concerns that will be a part of OA scholarly book collecting and preservation: defining the object collected, file format and format standards, discovering and curating available resources, copyright and licensing, integration into library discovery systems, and developing or purchasing the technological infrastructure necessary to collect and preserve these resources. In this sense, OA scholarly ebooks behave much like the websites that have so challenged library collection development and preservation efforts.12

Processes and workflows in the world of print monographs also translate to collection of OA as well as leased and purchased access to ebooks. Over many years libraries have developed relationships with vendors and publishers and established collection patterns and policies that relate to collecting books in other formats. Because of the scale of book publishing, libraries have historically turned to book dealers to help them keep a current picture of domestic and foreign book publishing; to automatically collect well-defined subsets of that publishing output; to receive notification of books of likely interest; to centrally select, purchase, pay for, and catalog those books; and to track purchase histories to avoid unintentional duplicate purchases. MaRLI has applied modified versions of these processes to shared ebook purchases and, for purpose of efficiency and uniformity, wants to apply these processes for the collection of OA books.13

Print gift books also informed thinking about the approach to OA monographs. Though gifts can be a great boon to a library’s collections, the dribble of donated books of varying quality and content has often represented a great burden to libraries, because each gift must be vetted for quality by subject specialists; checked against the collections to determine local needs; and cataloged, barcoded, and shelved. Aditionally, libraries often lack space to accommodate large gift collections. The costs of these processes have caused many libraries to go through moratoria on accepting new gifts, and some libraries have ceased accepting unsolicited donations altogether. Libraries have learned to be discerning about accepting free materials because they are not in the end free. This lesson remains true for open access monographs.

Against this background, MaRLI aims to pilot the collection of OA scholarly monographs with a program focused on Argentina. MaRLI believes that because of the move to OA publication by some academic publishers, libraries are failing to collect this material at the levels managed in the print world while the scale of OA output has increased.14 Despite the availability of broadly scoped search interfaces and protocols for finding titles hosted on institutional repositories, university websites, or author pages, the library services required to comfortably consider these titles a part of collections are universally absent. In addition, collection development librarians have no sense of what is missing from the collections and titles expected to be discovered by search engines because there is no picture of the universe of titles available. This is, in fact, true across all OA books, but the issue is especially pointed for the titles MaRLI sought to collect.

Collection development librarians quickly realized that this goal may require library partnerships with multiple consortia, and potentially with other libraries, as well as working with book vendors knowledgeable about the local academic publishing world to develop a shared OA monograph collection. In addition, MaRLI librarians prefer a platform for delivering OA books to users and, ideally, preserving and making these titles available universally as a more stable, more diverse, and quantitatively richer collection of titles than the native platforms where OA titles initially reside.

From our understanding of the issues detailed above, MaRLI identified key areas that a partnership among consortia, libraries, and vendors needs to address: issues of awareness, selection, rights management, stability of access, discoverability, and consistency of user experience. MaRLI believes it has found a rational, economically sustainable, and relatively comprehensive process for addressing them. For Argentine titles, MaRLI asked Garcia Cambeiro to provide the picture of OA monograph output in a defined universe that is as complete as possible and transparent about gaps. MaRLI also requested a mechanism for selecting titles from that universe, and rights negotiated with publishers that allow MaRLI participants to use the books purchased according to the academic community’s needs, which at this point likely include rehosting content for both preservation and access. Additionally MaRLI requested MARC records according to specifications, inclusion of a persistent URL (persistent because of the re-hosting in the previous point), and duplication control between the electronic and print versions of titles held among partner libraries. These services come at a cost, and MaRLI expects to share that cost consortially.

The model for Argentine OA monographs involves meeting the requirements above through the involvement of a vendor, whose fees are paid by a consortium of libraries. MaRLI believes that this is the most economical and sustainable way to manage OA scholarly monographs.

The current plan for collecting these OA monographs is akin to crowdsourcing.15 Following the MaRLI model, libraries would pay for vendor-mediated OA monographs. If enough libraries participate in a collaborative program, OA monographs can be re-hosted on a stable platform.

If MaRLI is to interest HathiTrust in this collaborative OA model, MARC records produced would point to this stable platform, and preservation activity can cover these selected titles. In addition, if the hosting platform has relationships with discovery tools, including searching over the open web, these OA collaboratively curated titles should become more visible throughout the world. The participating libraries have the advantage of determining which OA monographs should be included in the program they finance, but the benefits are truly universal.

Contributing libraries allow for the processing of a number of OA monographs parallel to their financial commitment. Currently, NYU alone has committed a small amount to a pilot as a proof of concept, but the final platform for these OA monographs is currently unclear. Discussions within MaRLI, with HathiTrust and broadly across many libraries with historically strong foreign language collections, are ongoing in developing a strategy for collective curation of OA monographs.

While this chapter outlines ideas and a possible approach to collective curation of OA monographs, implementation of a practical approach is still evolving. The key is to find libraries willing to help a vendor recuperate costs for the services seen as necessary to processing OA monographs. Those libraries will need to be committed beyond fulfilling traditional collection development functions that have exclusive access to books just at their home institutions. As conversations progress, libraries willing to make this commitment are being identified. Through this process, OA titles in “the wild” can be transformed into a large, carefully curated part of an even greater collection of curated titles for which library efforts at integration pay the largest dividend. Although there will be what some consider “free riders” in this process, they remain important beneficiaries of a concerted effort well in line with the most worthy traditions of librarianship.

The challenges of collaboratively collecting OA scholarly monographs are the latest, and perhaps most difficult, MaRLI has faced. These challenges in collaborative collection building are not insurmountable and do not seem totally novel.

CONCLUSION

Each of the collaborative collection development challenges MaRLI has addressed is meant to expand access to monographs and increase the visibility and use of monographs crucial to the scholarly process if not to every, or even very many, scholars. The university press monograph, the foreign language monograph, and the rising category of the OA monograph published outside traditional publishing (but not scholarly!) workflows each represent new formats of material that MaRLI libraries have aggressively collected for years. MaRLI’s collaborative collection development efforts are attempting to ensure continued collection of these scholarly monographs at risk because of rapid technological and structural change. One part of this process is developing new mechanisms to find, acquire, process, and pay for this content. The other part is a conscious attempt to support the viability of publishing this important material.

Notes

1. That the consortium includes a public library may seem odd to those unfamiliar with the NYPL’s history as both a traditional public library and a major research library serving both unaffiliated scholars and, because of the breadth and depth of its collections, scholars in general. See Phyllis Dain, “‘A Coral Island’: A Century of Collection Development in the Research Libraries of the New York Public Library,” Biblion 3, no. 2 (1995): 5–75.

2. Three Renowned Research Libraries Join Forces to Better Serve Their Users, press release, March 18, 2011, www.nyu.edu/​about/​news-publications/​news/​2011/​03/​18/​three-renowned-research-libraries-join-forces-to-better-serve-users.html.

3. The committee members are Angela Carreño, Head of Collection Development at New York University; Jeffrey Carroll, Director of Collection Development at Columbia University Libraries; and Denise Hibay, New York Public Library, Susan and Douglas Dillon Head of Collection Development. These three serve as the representatives of the full range of collection development decision-makers throughout their libraries.

4. Even the popular press has started to comment upon the threat to the university press; see, for example, Scott Sherman, “Under Pressure,” Nation 298, no. 21 (2014): 19–24. For a glimpse into the sparse holdings of foreign language monographs among ARL libraries, see Mary E. Jackson, Lynn Silipigni Connaway, Edward T. O’Neill, and Eudora Loh, Changing Global Book Collection Patterns in ARL Libraries, CRL Research Network Report, March 2007, www.crl.edu/​sites/​default/​files/​attachments/​pages/​grn_global_book.pdf. Given the overall trends in monographic purchasing, holdings can only be sparser at present; see, for example, Alex L. Holzman, “From the University Presses—Open Access Monographs and the Scholarly Communication Ecosystem,” Against the Grain 24, no. 6 (2012): 56–57.

5. Elisabeth A. Jones and Paul N. Courant, “Monographic Purchasing Trends in Academic Libraries: Did the ‘Serials Crisis’ Really Destroy the University Press?,” Journal of Scholarly Publishing 46, no. 1 (2014): 43–70. Although they do not attribute the decline in monographs purchasing to DDA models and usage-driven decision-making in this article, they did so in an earlier presentation; see www.slideee.com/​slide/​monographic-purchasing-trends-in-research-libraries-did-electronic-journals-really-destroy-the-university-press.

6. Association of American Universities and Association of Research Libraries, Prospectus for an Institutionally Funded First-Book Subvention, June 12, 2014, www.arl.org/​storage/​documents/​publications/​aau-arl-prospectus-for-institutionally-funded-first-book-subvention-june2014.pdf.

7. Shortly after the MaRLI proposal was accepted, the Triangle Research Libraries consortium instituted a very similar program that incorporated a consortial approach to cataloging and payment. See Charles Pennell, Natalie Sommerville, and Derek A. Rodriguez, “Shared Resources, Shared Records,” Library Resources and Technical Services 57, no. 4 (2013): 227–238, and “OUP, MaRLI, TRLN Reach Agreement on E-Book Use,” Advanced Technology Libraries 41, no. 11 (2012): 9–10.

8. A prospect made difficult, like so many processes, by the impossible proliferation of “unique” identifiers in the (E)book ecosystem.

9. Dan Hazen, “Provocations and Irritations for the Globalized Research Library,” paper presented at The Global Dimensions of Scholarship and Research Libraries: A Forum on the Future, Duke University, December 5–7, 2012, www.crl.edu/​sites/​default/​files/​attachments/​events/​Duke%20Conference%20Hazen%20paper.pdf).

10. Sometimes forward-thinking meant simply that the vendor was willing to sell the electronic monograph and had the technical wherewithal to deliver the title, and had established trust with publishers so that they could even broach the subject of the ebook.

11. Two of the three members of the MaRLI Collection Development Committee have backgrounds as subject librarians in Latin American Studies, and the third representative works closely with the Director of Columbia University’s Area Studies/Global Resources Division, who has a background in Latin American Studies.

12. OA monographs challenge us less than websites, however, in that they are not (for the most part) edited or updated on an ongoing basis, they are discrete units, and they aim for lasting availability. In this sense, libraries should carefully examine the processes, procedures, and tools in place to deal with websites. On the other hand, because the category of books is much more narrow than websites and is, in addition, a category (if not a format) that libraries have long collected, OA monographs present unique challenges. Although library users have set ideas about where to find library books, user expectations about the idea of archived websites remains a novelty.

13. It is crucial to note that MaRLI paid service fees to vendors for their assistance in these areas, and expects to pay for those services even when the content itself is free. It is also crucial to note that in part libraries have had an implicit cost-sharing arrangement for these services with publishers, which have offered print vendors discounted prices on their books out of which those vendors have sustained themselves. When the books we ask those vendors to process have no discount (a discount on free is of course zero), libraries should expect fees to those vendors to be higher.

14. Juan Pablo Alperin, Dominique Babini, Gustavo Fischman, eds., Open Access Indicators and Scholarly Communications in Latin America (Buenos Aires: Unesco, 2014), closely examines the prevalence and relevance of open access publishing for Latin American journals and, more relevant here, books.

15. It is somewhat similar to the unglue.it model. See https://​unglue.it/​faq/ for background.