Christopher Baker
I hate those eBooks. They cannot be the future. They may well be.
—Maurice Sendak
One thing is clear about the role of eBooks in public libraries. That is, not much is clear. There are major players, applicable best practices, booming statistics, growing support resources, and increasingly savvy users, yes. But the core model for library service: buy book, lend book remains fraught with new and changing ifs, ands or buts when applied to eBook collections. Not only on the front lines of collection development, but for library support staff, newly acquainted users, and those not-yet-cardholders public libraries are always reaching for. Public libraries and librarians have approached these challenges head on. We have faced and continue to face the reality that eBooks, while a major boon to the business of public libraries, have spurned some of our professions most controversial, insightful, and innovative decisions and discussions and they are definitely not going anywhere.
eBooks are steadily converting consumer markets as librarians struggle to provide content and advocate for access. While librarians have long championed eBooks and eReading, the growth in eReading has remained largely in the consumer market, outside of library walls. Library users purchase devices and quickly find content to fill those devices. In most cases they end up buying eBooks from major retailers instead of borrowing them from a library. A recent Pew study found that a majority of library cardholders are not even aware that eBook lending is a service provided by their local library (Zickuhr et al. 2012a). This is a startling revelation for librarians who have worked to build, market, support, and advocate for eBook collections. Users are finding eBooks, however. The Association of American Publishers BookStats 2012 report found that in 2011, eBooks ranked highest in format selection for adult fiction (Sporkin 2012). And in early 2011, Amazon reported that Kindle book sales had marginally outpaced paperbacks but were outselling hardcovers by three times (Carnoy 2011). Coupled with the rapid adoption of tablets, smartphones, and eReaders, with both tablet and eReader ownership in American adults nearly doubling, from 10 to 19 percent, between December 2011 and January 2012 (Rainie et al. 2012), it is clear that users are well beyond the tipping point for eBooks and eReading. Public libraries are as well, but have had a very different experience thus far.
Like individual users, public libraries have invested time and energy to build and grow eBook collections, learn the ins and outs of associated devices and technology, and enjoy the benefits of a collection available on demand—anytime, anywhere. But public libraries have been yoked with conditions and limits not placed on individual users: paranoid publishers hindering access to in-demand and popular titles readily accessible in physical formats, a complex, multifaceted, and ever-evolving download process, and the very real limits of collection development budgets in the face of the current eBook pricing and lending models offered to public libraries.
That is not to say public library eBook collections are not popular despite these issues. In the final week of December 2010, OverDrive-powered digital library sites were literally crushed by the demand for eBooks from users eager to download to devices received as gifts over the holiday season (Stasiewski 2010). eBook usage at public libraries has grown rapidly since, quadrupling between 2010 and 2011 alone with projections of a 67 percent increase in circulation between 2011 and 2012 (“eBook Usage” 2012). As evident by these numbers, eBooks in public libraries are definitely moving. However, librarians who are responsible for delivering eBooks cannot help but imagine what could be with only a few changes. These changes might include the following:
• access to other major publishers’ eBook catalogs;
• a less complicated download experience; and
• a more library-friendly model for eBook acquisition and lending.
eBooks are the public library’s greatest catch-22. They are what is next, they are in demand, they place no burden on physical space, and they are available when buildings are closed. But, they can also be difficult to purchase, are difficult for users to access, and require no small amount of ongoing support from library staff.
While some public library systems were early adopters of eBooks, the majority have only recently began collecting and investing heavily in digital collections. However, the addition of eBooks has been rapid within the past two years. According to the 2012 Library Journal and School Library Journal’s eBook Usage in U.S. Public Libraries report: “Almost nine in ten public libraries currently offer eBooks to their users, and 35% of those that don’t are in the process of adding them” (“eBook Usage” 2012, 4). That leaves very few library systems that have yet to jump into the eBook game. Public libraries that have not added eBooks cite the lack of funds to support a collection, reservations over available eBook platforms, or a user base without means to support such a collection as the major reasons they have not added eBooks (“eBook Usage” 2012, 21). While these are reasonable concerns considering the current state of eBooks in libraries, public libraries that are reluctant to add eBooks may soon find themselves with a similarly reluctant user base who believes the library is not responsive to their information needs or requests. As a generation of users raised on screens and on-demand media inches closer and closer toward becoming supportive library users, public libraries must recognize that digital content, such as eBooks, are key in addressing the information needs of today’s and, more importantly, tomorrow’s user.
It is evident that users are personally interested and invested in eBooks and eReading and public libraries have made heavy investments to address those interests. It is also evident that while usage and demand for library eBooks continues to grow rapidly, current trends and limits in eBook collection development have brought libraries through a crossroads and onto an unfamiliar path. The traditional model for library lending is not concrete and straightforward for eBooks. This is, unfortunately, predicated on the fact that the traditional idea of ownership is not applicable in current eBook purchasing models. Because, unlike physical books, libraries are unable to own eBooks, instead, libraries license access to eBook content. Libraries are in eBook lending purgatory, incapable of fully reaping the benefits of electronic access and delivery while experiencing growing demand and user interest in digital content. Major publishers have had differing reactions to the idea of eBook lending in libraries, from complete avoidance to limited circulations and increased pricing to justify their sales to libraries. Although libraries have never controlled book publishing or publishers, hard work and legislation have given libraries rights to purchase and lend physical books for as long as items can circulate. Limits currently in place to control eBooks in public libraries go much farther than merely damage done to physical books. While physical copies can be replaced or repaired, the complete refusal of sale by major publishers threatens the public libraries’ ability to even provide access to popular and relevant titles and information. For users, the proliferation of different eBook formats and restrictions placed on eBooks limits their ability to connect with the information they want or need using the device they have in hand.
Public libraries are in a precarious position in regard to eBook collections. Because libraries must be responsive to customer demand for access to popular and current titles, the leverage in the eBook debate is heavily skewed toward publishers. This lack of leverage has mitigated the voice of the public library as an influence on library lending of eBooks. But librarians have not given up. The role libraries play in eBook collection development and management is changing, from a simple access point to advocator, creator, and negotiator. This change is evident in the profession’s constant calls for increased access, in the creation of home-grown platforms, and in libraries’ work to adjust and advocate for new models and sources for library lending. Yet, libraries are still not fully in the eBook spotlight. To play a larger role, libraries must continue to engage publishers, vendors, and users alike to ensure each understand how important they are to the success of the eBook. For publishers, libraries create readers who then become customers. For vendors, libraries advance platforms and products through trial and use. And for users, public libraries tirelessly advocate for free, open, and easy access to information in all formats.
There is no doubting the public libraries’ important role in the rise of eReading over the past few years. However, in the face of growing demand and limited access, public libraries must take command of and work to expand that role. To draw a literary allusion, public libraries are poised as the protagonist in conflicts that will influence the future of eBook lending for users. To be clear, these conflicts are not presented in an antagonistic manner, but rather as a way of grouping the issues public libraries struggle with and must overcome to realize the full potential of eBooks and what they mean for users. These conflicts are presented with both internal and external concerns and address a range of issues surrounding eBook collection development, management, and support.
The current climate for eBook collection development is not wholly inhospitable to public libraries but it does not support an expanded role for libraries beyond passive content licensee. There are a number of vendors at the ready to assist public libraries in starting or expanding eBook collections. These vendors act as content aggregators and distributors for publishers, giving libraries access to a single marketplace and platform from which to build and manage eBook collections. With adequate funding and planning, eBook collections can be up and running in public libraries with very little effort beyond basic selection decisions and ensuring user support.
OverDrive was the first, and remains the largest, vendor to offer downloadable eBooks to public libraries. Its original vision for library lending and ongoing response to changes in formats, availability, user habits, and devices continues to set the bar for what public libraries expect in an eBook vendor. Other major eBook vendors include 3M, whose Cloud Library offers synced delivery across users’ mobile devices, Freading, a pay-per-use platform for library lending powered by PDA, and Baker & Taylor’s Axis 360 digital media library and accompanying Blio reader, which also offers cloud-based delivery of content. Public libraries should examine each vendor’s model, in depth, prior to any purchase. A number of factors should be considered carefully, including the availability of suitable publisher content, the compatibility of available formats with users’ devices, and the sustainability of costs associated with creating and maintaining successful eBook collections.
There are also deeper factors to consider before investing in an eBook collection or a new eBook platform. The content those vendors are able to provide is heavily controlled by the relationships they have with publishers and current limits major publishers have placed on selling to libraries. Those relationships are a key element in how well libraries are able to serve their reading public once the decision is made to add an eBook collection. Above all, and a focus of great concern over 2012, vendors do not and cannot control major publishers’ power in removing, adjusting, or altering the availability, prices, or formats of eBook titles in their marketplace and even in previously purchased digital collections. These factors should be carefully considered before adding new, or supplementing existing, eBook collections.
Of the big six publishers, or five with the impending Penguin and Random House merger (Pfanner and Chozick 2012), two, Simon & Schuster and Macmillan, do not make their eBooks available for library lending through any vendor platform. Public libraries are granted access to other major publishers’ title lists through third-party vendors such as those mentioned earlier, but that access has increasingly come with a caveat. In 2010, HarperCollins gave libraries access to the entire catalog of titles but altered its sales model to cap the loan of each new eBook purchased at 26 loans before a new copy could be purchased. Luckily, libraries get a discount on subsequent purchases. Hachette removed frontlist titles from library catalogs in 2011, and more recently increased the price of its backlist titles by just over 100 percent. Penguin had stopped selling eBooks to libraries in February 2012, after a tenuous few months that first saw the publisher suspend the sale of new titles to libraries, and then also request that the Amazon Kindle format’s Whispersync delivery be disabled for all Penguin titles in OverDrive-powered collections. However, Penguin has recently returned to the library market through a pilot partnership between 3M’s Cloud Library, New York Public Library, and the Brooklyn Public Library. Like HarperCollins, Random House gives libraries access to its entire catalog of eBook titles, but increased the cost of its eBook titles by 300 percent in March 2012.
Publishers have justified these restrictions and changes with a variety of explanations, from the need to make prices reflect the nature of a copy that never degrades to the loss of friction due to the remote nature of the eBook checkout and download transaction. It is not necessary to pick apart the details of each explanation, they point to the same end: major publishers believe that library purchasing and lending of eBooks cannibalize sales and place eBooks in danger of being poached instead of purchased. It is beyond the public libraries’ reach to pretend to be able to change the current trends in sales restriction, loan limit, and pricing decisions from major publishers. First and foremost, libraries must serve their users. As Jamie LaRue asserts in a recent Library Journal article, when it comes to serving users, “We’re still going to buy some of the content our patrons want us to buy. And they’re still going to ask for what has been advertised. That’s the mainstream” (LaRue 2012a, 33). What public libraries can do is change the way they approach the eBook market, beginning with a look at how collection development can enhance and expand digital collections to better serve users while also pushing public libraries’ role in the eBook market forward.
“Technological innovations changed how music and songs are bought and consumed today” (Sen 2010). This simple statement, from an article exploring changes in music in the digital age, may seem like common knowledge to anyone with a current relationship to music, but it should also strike a chord with librarians who have any stake or interest in the future of eBooks in public libraries. eBook collection development dollars are still being used to support outdated and inapplicable models that do not address the technological innovations eBooks offer. Major publishers’ current actions toward controlling the eBook market mirror the actions of major music labels faced with the growth of downloadable music and file-sharing. Just like major music labels, major publishers have scrambled for solutions and continued to adjust toward what they feel are workable models. While they adjust internally, eBook channels outside their control have grown. To keep up with eBook demand and usage in consumer markets, eBook availability from independent publishers has skyrocketed and self-published authors have embraced eBooks to great success. Libraries can play an important role in this sea change. The creators and distributors, devices, and user attitudes surrounding eBook content have shifted in the same way they did for music. This comparison may not have proved as strong to public libraries that were, only a year ago, celebrating higher eBook circulation statistics, greater compatibility through the addition of the Kindle format from OverDrive, and comparatively lower-cost eBooks from a wider selection of major publishers’ catalogs. However, factors currently controlling and influencing eBook publication, distribution, and use have lined up in the same way they did for the music industry. To stake a claim in the eBook paradigm of tomorrow, public libraries need to heed these flashing signposts.
As noted in an ALA TechSource post by Patrick Hogan, “traditional publishing and distribution channels now control only about 12 percent of new content” (Hogan 2012). Collection development in public libraries can, and should, evolve to embrace new channels for eBooks. The independent and individual streams of publishing are growing exponentially and public libraries have a rare opportunity to play a major role in the promotion and distribution of this content. Public libraries define themselves as content aggregators, in physical terms and as a profession, and should waste no more time in claiming this role in the eBook market. By doing so, public libraries open new doors for access, engage tomorrow’s authors today, and cement their place in actively representing the interests and needs of content creators and users alike. The good news is that a few public libraries are already following this path and are well on the way toward reasserting the power of libraries as content aggregators for all types of information.
Diversifying collection development with eBook content from independent publishers could prove the first step to move public libraries in reclaiming that role. As noted by Douglas County Library Director Jamie LaRue, “this stream of publishing is now roughly equal to commercial output yet significantly underrepresented in [public library] collections” and independents “produce a lot of fine, far more diverse writing of often keen local interest” (LaRue 2012a, 33). Independent publisher have embraced the innovations that eBooks offer in expanding their distribution. Additionally, independent publishers have reaped the benefits of an influx of established writers as major publishers shed the traditional midlist to focus on top-selling authors and blockbusters (Deahl 2010). Above all, independent publishers are excited to work with libraries and recognize the inherent value in the library’s creation of readers and customers. While public libraries will always advocate for increased access to eBooks from major publishers, eBooks from independent publishers address the library’s immediate need for eBook content and help all libraries meet the ultimate goal of collection development: connecting users with information.
There is also opportunity to expand on that goal. Self-published authors are a subset of the eBook market that has seen exponential growth in the past few years without registering a blip on collection development radars. Granted, self-published holds a certain connotation in collection development. Often bringing to mind works of questionable quality, in whichever way one defines quality, be it grammar, style, subject, or other. These wholly independent authors are taking full advantage of distribution channels offered up by eBooks and are doing so with great success. According to data recently compiled by Bowker, the release of self-published eBooks grew by 129 percent between 2010 and 2011 (“Self-Publishing” 2012). By ignoring this trend, collection development librarians are ignoring a great amount of the digital content being written and released into the eBook content stream. While statistics surrounding self-published eBooks in consumer markets cannot assuage quality concerns or influence changes in collection development policy without a much larger discussion, they can go far in proving to libraries the value users find in creating and consuming self-published works. Retailers such as Amazon, Apple, and Barnes & Noble have already tapped into this stream of content by creating publication and distribution platforms to engage independent authors. Public libraries have an opportunity to play publisher and distributor as well by utilizing open platforms that encourage and engage self-published authors alongside other digital content.
In early 2011, Douglas County Libraries and Red Rocks Community College Library partnered with the Colorado Independent Publishers Association (CIPA) to deliver eBooks through a library-managed platform that promotes accessibility and findability through library catalog integration. It also allows users to purchase titles directly from the publishers (Kelley 2011). Douglas County has since expanded the number of publishers it works with to include other independent publishers and publishing groups, and content from publishers such as Gale, Marshall Cavendish, ABDO, and Lerner Digital, who are already well known to public libraries (Sendze 2012). Another benefit Douglas County realized from working with independent publishers is the library’s ability to own the eBooks its purchases. This is no small detail when compared to current library sales models available from major publishers. By creating its own platform, Douglas County has made itself the key component in a mutually beneficial relationship connecting users with publishers. Sendze points out that by establishing agreements with publishers to purchase eBooks and manage the digital rights of those eBooks, public libraries, like Douglas County, can “emphasize our role as protectors of intellectual freedom and major players in the book-buying industry” (Sendze 2012).
Readers can learn more about Douglas County’s innovative approach to collection development by reading the chapter that James (Jamie) LaRue has contributed to this book (see “Self-Publishing: Does It Belong in the Collection?”). Public libraries should follow Douglas County’s lead and support platforms for eBook lending that expand available channels of collection development and enhance user access to eBooks. By working with independent publishers, public libraries can claim the role of content aggregator in a market with a growing number of middlemen. By reaching out to self-published authors and managing the publication and distribution of user-produced content, public libraries expand that role and place themselves on the crest of a wave of change currently hitting the eBook market.
Another important aspect of the call for public libraries to expand eBook collection development channels, and move toward in-house platforms for eBook management and lending, is the need for a standardized eBook format or standardized form of eReading access that can serve library users seamlessly across devices. For retail consumers, the device compatibility of eBook formats is not often of concern as retailers have created a seamless process for delivering and syncing eBooks across devices via proprietary readers, Web apps, and mobile apps. Consumers can purchase an eBook from Amazon, send it to their Kindle, access it from their Web browser, and resume reading later on their Android smartphone. The same is true of the Barnes & Noble’s Nook experience, among others. The retail model for eBooks removes the customer from format selection and, in doing so, rids eBooks of all physical connotations, respecting their true nature and pushing eBooks as on-demand content rather than in-demand, checked out, on hold, items.
Unfortunately, the direct comparison of eBooks to physical books continues to weigh down public libraries and eBook vendor models. “One copy, one user” remains a predominant theme in eBook sales models. HarperCollins’ cap of 26 loans per eBook purchased was a clear and early example of publisher attempts to apply physical limits to digital content. To be fair, HarperCollins’ model may be the best currently available to public libraries precisely because it follows the eBook to book comparison to its natural end: giving libraries the power to forgo replacing under-utilized titles while also offering a discount on subsequent purchasing of expired titles. But as Jamie LaRue points out in a blog posting, when it comes to major publisher models, “ ‘less bad’ isn’t ‘good’ ” (LaRue 2012b). The comparison to physical items served as a familiar starting point for both libraries enamored with the idea of eBooks and publishers accustomed to print. It has since been used to justify publishers’ restrictions and limits while steadily increasing the burden placed on public libraries in explaining the one copy, one user model to users steeped in an on-demand, immediately accessible media marketplace. The most frustrating and ironic aspect of this comparison is major publishers’ refusal to recognize the issues that accompany eBooks purchased on this basis as the very real friction they so often say is needed for digital content. Public libraries accepted this model early on to gain access to innovative and, seemingly, ubiquitous channels of information but are now bearing the brunt of decisions made without a full understanding of what eBooks could and would eventually mean to users.
Public libraries continue to call for increased access to major publisher catalogs and, given the current state of eBooks in public libraries, librarians should make their collective voices heard. However, it may be time for a strategic change in the tone of that call. New models of access should be proposed to major publishers in an effort to bridge growing gaps and return popular and in-demand content to library eBook collections. Public libraries deliver information and content to users in a number of ways and formats. Never have two such dissimilar types, book and eBook, been so strongly linked. Regardless of the name, eBooks are not books. They are born digitally, they are managed digitally, and they are delivered digitally. Yes, eBooks can be purchased, they can be checked out, and they can be returned like physical books. But these characteristics have been placed upon them. At their core, eBooks are similar to another type of content readily found in library collections—eContent accessed through databases and other online resources.
Like eBooks, public libraries purchase access to eContent through vendors and associated platforms. Like eBooks, eContent is accessible beyond the physical and temporal confines of library branches. Unlike eBooks, providers of eContent have fully realized the digital collection’s power in delivering information to users on demand. They have also realized the value of that access to public libraries. Subscription models for databases and other online resources allow public libraries to purchase access to the eContent they need for the specific period of time it is needed. Similar models of purchasing could work very well for the popular and best-selling titles offered by major publishers that do not often have the staying power suggested by the investments needed to meet initial demand. Subscription models for eBook collections could first work to shift the conversation with major publishers away from the idea of library ownership and toward one of, more simply, access: access to the eBook content public libraries need, when they need it.
Subscription models could help to alleviate some of the hesitancy felt by major publishers in selling copies that never degrade while also addressing the very real demand for popular and best-selling content felt by public libraries. Penguin is exploring a new sales model in its pilot program with New York Public Library and Brooklyn Public Library through 3M’s Cloud Library that, in part, seems inspired by subscription models for other eContent. Unfortunately, it does not go far enough. Penguin’s relaunch of eBooks in public libraries allows libraries to purchase a one-year license for each eBook copy purchased, with a six-month purchase embargo for new titles (Hoffelder 2012). Like HarperCollins, Penguin’s new model does allow libraries to forgo re-purchasing expired eBooks, helping to weed out-of-date or under-utilized titles and copies. Yet, it also continues to push the idea of the eBook as book. While public libraries welcome the return of access to a major publisher’s titles, it is important to consider the long-term effects of continued acceptance of models that do not further the usability of eBooks for library users. Bringing major publishers back into the discussion is a move in the right direction, but public libraries must assert that, like other eContent, the investment they make in eBook collections should pay dividends in access and convenience for their users.
The eBook to book comparison issue reaches beyond libraries and publishers. Most major eBook vendors also base their distribution models and platforms on the idea that in public libraries an eBook is simply another item bound by the lending limits of physical items. However, eBook vendors are making platform advancements in direct response to issues of compatibility and availability realized through library use and trends found in the consumer eBook market. OverDrive is the only vendor currently offering Kindle compatibility and is gearing up to offer a new method for delivering EPUB eBooks in library collections. OverDrive Read, similar in nature to the Kindle Cloud Reader and marketed with the tagline “See Book, Read Book,” sees OverDrive capitalizing on advancements in HTML5 to deliver eBooks to any device with a Web browser, including tablets, smartphones, and other Web-enabled mobile devices. Since there is no download, there is no need for authentication beyond a user’s library credentials. Mirroring eBook retail strategy, 3M’s Cloud Library offers synced content on up to six devices through apps available for both Android and iOS devices. Baker & Taylor’s Blio application is marketed as “a full-featured eReader for people who care about what books look like” (“Blio” n.d.), and is available for download to computers, tablets, smartphones, and other mobile devices. These enhanced features are clear indications that eBook vendors closely monitor user interactions with eBook content. Public libraries can be certain that major eBook vendors will continue to be responsive within the models and platforms that they control because they have a direct and vested interest in the success of eBooks in public libraries. As discussed earlier, what they cannot control and what continues to affect users is the publishers’ control over the availability and compatibility of eBook titles.
For libraries and library users, the availability of eBook formats varies across major vendor platforms and remains key in vendor selection. Alongside new vendor efforts in delivering content like those described earlier, vendors also support the download of eBook files direct to device or to desktop for transfer to device. OverDrive offers, when available, EPUB, PDF, and within the last year, the Kindle format for eBook titles purchased, Baker & Taylor relies on an adopted XPS file format to deliver eBooks to its Blio mobile apps, 3M’s Cloud Library uses the EPUB format for both cloud delivery and desktop transfer of its titles, and Freading offers EPUB and PDF formats for downloads, transfers, and sideloads to devices. What do all these formats and delivery options mean for public library users? They mean steps, as in step-by-step. They mean that while vendors are working toward solutions that attempt to replicate the user experience of retail eBooks, public libraries are tasked with supporting and defending convoluted processes for access to content born with the capability to be delivered seamlessly and immediately.
To that end, user support is crucial for any library implementing a new or additional eBook platform. The formats, accompanying restrictions and limitations, and eBook delivery methods (i.e., the download experience) that come along with each vendor platform can influence a user’s ability to interact with an eBook collection on all levels, from the user’s initial selection of a title to the user’s selection of device on which to read that title. In a 2001 episode of the BBC radio series the “Hitchhikers Guide to the Future,” the late Douglas Adams asserted, in regard to eBooks, that “lovers of print are simply confusing the plate for the food” (Adams 2001). This forward-thinking remark is fitting for today’s users’ view of public library collections—content is content regardless of format. With that in mind, libraries should not only be working toward increased access to food for their users but also working to ensure access to and compatibility with the different plates that users consume that food on.
Users with dedicated eReading devices, tablets, or smartphones are immersed in a media culture of instant gratification. See app, buy app. Need book, buy book. Want music, buy music. They are not faced with limits on the number of copies available for reading, viewing, or listening. Users bring this experience with them when they approach eBooks in the library. Often, frustration with either the availability of eBooks or the process by which an eBook is obtained stops a user before he or she gets started.
The checkout and download process in public libraries has come a long way in the past few years, and vendors have worked hard to increase the compatibility of library eBooks across a growing number of consumer devices. OverDrive’s addition of Kindle compatibility in 2012 is a perfect example of this type of work. The addition of the Kindle format allowed libraries to reach an entirely new audience of readers, those already using the No. 1 eReading device in the United States. Unfortunately, that increase in compatibility also came with a cost: Penguin’s removal of eBooks from OverDrive. The Kindle checkout and download process via OverDrive is one of the easiest forms of eBook access currently available to public library users. OverDrive’s employ of Amazon’s Whispersync technology has made library eBooks, once checked out, instantly accessible on a plethora of devices. While other publishers welcomed an influx of new readers, Penguin responded by citing security concerns as its motive for removing Kindle compatibility for Penguin titles. Public libraries should take note. While publishers have a justifiable interest in protecting their content, they also have an interest in controlling the compatibility and ease of access of that content in an effort to protect sales in consumer markets. Faced with this reality, public libraries can only do so much to match the compatibility needs of users. However, the steps they can take are powerful in ensuring they meet the device needs of current users and expand device compatibility for future users.
Despite the growth in device compatibility from eBook vendors, thanks in great part to proprietary apps for smartphones and tablets, there is still a gap between the devices many eBook vendors support and the popular devices users select. Of the major vendors selling eBooks to public libraries, only OverDrive currently offers eBook compatibility with the Amazon Kindle—the eReading device chosen by 62 percent of eReader owners (Rainie et al. 2012). OverDrive’s partnership with Amazon was a boon to public libraries and helped bring library eBooks to a new segment of users. Similar compatibility has yet to come from any other major eBook vendor. Public libraries have responded to issues of availability and compatibility by adding other vendor platforms to enhance and grow eBook collections. While this may serve as a stopgap measure to deliver either more eBook content or compatibility for more eReading devices, it does so at great fiscal expense to those libraries and fails to address compatibility issues head on. Investing in multiple eBook platforms forces libraries to hedge bets on content over user experience, sending users in different directions for eBook content that should easily come to them. Public libraries need to recognize that throwing money at eBook issues has only served to complicate matters thus far. By adding multiple eBook platforms, libraries waste valuable time and money chasing unraveling purchasing models, assorted delivery methods, and unpredictable eBook content down unfamiliar paths.
Such an investment of time and money should instead be made toward creating the best possible eBook experience for current users with an eye on how library eBooks may serve future users. The compatibility of public library eBook collections should, above all, mirror the device demographics of current users. Investments in eBook platforms should never influence a user’s selection of eReading device. Instead, public libraries should focus current efforts on building eBook collections on platforms capable of delivering the most content to the largest number of users. Compatibility is key. Regardless of eBook delivery method, be it via app, download and transfer, or online delivery, any investment in a vendor platform that excludes an entire segment of users, based on their choice of eReading device, is a failure of collection development to reflect the needs of the community it represents. To that end, public libraries must consider how the selection of an eBook platform will affect their ability to evolve alongside changes in the eReading habits and device choices of their users.
The devices used for eReading continue to change as users adopt and adapt to new technology. A recent study of young Americans’ reading habits by the Pew Internet & American Life Project found that users under the age of 30, who read eBooks, read them on a computer or cell phone rather than on a dedicated eBook reader or tablet (Zickuhr et al. 2012b). This shift in where eReading takes place could prove important to public libraries as they look for an eBook solution that addresses both compatibility issues for users and accessibility issues involving both major publishers and vendors. As eReading shifts toward Web-enabled devices including smartphones, tablets, and computers, library eBooks should shift in that direction as well. The immediate, online delivery of eBooks via HTML5 already takes place in retail markets and is set to debut from at least one major library eBook vendor. The delivery of eBooks via Web browser on a user’s device could go far in simplifying the eBook experience for users, removing cumbersome steps found in current methods as well as the need for proprietary apps for access and secondary accounts for authentication. Beyond a less-complicated user experience, browser-based eReading could prove revolutionary for libraries in their call for access to major publisher content simply because it treats eBooks like content, not as files that need to be downloaded and delivered through proprietary apps or third-party relationships but as eContent that can grow and change as users, technology, and devices change. The support and advancement of browser-based reading could rekindle the current conversation between libraries and major publishers by championing access above all else. For publishers, browser-based reading could mean access to new models of eBook sales based solely on the demand for access to that content without the dangers they find in current library lending models. For libraries, browser-based reading could mean new models for access to the eBook content they need, when they need it. And for users, browser-based reading could mean simple and straightforward access to library eBooks.
With all of the challenges eBooks still present for public libraries, they may often sound like more trouble than they are worth. But any public library launching an eBook collection will quickly realize the benefits of an eBook collection. First and foremost, the allure of eBooks draws in new library users, providing library services to a larger portion of a library’s community. eBooks allow libraries to deliver information and popular content to users outside of regular hours and far beyond the walls of the library branch. They also give libraries the opportunity to engage users on mobile devices, extending the presence of the public library to the smartphones and tablets users have in hand. Internally, eBooks raise the technical skill set of library staff through user-support, device troubleshooting, and the need to stay abreast of changes in eBook services. And without a doubt, eBooks put an edge on traditional library services, showing users that public libraries are dedicated to delivering convenient access to information in new and innovative ways.
Public libraries need to only look at the current use of and growing demand on eBook collections to realize they have quickly become an important part of library collections. eBooks are not going anywhere and for good reason. Regardless of current challenges, there is no denying that eBooks have a bright future in public libraries. It is up to public libraries to determine just how much of an influence they want to have on that future. Above all, public libraries must take steps to ensure eBook accessibility remains at the forefront of all discussions. A collective voice is needed for public libraries expressing the importance of access above all else. Boiling the conversation down to this simple concept gives public libraries a cornerstone on which to build a new future for library eBooks. Access will always mean the ability to deliver needed and requested content to library users, but in the context of eBooks, access is a malleable concept depending on its application. Looking forward, access could mean owning eBooks and developing in-house platforms for discovery and delivery, but by the same token, access could also mean subscription models from major publishers that allow simultaneous use of their catalog for a set period of time. As Christopher Harris points out, regardless of the comparison to physical books, “libraries don’t purchase digital content, we license it” (Harris 2011) and a license grants access. Public libraries must change their approach in eBook collection development, shrug off old models and comparisons, and show flexibility in what access can mean for libraries and library users.
To show true flexibility, librarians must also be adaptable in their approach and response to changes in the eBook marketplace. The backlash seen in 2011 over HarperCollins’ limit of 26 loans seems a glaring overreaction when compared to the response to recent, and more extreme, changes from major publishers: the total removal of catalogs or price increases that place a much heavier burden on strained library budgets. These changes point to the same end: the honeymoon period for eBooks and public libraries is over. The real work has begun. To thrive, this relationship will require what all great relationships require, communication and compromise.
Understanding is key in both directions. Libraries, publishers, and vendors will continue to experience growing pains as the creation, distribution, and utilization of eBook content evolves. But what has remained important thus far, and will remain important, is the public library’s commitment to its users. By championing user rights, public libraries work to create a clearer future for library eBooks. A future that sees ease of access and ease of use for library users grow through mutually beneficial relationships between libraries, publishers, and vendors. A future built upon one clear, and simple, concept: access.
Adams, Douglas. 2001. “E-book versus Paper,” an episode of the Hitchhikers Guide to the Future BBC radio series. BBC Radio (April 21).
“Blio—Don’t Just Read Books. Experience Them.” n.d. Available at: http://www.meetblio.com/. Accessed October 25, 2012.
Carnoy, David. 2011. “Amazon: Kindle Books Outselling Paperbacks.” Available at: http://news.cnet.com/8301-17938_105-20029839-1.html. Accessed July 23, 2012.
Deahl, Rachel. 2010. “Smaller Presses, Bigger Authors.” Publishers Weekly 257 (44): 4–6.
“eBook Usage in U.S. Public Libraries: 2012, Library Journal and School Library Journal.” 2012. Available at: http://www.thedigitalshift.com/research/ebook-usage-reports/public/. Accessed May 24, 2013.
Harris, Christopher. 2011. “Rethinking HarperCollins.” School Library Journal 57 (4): 15.
Hoffelder, Nate. 2012. “Penguin eBooks Return to Libraries—But Probably Never the Kindle—The Digital Reader.” The Digital Reader (June). Available at: http://www.the-digital-reader.com/2012/06/21/penguin-ebooks-return-libraries-but-probably-never-kindle/. Accessed July 23, 2012.
Hogan, Patrick. 2012. “Douglas County Libraries’ DIY E-Book Hosting.” ALA TechSource (March). Available at: http://www.alatechsource.org/blog/2012/03/douglas-county-libraries-diy-e-book-hosting.html. Accessed September 9, 2012.
Kelley, Michael. 2011. “Colorado Publishers and Libraries Collaborate on Ebook Lending Model.” Library Journal (March 17). Available at: http://www.libraryjournal.com/lj/home/889765-264/colorado_publishers_and_
libraries_collaborate.html.csp. Accessed September 9, 2012.
LaRue, Jamie. 2012a. “All Hat, No Cattle.” Library Journal 137 (13): 32–33.
LaRue, Jamie. 2012b. “50 Shades of Red: Losing Our Shirts to Ebooks.” Blog posting (July 30). Available at: http://americanlibrariesmagazine.org/e-content/50-shades-red-losing-our-shirts-ebooks. Accessed May 24, 2013.
Pfanner, Eric, and Amy Chozick. 2012, October 29. “Random House and Penguin Merger Creates Global Giant Random House and Penguin Merger Creates Global Giant.” The New York Times. Available at: http://www.nytimes.com/2012/10/30/business/global/random-house-and-penguin-to-be-combined.html?_r=1&. Accessed October 31, 2012.
Rainie, Lee, Kathryn Zickuhr, Kristin Purcell, Mary Madden, and Joanna Brenner. 2012. “The Rise of e-Reading.” Pew Internet & American Life Project. Available at: http://libraries.pewinternet.org/2012/04/04/the-rise-of-e-reading/. Accessed July 23, 2012.
“Self-Publishing Sees Triple-Digit Growth since 2007.” 2012, October 24. Publishers Weekly. Available at: http://www.publishersweekly.com/pw/by-topic/industry-news/publisher-news/article/54482-self-publishing-sees-triple-digit-growth-since-2007.html. Accessed October 31, 2012.
Sen, Abhijit. 2010. “Music in the Digital Age: Musicians and Fans Around the World ‘Come Together’ on the Net.” Global Media Journal: American Edition 9 (16): 1.
Sendze, Monique. 2012. “The E-book Experiment.” Public Libraries Online (January/February) Available at: http://publiclibrariesonline.org/issue/januaryfebruary-2012/. Accessed September 9, 2012.
Sporkin, Andi. 2012. “Bookstats 2012 Highlights.” Available at: http://www.frontgatemedia.com/news.php?mod=template&id=1066. Accessed May 28, 2013.
Stasiewski, Dan. 2010, December 28. “UPDATE: Significant activity on ‘Virtual Branch’ Websites.” OverDrive Blogs. Available at: http://overdriveblogs.com/library/2010/12/28/significant-activity-on-virtual-branch-websites/. Accessed July 25, 2012.
Zickuhr, Kathryn, Lee Rainie, Kristin Purcell, Mary Madden, and Joanna Brenner. 2012a. “Libraries, Patrons, and e-Books. Pew Internet & American Life Project. Available at: http://libraries.pewinternet.org/2012/06/22/libraries-patrons-and-e-books/. Accessed July 23, 2012.
Zickuhr, Kathryn, Lee Rainie, Kristin Purcell, Mary Madden, and Joanna Brenner. 2012b. “Younger Americans’ Reading and Library Habits.” Pew Internet & American Life Project. Available at: http://libraries.pewinternet.org/2012/10/23/younger-americans-reading-and-library-habits/. Accessed October 31, 2012.
Libraries are undergoing extraordinary change in a digital world. Digital indexes, journals, and monographs are now routinely incorporated into library collections and have become staples of library service. Other collection formats, such as film and video, have been slower to make the transition to digital, for reasons having to do with both the complexities of the technologies involved and the peculiarities of the video marketplace. This chapter provides an overview of the current state of streamed video collection development and acquisition. It focuses on the acquisition and licensing of new, streamed video titles and the incorporation of these into library collections, rather than digital conversion of existing collections or streaming of live events.
The author’s perspectives on streaming video emerge from years of experience with media for higher education curricular support. Acquisition and use of streaming video in public and K–12 libraries will vary to some degree, but the basic concepts covered here should hold true across many types of libraries.
Streaming is a means of delivering video content to computer desktops via an Internet connection. A host (server) delivers the file to the receiving computer (client). Settings at the host’s end determine whether access is available to multiple simultaneous users or limited to a single user. Unlike video downloads, which must transfer to the viewer’s desktop, streamed video plays almost immediately after the viewer initiates play; some content must buffer before streaming begins. Streaming video also differs from video downloads in that no copy of the file is stored on the end user’s computer, so files remain relatively secure (farrelly 2012).
Because of the generally large size of moving image data files, streaming videos usually employ file compression, a programming strategy that greatly reduces the size of the file through frame sampling and other complex means of reducing image redundancy. This compression can negatively affect image quality when compressed videos are played full-screen or projected. Even with compression, effective use of streaming video requires a robust, high-speed Internet connection. Dial-up connections generally are too slow to adequately deliver streaming videos.
Libraries’ accelerating evolution from hard copy to digital collections is in its third decade. From indexes to full-text indexes to e-journals and eBooks, libraries have recognized the value that digital formats provide. Digital’s ease of use, 24/7 access, ability for simultaneous use, service to multiple locations, and reduction in physical space requirements present multiple benefits to both users and libraries. Libraries have embraced this transition (not without some growing pains) to the point that digital content is a norm within contemporary libraries. These same benefits apply to streaming video. Within the digital content trajectory, moving to streaming video is a logical next step for libraries.
Other motivations for moving to streaming video stem from the nature of video use within the curriculum. The use of video in education is well established. It is rare for an instructor to devote class time to reading a journal article or book chapter. But the physical nature of media before the advent of video necessitated limiting its curricular use to in-class screening. Film (primarily 16 millimeter) as a medium was too fragile and complicated to withstand multiple out-of-class or reserve uses. While video formats alleviated some of this difficulty, both videotape and DVDs deteriorate from repeated use and still present obstacles to multiple viewings within the short time frame curricular use necessitates.
Higher education increasingly occurs online. In 2011 nearly one-third of all higher education students were enrolled in at least one online course (Allen and Seaman 2011). Another study reports that in 2011 overall online enrollment grew by nearly 6 percent (Primary Research Group 2012). According to the Survey on Academic Streaming Video, 83 percent of U.S. colleges and universities offer online courses (farrelly and Hutchison 2013). In this context hard-copy video formats are unusable. And while faculty anticipate using video more, often they cannot find quality or appropriate materials. Additionally, they are unable to find the content they need in their library and are instead turning to commercial sites (Kaufman and Mohan 2009).
Streaming video collections help to address many of the issues of durability and access. An instructor can still screen the content in class if desired, but also link to the content in a course management shell for asynchronous use. Additionally, many streaming interfaces permit linking to segments or portions of videos, enabling focus on key topics within the video, either in or out of class.
Arguments for adopting streaming video are not limited to academic libraries. The advent of home video resulted in huge increases in library circulation statistics, and DVD circulation remains strong in many public libraries. In a 2011 study, however, LJ Reports noted a shift in user preferences; 17 percent of respondents described using streaming services such as Netflix as their primary source for movies (Enis 2012).
The elements involved in selecting, licensing, acquiring, and hosting streaming video are foreign to most librarians engaged in collection development, already largely unfamiliar in general with media collection development. To understand issues in video collection development—seeking balance, just-in-time versus just-in-case collecting, quality, preservation, and other factors, the key text remains Video Collection Development in Multi-Type Libraries: A Handbook (Handman 2002). This title provides a strong foundation upon which to look at the evolution of media in library collections. Schools and some public libraries collected visual media in multiple formats for years. Through the second half of the 20th century the dominant media format was 16-millimeter film. Due to its expense, fragility, and need for frequent inspection and repair following use, the collection of 16-millimeter film was often limited to large public libraries, universities, or shared collections such as public library systems and consolidated school centers. It is important to note that films acquired and shared in these collections were licensed, generally for the life of the print, rather than purchased outright. The advent of home video in the mid-1970s, the release and mass-marketing of back catalog content (primarily Hollywood entertainment), and the concomitant development of rental stores resulted in films becoming available and affordable in video format. Driven by user demand, libraries began to acquire video as readily as monographs. In many ways the move to licensing streaming video is a pendulum swing back to the model of 16-millimeter distribution, with libraries not owning, but licensing content. For a detailed history of media formats in library or media center collections, see Lori Widzinski’s article, “Step Away from the Machine: A Look at Our Collective Past” (2010).
Librarians are quite familiar with contemporary monographic acquisition models. Making use of jobbers, approval plans, best-seller lists, review sources, and discounted pricing by monograph vendors, those involved in print collection development find it to be largely routinized. This is not the case with video acquisitions, outside of mass-market entertainment titles. To understand why, we must consider the differences between traditional print publishing and video distribution. In general, monographs come from publishers responsible for producing, marketing, and distributing the content, or from jobbers who distribute titles from many publishers. Generally, jobbers do not have exclusive distribution rights to the monographs they handle. Content from a given publisher’s catalog can be acquired from multiple sources such as Ingram, YBP (Yankee Book Peddler), Ambassador Books, Baker & Taylor, and Amazon. Jobbers offer a variety of value-added services, including approval plans which match elements of available product to an individual library’s scope of collecting, cataloging, shelf-ready processing, and frequently, content discounting.
The video marketplace most visible and familiar to librarians and consumers alike is the home video market. A much smaller and more obscure segment of the video marketplace consists of independent film titles distributed by small companies that primarily focus on sales to educational, institutional, or other specialty buyers. These distributors tend to have exclusive distribution rights for their titles. Films that one distributor carries are rarely available from another. Many films are self-distributed—only available directly from the filmmaker. A few library jobbers handle media titles, but these distributors generally handle only mass-market, mainstream entertainment titles, with limited independent, documentary, and educational offerings. These sell-through sources distribute only physical formats and do not have authorization to negotiate streaming rights. There are no profile-based video approval plans. Some companies provide for standing order arrangements, but these too focus on mass-market titles such as television series releases or Academy award nominees.
In their 2009 report, the first to examine changes in the use of video in higher education, Kaufman and Mohan (2009) note: “trends noticeable today . . . are as remarkable as the shift from the scroll to the codex over 2,000 years ago.” Certainly this shift is already evident in libraries with electronic indexes, journals, and most recently, eBooks. But many libraries, already resistant to media, have been slow to adopt digital video. This reluctance is partially the result of both marketplace issues (lack of hosted content, resistance of distributors to enter the digital realm, shifting licensing and pricing models) and the practical issues associated with the technology. Libraries must find ways to adapt their collection development processes and policies to meet shifting consumer use, expectation, and demand.
Libraries have multiple options for providing streaming video. Hosting, licensing, and pricing models are interconnected and differ from distributor to distributor. The models discussed here are not mutually exclusive; distributors may offer more than one or variations of these options.
Generally, libraries build video collections on a title-by-title basis; this is the common approach for streaming video as well. All the leading video distributors offer some form of title-by-title licensing for their titles. This approach provides a library the most control over content in its collection, but is labor intensive, and replicates a monographic selection process largely replaced with approval plans and patron-driven acquisitions programs. As noted previously, approval programs for video are very rare, and in the digital realm nonexistent. Instead, some distributors offer media databases or curated collections. A media database or curated collection provides an array of related titles as a package (e.g., The BBC Shakespeare Plays, Opera in Video, Filmakers Library, Films on Demand). These collections frequently offer added value through closed captioning, transcripts, segmenting of the videos into discrete units, semantic searching interfaces, and editing tools that permit users to build custom play lists from the content. With a curated collection the library has access to titles that might not have been selected otherwise, but still meet user needs. There is, however, also the probability that many titles in the curated collection will never be accessed. Distributors that offer curated collections may also license individual titles from the collections.
A third type of collection, less widely used, is the clip collection. Generally, a clip collection consists of short-form videos, often extracted from longer videos, addressing or covering a single concept. Used primarily in education, clip collections usually are a subscription product. Key clip collections include those offered by INTELLECOM and NBCLearn. Gary Handman’s 2010 article: “License to Look: Evolving Models for Library Video Acquisition and Access” addresses all of these models in greater detail.
Despite the source of the videos or the type of collection (single title, curated collection, clips), streaming video licensing falls into four dominant models: term, in-perpetuity, subscription, and pay-per-view. Most distributors offer a combination of options, and most licensing is negotiable. It is not uncommon for a distributor to require purchase of a hard copy before licensing streaming rights, though as the market evolves some distributors have begun to license digital files without requiring purchasing a hard copy.
A term license provides access for a specific period. Term licensing is essentially the de facto standard for streaming video, although the length of the term varies widely. Length of license and pricing vary according to the terms of the license. Term licensing is perhaps the most flexible of licensing options, permitting the library to provide access for a semester, a year, or even a single use.
In-perpetuity licensing is the model that many media selectors prefer. It most closely resembles the monographic purchase model in that it does not require repeated relicensing. Many video distributors have been resistant to in-perpetuity licensing, noting that they do not have distribution rights in-perpetuity. Media selectors counter that vendors sell DVDs without requiring return of the hard copy when the vendor’s distribution rights expire; thus digital files should be treated in the same manner. As the market evolves, more distributors offer in-perpetuity licensing, now sometimes referred to as life of format licensing. This refers to the life of the codec (a program or device that encodes or decodes a data stream) employed to create and manage the digital master. This approach mirrors the life of film licensing employed years ago with 16-millimeter films. If a new digital standard emerges, replacing the file with a new format will require relicensing of the title.
Subscription licensing, essentially a form of term license, provides access to an entire collection or array of videos from a distributor, usually on an annual basis. With a subscription license the library pays a flat fee for use of the collection (or collection subset) for the licensed period. In a subscription license the distributor serves the content, and new titles are available as they are added to the collection. If the distributor loses the right to offer a title, it is pulled from the subscription and is no longer available. Licensing agreements for subscription collections should include provisions to accommodate purchasing discontinued titles before the distributor loses distribution rights. Only a few distributors offer subscription models, usually those with widely scoped, interdisciplinary, or clip collections. As with licensed abstract and indexing services, nonrenewal of the license results in complete loss of access to the collection. There are no provisions for assuring continued access to a previously subscribed to back file.
Pay-per-view is a licensing model common in the consumer mass media marketplace. Amazon Instant Video and cable television providers employ this model for feature films and other specialty content, with viewers paying varying amounts for instantaneous viewing. With most pay-per-view licenses the content remains available for a short time period. A few educational content distributors offer this model, suggesting that it mirrors textbook purchasing. Some librarians argue that pay-per-view offloads financial commitment to the end user. If the cost were to be absorbed by the library, however, pay-per-view has the potential to be both prohibitively expensive and fiscally unpredictable. Pay-per-view licensing offers no permanent benefit to the community, as no content is added to the library collection.
Swank DigitalCampus licenses feature films for curricular use, based on course enrollment, number of titles, and length of license. This model supports education but is not applicable to general purpose use of feature films. Criterion has entered the U.S. market with a similar model to support curricular use of feature films. A PDA model implementation in the mid-2000s would not scale (farrelly 2008). But as streaming video matures in the library marketplace, larger streaming collections become available, and libraries gain experience with PDA models, it is possible that such a purchase model could prove workable for video. Two emerging approaches approximate a PDA model. Alexander Street Press offers evidence-based acquisition, with funds from an upfront spend applied to the purchase of accessed titles, with the library determining how those funds are applied. Kanopy’s access model results in term licenses for titles used. Similarly, there are no viable bookshelf models for streaming video. (A bookshelf model licenses an established number of titles for a predetermined period, but allows changing out the specific titles at intervals.) As streaming video is a relatively new means of delivering library media content, other licensing models may evolve or merit exploring.
Licensing and pricing are interdependent. As with licensing there are no industry standards for streaming video pricing. Some companies require purchase of a hard copy as part of the licensing process. Others include a term license for streaming with purchase of the content on DVD. Curated and clip collection vendors may base pricing on FTE (full-time equivalent) and/or Carnegie classification. At least one distributor calculates pricing on a sliding scale determined by the number of minutes licensed and the term of the license. Other distributors have offered package deals to library consortia. With these pricing models, generally the greater the number of titles or minutes of content, the lower the cost per video. Almost all distributors offer a variety of pricing models. Negotiation is critical as distributors and librarians continue to seek approaches that work for both parties.
Regardless of the source of the digital files, a key consideration in providing streaming video is hosting. Streaming video is not simply a matter of storing a file on the Internet and providing a link. Distributors expect that access will be authenticated and thus restricted to the library’s clientele. They also expect that the content will be protected from downloading, duplication, and other forms of content theft. Users expect to be able to access the content easily—on their desktops, in the classroom, and through their mobile devices. This requires transcoded files that will work with multiple computer operating systems, browsers, and playback utilities such as Flash, Mac IOS, and so forth. These expectations, as well as technology capabilities and financial capacity, impact whether a library’s digital media will be vendor-hosted, hosted locally, or outsourced to a third-party service.
Vendor hosting is the simplest and generally most cost-effective approach for libraries, mirroring how libraries provide access to databases and e-journal content. In this mode of content delivery, streaming files are served from the distributor’s site, and this distributor assumes responsibility for assuring security and the transcoding of files. As codecs change, the vendor is responsible for upgrading the files and assuring compatibility with users’ devices. Because of the resources invested in their platforms and the relationship with the original video producer, streaming video distributors can more easily provide added content and functionality such as transcripts, semantic searching, closed-captioning, and tools to create customized playlists. To fund platform development, some streaming media providers, no matter what licensing or pricing model they employ, may charge a maintenance fee. While the number of distributors providing hosting continues to increase, not all vendors host their content, requiring libraries to find their own hosting solution.
With self-hosting the licensing institution (library, technology unit, etc.) is responsible for all the processes and support structure to serve the file, secure the file, and enable playback. While some vendors will provide a digital file for self-hosting, others require the library to digitize and transcode the file from a hard-copy DVD. In such scenarios, closed-captioning, transcripts, and segmentation of the video may be too labor-intensive or expensive to be taken on by the library. As of this writing there are few third-party companies or turn-key solutions to perform these functions, so most self-hosting solutions are home grown. A few open-source products are available, but they require knowledgeable programmers to implement fully. Among these products are Omeka, Kaltura, iTunes University, and Vimeo. As the need for hosting solutions grows, new products are certain to be developed. ShareStream, Video Furnace, Media Hub, and Avalon Media System (in development by Indiana University) are some of the emerging tools that show promise for simplified streaming video hosting.
Libraries that lack the resources for in-house hosting may turn to third-party providers to perform these functions. One of the largest third-party companies is used by some of the largest video distribution companies. When using a third-party solution, it is important that the license from the original content provider includes language to permit such hosting and the storing of files off-site. Given the expense of building and maintaining a self-hosted solution, it is likely that the need for third-party hosting will increase, and that new companies will emerge to provide this service.
“Findability precedes usability. You cannot use what you cannot find” (Greenfield 2010). Regardless of where or how a library hosts streaming content, it is of no use to users unless they are able to identify, locate, select, and connect to that content. A library’s streaming video collection most likely will be housed across multiple locations and interfaces, including distributor platforms and local servers. It is essential to establish tools to identify and access this content without requiring users to search multiple access points or interfaces. In the author’s view this is best accomplished through full MARC records in the library’s catalog with direct links to the streaming file. Yet nearly a quarter of libraries that stream do not provide title-level catalog records for their streaming videos (farrelly and Hutchison 2013). Some distributors provide some form of bibliographic data for their streaming titles, most commonly larger vendors who offer subscription services or curated collections. The quality of this data varies widely. Some provide detailed and complete MARC records, including Library of Congress subject headings. Others offer little more than general subject assignments or records, which, while adequate for K–12 or public libraries, may be insufficient for a research library. Other distributors may provide only a metadata file from which the library can build catalog records. Some librarians argue that library catalogs should not include records for subscription content that may go away if titles are removed by the vendor or the library’s subscription lapses. Yet without catalog records the content in such subscriptions is unlikely to be discovered and used. Since renewal of resources is often based on use data, mechanisms to drive discoverability and resultant use are essential. Vendors may see the provision of catalog records as an expense that does little to increase sales. Librarians need to continue to impress upon distributors the importance of bibliographic data for generating use—use that subsequently drives renewal. Libraries with the requisite cataloging skills will need to examine vendor-provided bibliographic data to determine its suitability for local needs, and arrange for other solutions if it does not.
Some libraries opt to edit the existing catalog record for VHS or DVD video already in the collection to include a link to the digital version. Others may find it satisfactory to modify an existing analog record for the digital file. Libraries planning to generate original cataloging for digital video files will want to refer to Best Practices for Cataloging Streaming Media (OLAC 2009).
Of course the catalog is not the only way to discover and access resources. Discovery tools such as ProQuest Summon, EBSCO Discovery Service, III Encore, and Ex Libris Primo, now being widely implemented in libraries, also serve to identify content and connect users. Some larger distributors already work with discovery tool vendors to assure that their content is included. Libraries also need to work with discovery tool vendors to impress on them the importance of making video content discoverable and to ensure that their own collections are activated within the tools.
As more libraries acquire streaming videos, the likelihood of usable records for copy cataloging through OCLC and other record providers will increase. As more distributors offer subscription and curated collections, the prospects of subject indexes identifying streaming content and providing access through link resolvers will also certainly increase.
Principles of video collection development are well described in multiple chapters of Gary Handman’s Video Collection Development in Multi-Type Libraries: A Handbook. Individual chapters address public, school, and academic library video collections, while one chapter directly addresses the need to develop collection development policies to address new technologies (Scholtz 2002). The principles addressed in this text remain relevant to streaming collections. Streaming video collections should continue to mesh with existing collection development policies. The scope and nature of the content collected, balance, fit with curriculum or user population, and myriad other points in a library’s collection development policy should still apply to streaming titles collected. For a librarian not already familiar with educational and documentary media publishing and distribution, the process of finding, evaluating, and selecting video content, streaming or not, may be baffling. The ALCTS (Association for Library Collections and Technical Services) Guide to Video Acquisitions in Libraries: Issues and Best Practices (Laskowski 2011) addresses the essentials, including appendixes covering review and selection tools, and an annotated list of video vendors. There are some vendors who specialize in the distribution of streaming video. The ALA’s Digital Video Collections Guide describes numerous quality streaming video collections (Spicer 2012). Although structured with an academic focus, the Guide lists collections that are also applicable to public and school libraries. The Guide is an open access resource; others are invited to contribute descriptions of other digital video collections with high-quality content.
Feature films and television programs present unique challenges for a streaming collection. Consumer interfaces such as Netflix are driving the development of streaming options for personal viewing of mass-market entertainment titles. There are fewer options for libraries, including Swank’s DigitalCampus and the Criterion platform mentioned earlier. Hoopla, a platform developed by Midwest Tape, may prove to be a viable interface for streaming mass entertainment titles for public libraries using a pay-per-view model. Some libraries and faculty have explored using Netflix or Amazon on Demand for curricular use of feature film. It is arguable, however, whether such use is permissible within the terms of use for these products. And such use generally falls within the limitations of the pay-per-view model already referenced. For feature films and television, general market forces rather than educational need drive the availability of content for institutional acquisitions; thus it is unlikely that many viable options will be available for libraries to stream content of this type in the near future.
There is no single guide, Web site, directory, or jobber for identifying and selecting videos. The National Media Market, a trade show featuring documentary and educational video, however, provides an excellent opportunity to preview and select titles from many of the top distributors. Copies of newly produced and released, as well as popular, titles are available for on-site preview, and distributors often negotiate package deals and discounts with attending selectors. Nearly all distributors participating in the National Media Market offer streaming options. Another outlet for streaming video information is the VideoLib discussion list. A general forum for all matters pertaining to video in libraries, list members frequently recommend titles, identify sources for hard-to-locate titles, and offer other advice pertaining to video collection development and management.
Librarians interested in developing a streaming video collection may wish to start by looking at use data for their existing physical video collection. Providing streaming access to heavily used titles, titles frequently used for course reserves, those needed at more than one location, and titles with broad multidisciplinary appeal are good candidates for a streaming collection. Also, as VHS and DVDs need to be replaced, streaming, if available, should be strongly considered.
If jumping into licensed streaming is not viable, openly available collections of high-quality, freely accessible videos that do not require licensing are a possible option. The Digital Video Collections Guide mentioned earlier and the author’s Libguide on Streaming Video (farrelly 2012) provide descriptions of and links to many such collections.
Early models for streaming video placed inordinate burdens on libraries. It was not uncommon for distributors to require hard-copy purchase before allowing a streaming license, to charge double the price (or more) of the hard copy for a digital file, to require renewal at the same price at the end of a licensed term, or to require the library to digitize and host the content. Had journal publishers promoted e-journals with similar requirements, it is unlikely that libraries would have adopted e-journals at all.
To some degree many of these requirements still persist in some licensing models, but the market is changing. As late as 2007 less than half a dozen distributors for educational or documentary content provided streaming video options. Now virtually every video distributor offers some form of streaming licensing. Still, licensing presents a hurdle for distributors who are relatively new to the game and for smaller, independent production companies.
Streaming access offers considerable value to libraries and their clientele. Twenty-four-hour and remote access, simultaneous use of a single file by multiple users, segmentation and user editing tools, and the ability to link or embed video into course pages or learning management systems expand the flexibility and usability of video collections for users. Libraries benefit from a huge decrease in staff time devoted to processing, circulation, and shelf maintenance for videos. Other significant benefits include eliminating the need for multiple copies and the costs and efforts related to replacement of lost, stolen, and damaged videos.
Some of the cost benefits of streaming video are countered by the conversion of staff time from collection maintenance to efforts in licensing and negotiation, tasks ordinarily performed at a higher skill level and pay grade. Significantly, term licenses that require paying repeatedly for the same content move video acquisition into the sphere of serials and other continuations of digital content that now consume an ever-increasing portion of library acquisitions budgets. As the streaming video market matures, more distributors are accommodating licenses in perpetuity, but libraries subscribing to curated collections run the risk of replicating in video the current problems with the Big Deal in e-journals packages.
There are expenses associated with streaming video that other digital content does not present to libraries. The major additional expense is hosting. For content not hosted by the distributor, the library faces the ongoing costs of local hosting. Emerging market forces, however, will minimize this ongoing cost as more distributors provide hosting, turn-key hosting solutions emerge, and third-party vendors provide alternatives to hosting locally.
Unlike other digital content providers some video distributors do not provide the digital file, off-loading to the library the responsibility of generating a digital file from a purchased DVD. These locally generated files most likely will lack closed captioning or chaptering, without additional expense incurred by the library.
There are tradeoffs with streaming versus physical copies. Not insignificant is the reduction in image quality of files compressed for streaming. The streaming approach may be less desirable in a classroom setting where image projection magnifies the issues of compression and the equipment and tech support requirements for classroom use are more complex and expensive than for DVD. But streaming delivery is largely a lean-in activity, meant for personal rather than group screening. With the growth in use of tablets and other mobile devices for personal viewing screens, the issue of enlarged image quality is less important. Additionally, compression quality is improving rapidly. Commercial streaming services already provide HD (high definition) streams at a bit rate comparable to standard DVD. As Internet service providers increase bandwidth capability, streaming image and delivery are certain to improve (Brandon 2011).
As libraries acquire streaming video, other service issues arise, not the least among these is the value of shared collections and ILL. If content is locked behind an authentication firewall, will walk-in users be able to access it? Libraries have long depended on ILL to fill in collection gaps and to meet specialized needs of users. If collection access is limited to authenticated users, libraries will be unable to loan their streaming content. More importantly, however, they will not be able to borrow what is not in the local collection. E-journal licenses already address this issue; eBook communities of practice are beginning to. It remains to be seen how the problem will be addressed for streaming video.
Since the advent of home video, library circulation statistics have benefitted from video loans. As video use migrates from hard-copy circulation to online viewing, libraries will need to reconsider how to report this use and measure success in meeting user needs. Video use data collection is complicated by the plethora of portals through which users access streaming video and by the huge variances in how vendors collect and report use data. Use remains an undefined term. Some hosting services employ COUNTER-compliant (Counting Online Usage of Networked Electronic Resources) reporting, but the majority do not. The need to establish both industry and library standards for measuring use of streaming video collections is an issue still to be addressed.
A historic view of libraries is that they serve a major role as the curators, protectors, and guardians of cultural heritage. Libraries already are addressing issues with the preservation of books and journals. Services and protocols such as JSTOR, Project Muse, Portico, and the Western Regional Storage Trust have been developed and refined to assure preservation of the journal record. Project Gutenberg, the Google Books project, and the Hathi Trust are programs to protect and extend the life of monograph content. No such coordinated efforts are in place for preserving the video record. The Internet Archive, the Prelinger Archive, and the Vanderbilt Television News Archive are all efforts to assure description and digital access to some video. A more robust, cooperative effort to provide for long-term accessibility of educational or documentary film or video content is a major development that awaits resolution in the 21st century.
Streaming video is a relatively new format for libraries but has already reached the tipping point: 70 percent of academic libraries now provide streaming video (farrelly and Hutchison 2013) up from 33 percent in 2010 (Primary Research Group 2010). But streaming video has considerable distance to go to achieve its full potential, and is not without shortcomings and issues. As with other digital formats in libraries, these issues and shortcomings will be resolved as the delivery models evolve and mature. There can be no doubt, however, that streaming already is a major force in shaping user expectations and library video collection development.
Allen, I. Elaine, and Jeff Seaman. 2011. Going the Distance: Online Education in the United States. Available at: http://www.onlinelearningsurvey.com/reports/goingthedistance.pdf. Accessed March 6, 2013.
Brandon, John. 2011. “Video Format War: Blu-ray vs. Streaming,” PCWorld (October). Available at: http://www.pcworld.com/article/241215/. Accessed March 6, 2013.
Enis, Matt. 2012. “Patron Preferences Shift toward Streaming,” Library Journal 137 (14): 18.
farrelly, deg. 2008. “Use-determined Streaming Video Acquisition: The Arizona Model for FMG On Demand,” College & University Media Review 14 (1): 65–78.
farrelly, deg. 2012. Streaming Video: Internet Sites. Available at: http://libguides.asu.edu/StreamingVideo. Accessed March 6, 2013.
farrelly, deg, and Jane Hutchison. 2013. “Streaming Video in Academic Libraries: Preliminary Results from a National Survey” (speech). Presented at the Charleston Conference, Charleston, South Carolina.
Greenfield, Mark A. 2010. “The End of the Web as We Know It and I Feel Fine” (speech). Keynote address presented at the CCUMC Conference, Buffalo, New York. Available at: http://hosted.mediasite.com/mediasite/SilverlightPlayer/Default.aspx?peid=5e6f12a14a264567876f0807ef461a471d. Accessed March 6, 2013.
Handman, Gary, ed. 2002. Video Collection Development in Multi-Type Libraries: A Handbook. Westport, CT: Greenwood Press.
Handman, Gary. 2010. “License to Look: Evolving Models for Library Video Acquisition and Access,” Library Trends 58 (3): 324–34.
Kaufman, Peter B., and Jen Mohan. 2009. Video Use and Higher Education: Options for the Future. [New York?]: Intelligent Television. Available at: http://library.nyu.edu/about/Video_Use_in_Higher_Education.pdf. Accessed March 6, 2013.
Laskowski, Mary. S. 2011. Guide to Video Acquisitions in Libraries: Issues and Best Practices. Chicago: Association for Library Collections & Technical Services.
OLAC Cataloging Policy Committee & Streaming Media Best Practices Task Force 2009. Best Practices for Cataloging Streaming Media. Available at: http://www.olacinc.org/drupal/capc_files/streamingmedia.pdf. Accessed on March 6, 2013.
Primary Research Group. 2010. Survey of Academic Libraries. New York: Primary Research Group Inc.
Primary Research Group. 2012. The Survey of Distance Learning Programs in Higher Education. New York: Primary Research Group Inc.
Scholtz, James. C. 2002. “Developing Video Collection Policies to Accommodate Existing and New Technologies.” In Video Collection Development in Multi-Type Libraries: A Handbook, ed. G. Handman, 245–76. Westport, CT: Greenwood Press.
Spicer, Scott. 2012. Digital Video Collections Guide. Available at: http://connect.ala.org/node/183711. Accessed March 6, 2013.
Widzinski, Lori. 2010. “Step Away from the Machine: A Look at Our Collective Past.” Library Trends 58 (3): 358–77.
American Library Association Video Round Table. Available at: www.ala.org/vrt.
National Media Market. Available at: www.nmm.net.
VideoLib Discussion List. [email protected].edu
Academic library collections are evolving away from browsable shelves of print materials to a mixture of online and traditional media. This poses challenges for the patron’s ability to discover the available resources and to choose the most relevant from among them: challenges that require collaborative work between public services librarians, materials selectors, and cataloging staff. Transformative changes in the nature and complexity of the collections and in the nature of the metadata that supports discovery of a collection’s strengths require that collection managers learn more about cataloging, and catalogers learn more about how collections are used.
Discoverability is an essential part of what librarians do. While historically cataloging has been perceived as being focused on describing materials, the ultimate aim of cataloging is to help people find relevant works via description and organization, and convergently access them. To that end, it is desirable for librarians, regardless of specialty, to collaborate with their cataloging colleagues to maximize description of materials in their collections to the best extent possible. For example, subject specialists will know best which eBook collections would benefit most from supplemental metadata; reference librarians can provide details on which materials are most difficult for users to locate in the catalog as currently described; and ILL staff can help identify which materials are not being retrieved by linking software and possibly prioritized for addition of linking fields such as ISSNs or ISBNs. Given the size of most collections, it is important to prioritize efforts to maximize benefits to users.
Knowing what kinds of metadata improvements, also known as enhancement, can be undertaken locally requires understanding not only some basic cataloging but also how your catalog software functions—what fields are indexed, searchable, and displayed. As new catalog systems are implemented, librarians need to select which fields in existing records should be indexed and how they should be indexed (e.g., alphabetically for browsing, by keyword, or by numeric identifier). These same decisions come into play when choosing what types of enhancements are worthwhile. If your catalog software does not currently support indexing and searching of ISBNs, then that type of enhancement is probably not worth making a priority. Looking to the future, however, retrieval systems will evolve. For example, while most catalog software currently in use does not take advantage of the hierarchical structures inherent in tools such as the Library of Congress subject headings and related classification scheme, future discovery systems may be able to better capitalize on the usefulness of these structures for retrieval.
Good metadata improves discoverability and retrieval; it also reduces the number of questions users bring to our service points, both virtual and physical. Full text, while useful, is not the same as catalog access. Catalogs typically utilize controlled vocabularies that can aid users in narrowing their searching along subject lines. The biggest challenge users face in the evolving information environment is information overload—too many results—and good metadata is one method librarians can employ to aid users in sorting through overwhelming numbers of results. Next-generation discovery tools and databases prompt users to retrieve resources relevant to their needs with faceted results derived from the underlying title-level metadata. All library users benefit from good subject analysis through enhanced discovery points and controlled vocabulary that pulls together materials despite disparate terminology and languages. This latter point is significant to remember because materials in other languages and scripts will not be retrieved through use of keywords without metadata to provide English-language search terms.
Beyond providing a useful hierarchy, common thesauri, and keywords for searching, metadata provides a way for librarians to document the history of interrelationships between entities and works. It is one of the unique and valuable services that librarians provide. Via authority files and linkages among records, library catalogs essentially document the history and development of various works. It is via authority file linkages that users seeking works related to Mark Twain are led to works about Samuel Clemens. It is through the catalog linkages that users know that the title of a journal that began as Geological News Letter changed to GeoTimes, then became Earth. The latest iteration of widely accepted cataloging codes, RDA (Resource Description and Access), continues the long-standing efforts of the cataloging community to adjust the rules and standards of their craft to improve users’ success at finding the resources they are seeking. Perhaps the most exciting change brought by RDA is to the authority file, a system that assigns authoritative versions of the names of authors, historical persons, corporate bodies, and musical and literary works. Using the assigned form of name in the catalog ensures that references to the same person or work index together, so that patrons find the resources that can assist with their research. Name records written under the new RDA standard will carry much more information, such as notes on the history of a corporate body, and links out to Web sites in addition to the citations for the sources of information that were already there. The new authority file will also interact with sources such as Wikipedia and the Virtual International Authority File (VIAF), an authority file that merges information from over a dozen national libraries in Europe and across the globe.
The catalog is a shared resource that all librarians should and can contribute to. Among many librarians there is a sense that the catalog is the sole creation of catalogers and therefore territory into which a noncataloger should hesitate to venture. Yet much as librarians provide feedback to publishers and database vendors on the quality and characteristics of their products, they should be as, if not more willing to provide feedback to their cataloging colleagues who quite likely work for the same institution. After all, if we cannot talk with our cataloging colleagues when they are just down the hall, how much more difficult will it be when or if cataloging is centralized across organizations or consortia? Librarians put infinite care in selecting materials for the collections, so it behooves us to partner with cataloging staff to make sure that the resources are discoverable. The pathways to becoming an active participant in creating a superior catalog resource are myriad but must include at a minimum finding ways to communicate and work with cataloging staff. Illustrations of how librarians can learn more about cataloging, examples of why this knowledge is useful, and ways to contribute to the catalog are explored later.
As new materials are received, it is useful to not only examine the work to become familiar with the content but to examine the metadata as well. Through this method, you can become familiar with the common subject headings for works in your library as well as their accompanying classification assignment. This is also the best point at which to identify deficiencies in the record describing the piece, such as a need for additional subject descriptions or other access points. Given libraries’ reliance on copy cataloging, it should not be assumed that the metadata will reflect all the aspects of the work that your patrons need. For example, you may have purchased a work in response to a subject of local interest. It is important to examine the metadata to verify that the topic of interest is adequately described in the record so that it will be retrieved in a user’s search. If more access points are needed, the solution is simple. Ask your cataloger to enhance the record with additional headings or notes that reflect your local strengths. In some cases, the classification assignment may need to be adjusted to better collocate the resource with other materials in your collection. In other cases, the subject specialist and cataloging staff may need to consult about whether the item in hand is better treated as a serial publication or as a monographic one. It is important that you know enough about how the cataloging affects searching and retrieval so that your users benefit from the combined expertise of the library’s staff.
Knowledge of the major subject headings and classifications in your subject area is valuable for additional reasons. Approval plan vendors and bulk-content providers frequently organize their selection profiles based upon subject or classification schemas such as the Dewey Decimal or Library of Congress classifications. Selection tools such as the British National Bibliography are organized by classification scheme. Familiarity with the classification structure helps you ensure that all aspects of a subject are not only collected, but are also collocated in your library. For example, in the Library of Congress system, the G and Z classification numbers, maps and bibliographies, respectively, have validity across many subject areas. To illustrate, if you are trying to collect all major works related to forestry, it is important to realize that bibliographies on the topic may class in Z rather than in SD; maps and atlases on the topic will class in G. Additionally, it is useful to know of the changes in the classification schema over time. As an example, prior to the creation of the GE classification in the Library of Congress system, environmental science materials were classified in a variety of areas, including QH (biology), GF (human ecology), and others (www.loc.gov/acq/devpol/environ.pdf). Knowledge of how the classification scheme used at your institution works at the subunit level is useful in space planning, given that different aspects of a field grow at different rates. Imagine, for example, the challenges faced by the librarian doing space planning in the gaming section of the collection. How fast does the collection grow in the area for board games versus the section on video or computer games? Finally, data on collection usage is generally reported using call-number ranges so it is helpful to know how those correspond to the subject areas. Data on which parts of the collections are being most and least heavily used can inform not only collections managers but also can be used to guide discussions and priorities with cataloging staff. Potentially, records for materials in underutilized areas of the collections could be examined and targeted for metadata enrichment to aid in increasing usage.
The catalog record is also a place to record item-specific data such as signed or numbered copies, donor information, and other information valuable to current and future users. Other examples of information useful to record include preservation work performed and information related to the artifactual value (e.g., quality of binding) of the piece. The Institute of Museum and Library Services–funded project Publishers’ Bindings Online (bindings.lib.ua.edu) provides a colorful illustration of why such information is useful. As librarians move to shared archives of historic materials, it is important to retain the most complete, most representational originals. Accurate metadata is essential to facilitate this process. Historically, item-level metadata has received much less attention than bibliographic metadata, yet to adequately facilitate the merging and sharing of collections across institutions, item-level metadata is key. For example, does the work retain all accompanying parts and pieces, such as accompanying maps? Which copy of a film was purchased with public performance rights? Some libraries may have bound issues of a periodical yet not noted that the holdings were incomplete, that covers were removed and discarded, or that less-than-facsimile replacement pages may have been utilized (e.g., black-and-white photocopied replacements when the original, color version was lost). Such information can and should be recorded in the metadata.
Increasingly, libraries are able to not only purchase a work, but also its metadata record simultaneously. Major approval plan vendors such as YBP and Coutts provide MARC records, sometimes enriched with tables of contents and other data elements, at the time of purchase. Publisher-supplied metadata, considered by many publishers to be a freebie and not worth investing a lot of time, is unfortunately often less than ideal in terms of completeness and specificity. For example, while the general heading of “Engineering” may adequately describe a subject category among a publisher’s offerings of new imprints, it is less than helpful as a subject heading to be used to retrieve a work on dam building. eBooks (and other eFormats) pose particular challenges in terms of the quality and extent of their metadata. From a user’s standpoint, it is useful to know whether the work is in the form of an HTML file, PDF, Kindle edition, or other format as well as indicating any platform or license-driven restrictions (e.g., one or more simultaneous user, no downloading, limited number of uses). This information is typically lacking from the metadata record. Bulk purchases of eBook collections pose particular challenges. In some cases, the purchase is really a lease and records may only be temporary (and thus not worth enhancing), whereas in others, the size of collection is such that creation of metadata records on an item-by-item basis is unrealistic. Metadata records for works in collections can sometimes be purchased, in which case knowledge of the basics of cataloging improves your ability to evaluate these records for adequacy and judge whether they are worth the asking price.
Ultimately, it is essential to understand that the catalog record is not sacrosanct. It is okay to edit catalog records and most catalogers will welcome your input, interest, and expertise. Keep in mind, however, the importance of maintaining a balance between enhancing existing metadata records and other needs, such as retrospective work on uncataloged materials. No library has the resources to enhance every record; therefore a focus on the most important, unique, and hardest-working parts of your collection is required. Public service librarians need to share their knowledge of the collections and the ways they are being used with their cataloging comrades in order to set appropriate and reasonable priorities.
In addition to being able to purchase metadata records, several vendors provide fee-based content enrichment services for recently published material, typically providing tables of contents, summaries, genre terms, cover art, and so forth. Enrichment, however, can be done locally on a case-by-case basis. (The term enhancement typically refers to bringing a record up to date or to a minimum standard, whereas enrichment refers to adding various optional fields above and beyond the minimum.) Given the size of most libraries’ collections, librarians must be selective in choosing which materials are worthy of enrichment. Choices of what to enrich should be specific to your situation and institution, along with the decision of whether to contribute your improvements to the broader library community. Some may choose to enrich records in areas of collection strength; others might choose to enrich records based upon other criteria such as those that further the institution’s mission (e.g., works related to diversity). Records for items going to storage (or otherwise restricted from user browsing) are particularly important to examine for quality of the base record as well as enrichment. Does it have the typical hooks used by software for match and retrieval, for example, ISSN, ISBN, LCCN? For multivolume works, is it clear what the content range of each volume is? Consider the example of the collected works of Shakespeare: how easy is it to determine which volume contains the play Titus Andronicus? Does the record include information about related works, such as a teacher’s guide, or where the work is indexed? Are there accompanying pieces worthy of additional access points or metadata? An example might be an accompanying map or poster that has a unique title or subject focus. Is it useful to indicate that the work exists in a translated form or alternative format, such as online? What about the case where the alternate format does not exactly match the copy of record (which is increasingly the online version)? For some periodicals, the title of the online version differs from that of the print, retaining only the latest title for all volumes rather than documenting the history of title changes over the life of the publication. This presents problems for users and ILL staff when a citation to the former title does not match the metadata. In another variation, the online edition of a journal may lack sections contained in the print, such as book reviews or letters to the editor. The metadata record is an excellent place to record such information for the benefit of current and future users (including librarians).
Sometimes the metadata record simply does not provide information sufficient to help potential users identify the work. If the record is old or minimal, it may utilize obsolete cataloging terms or conventions. Standards and capabilities change over time. As an example, for many years the cataloging cooperative OCLC restricted the maximum size of a catalog record to 50 lines. As materials selectors strive to build collections on the topics of special interest to their faculty and students, the cohesiveness of strong, deep collections may no longer be apparent from the metadata that represents the library’s holdings to its local clientele and to the world of scholarship. A tune-up of older metadata may be in order. Perhaps the metadata utilizes outdated vocabulary—the subject heading “Groundwater” was once “Water, Underground”; the heading “African American” was once “Negro.” New concepts such as nanotechnology or artificial intelligence appear in the literature; however, controlled vocabulary to describe these new fields typically lags behind until usage coalesces around a new term. Few libraries go back to the records of those early works to update the metadata with the newly established subject headings. Metadata for serials may deserve special attention due to likely changes in the resource over time. The journal’s supplements spin off as separate publications, and then are reabsorbed. The scholarly society that publishes the work has changed its name. The title on the covers has varied slightly. These changes will not be reflected in the legacy record that was downloaded to your local catalog long ago. It is particularly important to keep your authority records up to date so that, for example, J.Lo’s early works will be found when searching for works by Jennifer Lopez. Perhaps the metadata record simply lacks an ISBN or ISSN, which are now commonly used to link related resources. Knowing the kinds of enhancements possible can greatly improve retrieval and use of materials.
Another consideration when prioritizing local efforts is whether another organization might be taking the lead on providing enhanced access. In many cases, such efforts follow geographic boundaries, for example, an Arkansas library enhancing metadata related to Arkansas works. If a government agency or consortium is in the process of digitizing certain publications, perhaps local enhancement efforts should concentrate on other materials that cannot be as readily found. In the end, the specifics of your local situation will influence what can and cannot be done. Accepting imperfect records for some of the materials in your collections is just going to be a fact of life in the future. It is also worth noting that, if portions of your collections remain completely uncataloged, your organization’s time may be better spent on describing those materials rather than on enhancing records already in existence, however imperfectly described. With space at a premium, few libraries can afford to retain and continue to store materials that are invisible to users and therefore receiving little or no use. In the end, catalogers must rely on the experience and subject knowledge of public service librarians to help determine the priorities for improvements in metadata.
The catalog is a collaborative endeavor continually added to by successive generations of librarians and others. Catalogers need feedback in order to maximize their impact and should partner with their library colleagues in building metadata records that will contribute to global efforts to record the fullest history of human knowledge. We will always need to provide access to the new but attention must also be paid to keeping what you currently have (your existing metadata) up to date, fresh and usable, responding to the new queries that your clientele bring to the collections. As such, noncataloging librarians can help improve the corpus of knowledge by keeping catalogers informed of difficulties users are having in locating particular resources, or of the various ways that users are searching for materials (color, size, geographic coordinates, etc.). It is helpful to alert cataloging staff to areas of emerging interest or research priorities that need special attention. Cataloging librarians should seek out opportunities to interact with public service librarians and cooperatively set goals for not only getting materials cataloged but also identifying resources deserving of enhancement or enrichment.
No one library can afford to do it all, when it comes to either collecting or cataloging. Fortunately, librarians have long recognized this latter reality and have a history of collaboration on shared metadata, as evidenced by WorldCat, which is a combined catalog of works contributed to by a multitude of libraries around the globe. Where most libraries currently share metadata, some are beginning to experiment with giving up their local catalog in favor of utilizing this global resource as their primary access tool. This raises the question: in a world of a single, shared catalog, who will be allowed to edit records? As more libraries move from shared cataloging to shared catalogers, communication among public service librarians and those catalogers privileged with writing and editing access will require ever more deliberate efforts by both parties.
Libraries are undergoing rapid change. Online resources have displaced collections formerly devoted to materials in tangible media (i.e., print, vinyl, etc.). Many academic libraries are moving toward sharing collections in remote storage facilities or jointly bundled in online collections. Reciprocal arrangements to collect in complementary topics and subdisciplines are becoming more common. Legacy collections, backed up online, are being reviewed for withdrawal or relegation to remote storage. ILL services, assisted by digital collections and scanning, can now swiftly fulfill many patron requests, freeing shelf space in the library for other uses. Even smaller libraries that have no current plan for shared collection coordination will be impacted, as their ILL options will change when loan partners and major research centers move more and more toward shared collections. Such increased sharing of collections will only increase the need for accurate, high-quality metadata.
Cataloging is an evolving craft. Subject headings change and evolve, as do cataloging standards and formats, but the essential goals of cataloging remain unchanged. We catalog in order to help people find information. We catalog in order to document relationships between works. We catalog to describe the uniqueness of items in our collections. In an increasingly interconnected information environment, we catalog in order to connect the information and resources in our local collections with the world of knowledge and the cultural and scientific heritage of each community and academic discipline. We catalog for the future.
Dempsey, Lorcan. 2012. “Thirteen Ways of Looking at Libraries, Discovery, and the Catalog: Scale, Workflow, Attention.” EDUCAUSE Review Online (Monday, December 10). Available at: http://www.educause.edu/ero/article/thirteen-ways-looking-libraries-discovery-and-catalog-scale-workflow-attention. Accessed October 30, 2013.
Smith, A. Arro. 2008. “Cataloging Heresy.” In Radical Cataloging: Essays at the Front, edited by K. R. Roberto, 291–99. Jefferson, NC: McFarland.
Tosaka, Yuji, and Cathy Weng. 2011. “Reexamining Content-Enriched Access: Its Effect on Usage and Discovery.” College & Research Libraries 72 (5): 412–27. Available at: http://crl.acrl.org/content/72/5/412.full.pdf+html. Accessed October 30, 2013.
Since its creation in 1876, the Dewey Decimal Classification system has been adopted by more than 200,000 libraries in over 135 countries. It has survived the invention of the airplane, two world wars, the rise of television, and the Information Age, making it a global standard for library classification, which has endured for more than a century. Yet a recent trend has some libraries replacing Dewey in favor of word-based models inspired by the Book Industry Standards and Communications (BISAC) subject headings published by the Book Industry Study Group (BISG).
One reason for this shift may be Dewey Decimal Classification’s heavy emphasis on Western, American, and Christian thought, which has become limiting and inadequate in our multicultural society. The topic of religion provides the clearest illustration of Dewey’s structural biases. People of all faiths utilize the library, but a vast majority of Dewey’s classes on religion focus on Christian topics; some major world religions are not even represented as a distinct class (Weinberger 2004; Olson 1998). Religion is just one of many topics where American or Christian ideas are dominant in Dewey Decimal class structure, leading some libraries to replace Dewey in order to present a more balanced system to their users.
Because of these biases, libraries globally are required to do a great deal of work to adapt the system to local needs. For example, much effort has been made to adapt Dewey to the subject needs of Islamic libraries (Sulistyo-Basuki and Mulyani 2008; Khan 2004). Librarians in some countries, like the Republic of Korea, have even built their own decimal classification to address local classification needs (Dong-Geun 2012). For some libraries, word-based systems offer an alternative to modifying Dewey to reflect the communities they serve.
Others recognize that arranging and accessing items classified using a decimal-based system like Dewey no longer reflects the way people access information in the era of keyword searching. Dewey was developed in a time before digitized materials, and linear arrangement was the key to findability. Dictionaries, encyclopedias, indexes, and card catalogs were organized in alphabetical order from A to Z. With its decimal order from 000 to 999, Dewey maintains that linear approach in the way subjects are arranged and materials organized on library shelves.
Modern keyword searching is often ordered by relevance, allowing the user to select the order in which results are displayed. The flexibility of customer-centered arrangement is not easily reflected in a Dewey-organized collection. In a Dewey collection, a user enters the continuum at one point and must navigate forward or backward to find what he or she needs, requiring the user to pass a number of irrelevant titles before reaching the relevant ones. In a keyword system, the user has multiple entry points and can often determine where those entry points appear. A word-based system offers library users more direct entry points to the collection with closer proximity to the most relevant titles they are looking for.
One library in Colorado decided to move away from Dewey Decimal Classification in part because of a shift in its core philosophies. When the Rangeview Library District in Adams County (Colorado) reinvented itself as Anythink, it created the experience model, where all of their services have been reexamined with the customer’s experience in mind. Anythink libraries are places where customers have meaningful, empowering experiences; they are no longer simply warehouses full of books and computers. Part of becoming an experience-based library was moving away from some of the features of a traditional library that Anythink identified as barriers to customer empowerment, inclusion, and interaction. Anythink eliminated overdue fines, built libraries for people instead of materials, changed their job roles, replaced battleship-style service desks with smaller perches—coupled with a roving model and heavy emphasis on hospitality—and switched from using Dewey Decimal Classification to a word-based system.
Giving up Dewey was not easy, but Anythink was fortunate to have a model to build from in the work of the Maricopa County Library District in Arizona. Maricopa County’s experience moving from Dewey to a BISAC system inspired Anythink in 2008 to create its own classification scheme called WordThink. WordThink uses the BISAC subject headings assigned to books by publishers as a guide for classifying materials into 83 classes and 385 subclasses. Classes are arranged in each library by neighborhoods, or groups of similar subjects, to create an intuitive flow between subject areas. Within each class and subclass, individual items are arranged on the shelf in alphabetical order by title. Work began on this reclassification in March 2009. It took more than 1,000 hours of staff time and almost a year to reclassify all of the items in Anythink’s nonfiction collection; the WordThink conversion was completed district-wide in early 2010.
Anythink’s intent was to create a system where customers were empowered to intuitively find the materials they were looking for. Rather than educate customers about a classification system that few outside of libraries understand on a deep level, Anythink chose to use a word-based classification scheme that allows its customers to encounter information on their own terms. Customers would no longer feel pressured to understand a secret librarian code in order to find materials.
Some may believe that the classification system used in a library has little impact on customer feelings of empowerment, but an online survey conducted by Barbara Fister in 2009 showed that more than 60 percent of library users who have trouble finding nonfiction materials are “intimidated by a classification system they don’t understand very well” and/or “want to go straight to the right shelf without having to look anything up” (Fister 2009, 24). In other words, customers are intimidated by Dewey’s secret code complexity, which becomes a barrier for people who want to find information on their own.
Anythink also saw a move away from Dewey as an opportunity to arrange materials in ways that would help create positive experiences for its customers. Breaking up the Dewey subject order allows library staff to build associations between materials, encouraging customers to interact with information in new ways. These associations could be on a grand scale in the way subjects are placed throughout the library’s spaces, or on a smaller scale through merchandizing and displays. As a customer browses shelving or displays, a related topic or beautiful item might catch his or her eye, spark an interest, and lead to a new experience. Anythink sees its materials as an important tool in building experiences for its customers, and the move away from Dewey was the first step in that process.
Since instituting WordThink district-wide, Anythink has seen a variety of positive outcomes in its libraries. These outcomes include more flexibility and customization for the nonfiction collection, improvements in findability, reduction in cataloging and processing costs, and the removal or minimization of barriers to customer empowerment and self-service.
A word-based classification system is customizable in ways that a standard code-based system like Dewey is not. Word-based systems accommodate new areas of knowledge simply by adding additional top-level classes and classes or subclasses can be easily created for subjects with high local interest. A customized word-based system also allows a library to create relationships between materials that aren’t available in standard Dewey classification. To create its own word-based system called WordThink, Anythink made some adjustments to the existing hierarchy of BISAC subject headings and decided to use words on spine labels that were as close as possible to the words its community was already using to describe subjects.
Not only did Anythink create top-level classes outside BISAC standards, like parenting, it also joined some subject headings together to create subclasses while leaving others to stand on their own. For example, users with interests in outdoor sports, such as hiking, fishing, and camping, were often interested in more than one of those topics, so Anythink combined materials with those BISAC subject headings into one WordThink subclass—“SPORTS OUTDOOR.” This is a great example of using a word-based system to reflect Colorado’s community interests. Spine label words were also carefully chosen to match terms that customers were already using to describe materials. For example, “FARMING” and “CAR REPAIR” were used as spine label terms rather than the matching BISAC subject terms “agriculture” and “automotive.” Anythink also created a Colorado history subclass under history unique to WordThink, another example of responding to community needs. Customizations such as these allow Anythink to tailor its collection and classification scheme to its customers in ways that were not available with Dewey Decimal Classification.
While Dewey allows a library to classify all of its materials under a single subject hierarchy regardless of target audience, word-based systems like WordThink can leverage the separate BISAC subject headings for children’s materials to offer a unique classification scheme to different age groups. At Anythink, there is a separate classification scheme for children’s materials and adult materials. Some of the subjects are the same in both schemes, but Anythink uses WordThink’s flexibility to emphasize certain topics in its children’s collection. For example, U.S. presidents get their own subclass of biography in the WordThink classification for children while similar titles are included in the single large biography class in the adult scheme. Spine labels are targeted to younger readers with different words like “BUGS” instead of “INSECTS” or “SCARY” instead of “HORROR.” These changes allow Anythink to create unique associations between materials and experiences for younger readers.
WordThink increases the flexibility of Anythink’s nonfiction collection in the way materials are arranged in the library’s physical spaces. With Dewey, there is a direct relationship between subject classification and shelf order. This relationship isn’t always clear to library users or library staff, but it is there nonetheless. Maintaining that order can limit a library’s ability to modify placement of materials within its physical space because a long, linear classification scheme fits best on long, linear shelving units. Before WordThink, Anythink’s nonfiction collection was spread over multiple long ranges of shelving with one primary beginning of the collection. Customers and staff looking for a specific item or subject needed to understand where they were in the 000 to 999 continuum before figuring out which way to move to find a specific item. Word-based systems like WordThink break that link between subject classification and shelf order, giving the library greater flexibility to design physical spaces that meet the needs of its customers. Shelving sections can be varied in length depending on the size or subject of the class. Popular subjects like cooking or nature occupy shorter shelves to facilitate easy browsing, while other subjects like philosophy or social science occupy taller shelves.
Anythink uses this flexibility as an opportunity to place certain subject areas where they will see the highest use or where they will help create experiences for customers through their relationships to other library services. For example, books on computers and computing are located near public computers whenever possible. Books in the parenting class, from child rearing to baby names, are located in or near the children’s area. Children’s books on fairy tales, poetry, and concepts are located close to picture books to group them by age appeal rather than Dewey class or subject. Homework help subjects in the adult collection, such as math, science, language skills and test prep, are located together in areas accessible to library users of all ages. Arrangements can even vary from branch to branch, depending on branch size and community interests. Since WordThink was launched in 2009, Anythink staff have also revisited the arrangement of materials at each branch and adapted the layout of materials to reflect customer interest. At one location, customer feedback led to a shifting of materials in the adult collection that made titles in the “REFERENCE” class more accessible. This kind of responsive shifting was facilitated by the flexibility of WordThink.
Merchandising is also easier under a word-based classification system. Materials are shelved at Anythink in such a way that each class or subclass begins and ends on its own shelf. Because of this, facing out items always results in highlighting similar topics. This also makes it easier for staff to pull materials for larger displays since staff can browse for related titles in once place instead of searching through the entire collection.
WordThink is designed for customers to navigate the collection self-guided if they choose, eliminating the need for staff to translate what they are looking for into a code-based call number. Customers now use natural language to access materials without the intermediate step. A customer looking for materials on diabetes can go directly to the shelf marked MEDICAL DIABETES without having to look up a call number. The time that staff spent cracking the code for customers is now spent on readers’ advisory, reference questions, and building meaningful relationships with customers.
The same keywords customers use to search the library’s catalog are reflected in the classification system, spine labels, and signage. Using BISAC subject headings and word-based classification also allows Anythink to add natural-language entries to the library’s catalog records. Some titles have become easier to find with the addition of BISAC subject headings as searchable keywords, since BISAC uses more common words than Library of Congress in its subject headings. At the same time, adding BISAC subject headings to the catalog groups titles in different ways than other subject headings used by the library. This allows customers to discover materials in another way if they choose to browse the catalog using the BISAC subject hierarchy.
Anythink has economized its cataloging and processing tasks with WordThink. While library staff do minimal original cataloging, it is much quicker to classify materials based on subject using WordThink than constructing Dewey numbers. Printing and applying spine labels is also easier since most of the nonfiction collection uses standardized call numbers and the same spine label for every item in a class or subclass. Staff can now print sheets of labels in advance instead of individual labels for each item.
Although a word-based classification system has been successful at Anythink, it may not be the best solution for every library’s collection. Like any classification system, it comes with its own set of concerns and issues that can affect implementation and usage. Consider the following issues before making the switch.
Collection size plays an important role in the suitability of word-based classification. Some of the findability and flexibility benefits may be lost when the number of items in any one category or subcategory exceeds a certain point. At Anythink, target category and subcategory sizes are set at one shelving range per category and about one shelving section or bay per subcategory. Maintaining a cohesive shelving arrangement becomes difficult when a category or subcategory has so many items that it spans more than one shelving range. When a category has too many items, Anythink considers revising that class to add additional subcategories in order to maintain the same level of browsability. Categories with fewer items are also easier to relocate or shift if necessary due to shelving needs or merchandizing plans. This becomes more difficult as the number of items in a category or subcategory increases.
A word-based classification system like WordThink is also more appropriate for collections that require only broad categories for classification, not specific and granular classes. Anythink has made the conscious decision not to break down its collection into too many subcategories, resulting in some categories and subcategories that have a fairly broad range of different topics, such as politics or diseases. For libraries that need more granularity, these broad categories may not suit their needs.
Furthermore, word-based systems do not yet have the detailed, written documentation like Dewey or Library of Congress. WordThink is currently documented through two spreadsheets of fewer than 15 pages each and a related translation key that translates BISAC subject headings into WordThink classification. While classification tasks are easier and quicker than with Dewey, there is the possibility that the scope of similar categories is not defined well enough to enable catalogers to precisely locate titles. For example, the distinctions between “SCIENCE” and “NATURE” or “HEALTH” and “MEDICAL” may be too nuanced for quick classification by someone unfamiliar with the system. Further documentation describing the differences between the main classes and subclasses would help improve the clarity of WordThink.
Word-based systems require different shelving arrangements and a stronger emphasis on way-finding tools like signs and maps. Traditional library shelving models of long linear ranges are not optimal for word-based systems. Instead, a variety of shelving lengths and heights may be the best fit so each category can have its own distinct shelving space. This is definitely something to consider for libraries interested in switching to a word-based system. Is there room for flexibility within the current shelving arrangements? Can the space be redefined to support nonlinear organization?
Signage is extremely important to define areas and place categories in context in a word-based system. At Anythink, each top-level category has its own large sign placed on the top of the shelving unit, and each subcategory has a smaller in-shelf sign where the subcategory begins. Detailed maps identify the shelving units for each category. These solutions have worked well so far, but more work still needs to be done at Anythink to improve customer way finding and make accessing the collection even more self-driven for customers.
Once you break the connection between classification number and shelving location, you introduce some shelf order issues not present in Dewey or Library of Congress. Ordering items by title as Anythink does is more difficult when the title of the work is not clearly distinguishable. Most commonly, this issue arises when subcategories contain many works with similar branded titles from the same publisher like The Complete Idiot’s Guide or Fodor’s. This issue could be solved by using a different data point to order materials on the shelf like author name or Cutter number, but the addition of those bits of information to each spine label would likely negate the time savings in processing that a word-based system provides. Anythink has decided to leave author, title, and Cutter number off its spine labels to maintain efficiencies in cataloging, processing, and shelving.
Probably the most far-reaching issue related to the use of word-based classification is the lack of a library standard. Though many libraries using word-based systems use BISAC as the foundation for their schemes, the actual implementation varies from library to library. At this time, not having a standard increases the learning curve for customers and staff who may be used to a different system from another library. Each unique system requires some degree of retraining to help customers and staff memorize the differences. Anythink has also struggled at times to find outside technical services vendors (cataloging, processing, collection analysis, etc.), willing to work with the library’s classification needs.
Word-based classification systems have gained a foothold in the United States, and libraries continue to explore new ways to organize their collections to improve customer service. Large library cataloging vendors are providing services using word-based schemes, and other businesses working with libraries increasingly see the value of adding support for non-Dewey classification. The development of a robust word-based classification community is well in progress although not yet as mature as the Dewey Decimal Classification community.
As flexible and customizable as word-based classification systems are, one of the biggest challenges to making such systems a more widespread choice for libraries comes from the lack of standards. In the United States, there are several distinct word-based systems currently in use. Some stick closely to the BISAC subject headings, whereas others, like WordThink, use them as a foundation but map those headings to locally defined classes and subclasses. It remains to be seen whether these systems will continue to develop in silos or whether libraries using similar systems will join together and develop a common classification scheme.
On an international level, one aspect of standardization is being addressed. In October 2012, the book industries of 15 countries joined forces on a move to combine several book industry subject categorization standards into one global standard. Thema, this new standard, is intended to facilitate the sharing of subject classifications between book markets currently differentiated by language, country, and culture. The long-term plan for Thema is to replace each market’s own standard with a global one, replacing BISAC in the United States and other standards in markets across the globe. Libraries currently using systems inspired by BISAC will need to evolve toward Thema as it is implemented worldwide. Thema may also increase opportunities for word-based classification systems to become more prevalent on an international level as more countries begin using this new global standard. Maybe libraries can look to the book industry for inspiration in developing their own flexible, global standard for word-based classification.
No classification system is perfect. Like any system, word-based classification has its own strengths and weaknesses. For Anythink, the strengths of flexibility, customization, efficiency, and empowerment outweigh the benefits Dewey brings as a worldwide standard with a high level of subject specificity. As more libraries develop their own word-based systems, vendor support continues to increase, and standards are adopted nationally and internationally, it will be easier for libraries to transition away from Dewey. Library users will ultimately be the driving force, as collections should be organized with their needs in mind.
Dong-Geun, Oh. 2012. “Developing and Maintaining a National Classification System, Experience From Korean Decimal Classification.” Knowledge Organization 39 (2): 72–82.
Fister, Barbara. 2009. “The Dewey Dilemma.” Library Journal 134 (16): 22–25.
Khan, Sher Nowrooz. 2004. “Expansion of DDC 21st Edition Number 297.63 for the Sirah.” Pakistan Library & Information Science Journal 35 (4): 16–25.
Olson, Hope A. 1998. “Mapping Beyond Dewey’s Boundaries: Constructing Classificatory Space for Marginalized Knowledge Domains.” Library Trends 47 (2): 233.
Sulistyo-Basuki, L., and Alit Sri Mulyani. 2008. “Indonesian Librarians’ Efforts to Adapt and Revise the Dewey Decimal Classification (DDC)’s Notation 297 on Islam.” Malaysian Journal of Library & Information Science 13 (2): 89–101.
Weinberger, David. 2004. “Free Dewey!” KMWorld 13 (9): 18–30.
Librarians have long desired to improve the user experience by offering search tools that are aligned with modern expectations, such as using a single search across multiple data sources. Libraries first met the challenge of searching library resources across vendor platforms by developing or acquiring federated search software, which broadcasts the user’s search query to many different search engines and returned results that the software attempted to combine into one results set. User satisfaction with this type of software was mixed, but most libraries and vendors acknowledged its inherent limitations, particularly related to relevance ranking and the use of search syntax and limiters across different systems (Cervone 2005; Jung et al. 2008; Randall 2006; Wrubel and Schmidt 2007). Federated searching did not pose many new collection development issues because each data source was still searched individually.
Discovery tools have provided a new, welcome alternative to federated search. A key strategic advantage of discovery tool software has been the combination of records from a library’s catalog into one index with journal article metadata in advance of the user’s search, interfiling results into a single interface, with facets available for refining results across the system. These tools tend to search extremely large collections of items; Serials Solutions’ Summon discovery tool included 950 million de-duped items as of May 2012 (Nagy 2012). Although the tools seem to be an improvement over federated search (Williams and Foster 2011), they still pose challenges for library collections access, growth, and maintenance. This chapter first discusses the ways in which discovery tool software is changing access to information sources. Then, it examines how collections are accessed inside and outside of the discovery tool, and the implications of discovery on collection development policies.
Figure 21.1
James Madison University Discovery Tool Purpose Statement, Fall 2011
An important consideration underlying all these facets is a given library’s conception of how the discovery tool fits within its suite of systems and resources. Every library should identify and articulate a purpose statement for its discovery tool to inform the numerous, inevitable decisions that will follow implementation. The statement will specifically identify the purpose of the discovery tool and how it relates to other means of discovery of library content, and therefore how it will affect overall collection development decisions. In 2011, a cross-departmental team at James Madison University (JMU) drafted a purpose statement and presented it to library management, where it was approved (Figure 21.1). It was recently reviewed by a different cross-departmental group that created purpose statements for all the library’s major systems, and has been used to inform numerous discussions. This chapter occasionally suggests that some decisions depend on a library’s particular situation; the discovery tool purpose statement clarifies the tool’s role in individual libraries’ collection development approaches.
Discovery tools provide access to multiple source types (books, eBooks, journal articles, government documents, theses, etc.) simultaneously after combining different databases into a unified index. This creates a search tool that is more aligned with user expectations, at least on the surface, but this new paradigm also raises questions and challenges. The very act of combining records from multiple databases ensures that discovery tools will face challenges in harmonizing different data structures to support search algorithms. Due to ever-changing publisher agreements and library subscriptions, the discovery tool will be able to search full text for some, but not all items. Also, the differing nature of the information source types searched by discovery tools presents issues related to search strategy and information delivery. Finally, while metadata will continue to be important, discovery tools have the potential to change the role of indexing and abstracting databases in the industry as well as at individual institutions.
At one time, database searching involved a single record structure. A&I companies or professional organizations such as the American Psychological Association (APA) or Modern Languages Association (MLA) created, edited, and enhanced metadata about relevant content in a disciplinary field, often using discipline-specific subject headings and thesauri. This provided a presentation of the literature in the field using the terminology of its scholars (Benjamin Jr. and Vandenbos 2006, 950); although it is important to note that not every index used descriptors or a controlled vocabulary from its inception (Alexander 2001). Additionally, some indexes and abstracts’ development was influenced strongly by public policy (Weiner 2009). Both the record structure and the descriptive vocabulary used created a subject-specific experience. Therefore it was easy for libraries to collect and promote access to these collections, as money was available, because of the subject specificity. In fulfilling their single-search experience, discovery tools have taken these subject-specific resources and combined them into one unified index, without much regard for their unique qualities.
This combination of different databases into one unified index has required librarians to rethink how users search these collections. EBSCO Discovery Service and WorldCat Local combine records from many different databases, striving to de-duplicate by choosing which database’s record to display; while Serials Solutions aims to “combines data from multiple sources—such as full text, abstracts, and subject terms—into a single, discoverable record” (Serials Solutions 2012a). These techniques are imperfect, because discovery tools still leave some duplication of items within results sets. For tools that choose one record in cases of duplication, librarians need to stay alert to which database records are chosen and whether all relevant metadata have been merged. For example, when EBSCO first merged H. W. Wilson databases into EBSCO Discovery Service (EDS), OmniFile records were used rather than those from subject-specific databases, such as Art Abstracts. Although EBSCO seems to have rectified the situation by merging the records, the important “Artist & Work” field of Art Abstracts was originally missing. This error drastically impacted the search results and the ability to limit those results.
Metadata from subject-specific indexes and abstracts are still important and relevant in the age of discovery. Metadata allow greater bibliographic control and search precision. Features such as faceted browsing and “find similar items” use metadata to supplement naïve users’ search habits. Also, previous studies have shown that subject headings greatly improve keyword search success (Gross and Taylor 2005). The advent of the semantic Web illustrates that although metadata structures are changing, online information requires descriptive metadata for effective use. Even large, public search engines like Google and Yahoo! are using schema to improve search results (e.g., http://schema.org) rather than relying on full-text indexing. In the age of discovery, records and their metadata are an important representation of a library’s collections.
Regardless of vendor, discovery tools search records that have different information structures from one another. With EDS, some databases only have their metadata indexed, while others index some or all of the full text associated with the record. Although Serials Solutions is customized to include records for just the library’s full-text holdings, the metadata and full text may be from different sources. As they explain, “In some cases, the Summon index may contain a record with citation metadata from an A&I provider as well as full text from a publisher. In other cases, the Summon index may contain a record and full text for an item included in an A&I database, but not contain the metadata from the database provider” (Serials Solutions 2012a). The situation grows more complex when the vendor imports the library’s catalog into the index, choosing how the numerous MAchine-Readable Cataloging (MARC) fields are mapped to discovery tool’s facets. For example, Figure 21.2 shows an example specification of how MARC fields are mapped to discovery tool fields, the type of indexing, and where the data in the fields displays in the results set. Depending on the discovery tool vendor, this mapping may be customizable by the individual library.
In addition to the questions of what fields are available in the records and whether the fields have controlled vocabularies, some discovery tool records have included full text in the index. Full-text searching adds an important dimension to discovery, but does not replace metadata. Users looking for articles about “student evaluation of teachers” versus “teacher evaluation of students” will benefit from the Education Resources Information Center (ERIC) subject headings, still present in discovery tools. However, users looking for esoteric information, such as the names of less-famous people, events, or works of art, will be glad if the discovery tool has indexed the full text. While full-text searching increases the number of results that are returned, indexing ensures higher relevance for those returned results. Therefore the user’s ability to discover relevant collections depends on the level of indexing.
Figure 21.2
Example of MARC field to Discovery Metadata Field Mapping
A traditional role of indexers and abstracters was to select relevant, high-quality content for a disciplinary field; in a sense, they defined some boundaries for a subject collection. This was often accomplished by marshaling content experts or a professional organization’s members to review the quality of sources included (Benjamin Jr. and Vandenbos 2006, 951; ERIC 2012). Today’s discovery tools offer no equivalent way to provide this focus. For interdisciplinary topics discovery is a boon, but for those looking for a specific field’s perspective on a topic or wishing to limit to the core sources in a discipline, a discovery tool is of little help. Vendors attempt to use journal subject headings or other groupings of journals, such as Columbia’s Hierarchical Interface to Library of Congress Classification (HILCC), the Serials Solutions Knowledgebase, and Ulrich’s Periodical Directory, to provide discipline-scoped searching (Serials Solutions 2012b).1 EBSCO released a similar feature in beta. In theory, this will improve the user’s ability to search within or across discipline-specific literature with ease. This type of system was actually envisioned by information scholars as early as 1982 (Lancaster and Neway 1982).
Librarians have several roles to play in addressing the collection management challenges created by the conflation of information. First, they need to continue to advocate for preserving accurate, complete description, and indexing in article databases and within the catalog. For example, the table of contents will be increasingly useful in discovery environments, where books are searched alongside articles. Collection managers may need to explore catalog record enrichment services, such as adding tables of contents from older book titles to the catalog, in order to enhance their discoverability. New techniques such as using linked data for metadata and sharing metadata are exciting developments (Coyle 2012). Linked data establishes descriptive relationships among the data, which supports more intelligent connections between systems. Librarians can also support developments that facilitate metadata browsing and hierarchical navigation of subjects (e.g., Bland and Stoffan 2008).
Second, collection managers should continue to encourage discovery tool vendors to leverage discipline-specific metadata and disciplinary journal collections in order to allow different views of the collection within the discovery tool. They will need to compare whether discipline-scoped searches are equivalent to those in a subject-specific index, and they will need to defend decisions related to discovery. Another challenge will be to define what non-subject-specific discovery means at each institution. For example, should libraries include online reference sources in a general discovery search alongside large, multidisciplinary journal databases? If the answer is yes, do the technical capabilities within discovery tools support it? Librarians may also need to weigh the discovery of online reference sources when making purchasing decisions.
Librarians need to test information retrieval of discovery tools for content in different subject areas and of different source types. For example, when JMU catalog records were first loaded into EDS, not all the MARC 740 subfields were included, meaning “Beethoven, Ludwig van, 1770–1827, Sonatas piano no. 23, op. 57, F minor” was reduced to “Beethoven, Ludwig van, 1770–1827, Sonatas.” This is a significant issue because access to resources the library purchased decreased due to limitations of the search. Also, initially the field was not hyperlinked, whereas now the composer is linked to perform a phrase-indexed author search and the title is hyperlinked to perform the same search limited by the title. Librarians need to communicate with vendors what the mapping of MARC fields should be, and how hyperlinks should be created. Since identifying source types challenges users, librarians should also test how the source type facet limits include or exclude results. If there is a government documents source type, does it include both online and print government documents? Does it include government documents cataloged by the library, or just the ones the discovery tool vendor has added to its main index? These details will influence the way in which a library’s collection is accessed in discovery space.
Finally, librarians should continue to promote discipline-specific search tools, and present the discovery tool as a way to cross-search a subset of the library’s collections. It is unlikely that discovery tools will ever provide a good discipline-specific search because they are designed to facilitate interdisciplinary searching. Bibliometric studies show that users construct fundamentally different keyword search strategies depending on the discipline. Yi et al. (2006, 1218) found that users of history databases used nonconceptual multiword search strings describing regions, people, and events, while psychology database users employed conceptual terms indicating disorders, abuse, development, and therapy. While discovery tools could provide convenient pathways into sub-interfaces that reproduce the discipline-specific tools, it is difficult to imagine how the molecular searching of SciFinder Scholar would be incorporated into the same search interface as the time period headings of Historical Abstracts in a way that would support usability principles. In this case, individual subject collection development policies could articulate whether databases should be included or excluded from discovery tools.
Looking farther into the future, it is likely that journal article metadata creation will become more standardized and streamlined. Currently, vendor-provided records for books and eBooks have some problems. One study found that 27 percent of records from Yankee Book Peddler and PromptCat exhibited errors affecting catalog access, and records for 39 percent of titles needed modification (Walker and Kulczak 2007, 61). In response, libraries have developed in-house cleanup solutions (Jay, Simpson, and Smith 2009; Sanchez et al. 2006) and workflows for large batch loads (Martin and Mundle 2010). A question asked by Sanchez et al. is still outstanding:
with the proliferation of eBook sources that use very basic cataloging or none at all, we will face larger issues of how, or if, we can continue to provide consistent, quality cataloging and authority control for these titles. If some entity does not provide cataloging for the universe of eBooks, will other methods such as basic Internet search engines be sufficient to provide access? (2006, 69)
Martin and Mundle (2010, 235) concluded that libraries, vendors, and eBook record services all have roles to play in ensuring quality metadata for eBooks. It is less clear what will happen with other formats, although history suggests they will continue to defy simple integration with article and book searching. Given these limitations, it seems that libraries will need to continue to provide access and discovery through multiple venues to ensure that the many facets of a library’s collection can be seen in virtual space.
Even if the challenges of metadata creation and integration were solved, the way discovery tools combine different information objects together brings its own challenges. In the physical world, it is easy to see the difference between a newspaper and a book; clearly the newspaper article will be brief, while the book offers an in-depth treatment of a topic. Determining the source type from search results lists has long confounded users (Wrubel and Schmidt 2007, 300; Jung et al. 2008, 387; Alling and Naismith 2007, 203–4; Ponsford and vanDuinkerken 2007, 171), but discovery brings this issue to the forefront (Williams and Foster 2011, 187–88; Fagan et al. 2012). Indexes and abstracts were historically dominated by periodical articles, with an odd book or book chapter here or there, while catalogs were largely monographic, with the occasional technical report. Information professionals knew that complex, well-thought-out search syntax was necessary for complete retrieval within an A&I (Younger and Boddy 2009), whereas simple searches to identify subject headings often worked better in a catalog (Drabenstott and Vizine-Goetz 1994, 266–67).
With discovery, all source types may be present or dominant in a particular result set. A simple search may retrieve both books and articles, and it may be difficult for naïve users to understand how refining a search excludes or includes certain information objects simply by their breadth of coverage. For example, when searching a broad topic where books would be more useful for providing background information, such as a search on “chemistry teaching” in JMU’s EDS retrieves a 167-page book for the first result, but an academic journal article focused on Jordanian teachers as the second. While librarians will understand that limiting to books will find more items with broad treatment, students may not immediately grasp this relationship. Discovery tools do not yet have affordances to permit easy interpretation of the variety of sources in library collections.
Compounding the difficulty of identifying the material type of individual results are the varying pathways to the full manifestations of the item and the confusion these disparate paths may cause. For some records, the full manifestation is a physical book unavailable in electronic format; for others, it is an online eBook, e-journal article, streaming audio file, or even a Web site. Other records will be for books the library does not own or for eBooks to which the library does not have access. The decision whether to include resources that index materials not available locally is related to whether the priority is to point users to the most relevant resources, even those that require a wait time, or to point users to resources that are available immediately. Additionally, in a discovery tool, a user looking for a specific book may wind up confused because book reviews for the sought title appear higher on the results list than the record for the book itself.
eBooks are also introducing complexity to discovery tools. A library’s eBook holdings are often on different platforms with different access restrictions and technical requirements, but are presented to the user in one interface through the discovery tool. Individual libraries may have less direct control over how these eBooks appear in discovery tools than they did within their catalog (Hartman 2012). These problems are not new with discovery, but again, discovery may increase the likelihood of confusion due to the sheer size of results sets.
Reference resources present a larger problem. While the problem of limiting a discovery tool to items from electronic reference collections seems solvable, including print reference, items in results seem more challenging. Reference resources were already challenging to find in library catalogs, leading librarians to create canned search strings (such as encyclopedia or dictionary or handbook or gazetteer) to combine with additional search terms. Even if an effective source type existed to limit to reference materials, they have a few additional problems. With the catalog and online reference databases combined, a single reference source in a results list may actually be a multivolume encyclopedia, while another may be a single encyclopedia entry. This makes it harder for the user to understand what is meant by reference information. Reference materials become more confusing to users when mixed in with journal articles and books, because they fulfill such a different function. Even scholars who understand the use of reference sources in providing background information may find them annoying when sprinkled in results lists. Depending on the mission of the discovery tool, institutions may decide to rout users to reference resources through mechanisms other than the discovery tool.
Audio and video items have similar problems to other formats in discovery tools because of the way their source types are tagged for retrieval. An additional issue with such items is that the label “full text” does not apply to audio or video—this has been a historical challenge in other interfaces, but is amplified in the discovery tool because streaming audio and video are full-text limiter options. Music librarians have been especially proactive in describing their needs for discovery (Music Library Association’s Emerging Technologies and Services Committee 2012; Music OCLC Users Group 2012).
Electronic resource (ER) records have always been an enigma to users. ER records describe databases, journal collections, eBook collections, or other aggregate resources, providing details about dates of coverage, scope of collection, descriptions, and terms of use. Most users want to search within a database or read journal articles—they do not care about the container in which the resource is marketed and delivered. Although there are a few users who want to know the dates covered by PsycINFO or the previous title of a given journal, most will only be using ER records to get into the product itself and perform a search. So, the fact that ER records appear in discovery tool results may be more confusing than ever before. However, excluding them from discovery may mean that a scholar just trying to find the institution’s access point to a specific database is led to believe the library does not have access. One study of a library’s Web site search showed 17 percent of searches were for database names (Fagan 2002, 248).
To overcome the challenges differing information source types raise, librarians and discovery tool vendors will need to work together. First, discovery tool vendors must strive to provide effective, accurate source type limits, including definitions of what these source types include. Of particular value would be the ability to limit results sets to just full text, to available eBooks, and to catalog items with an available status. Users should also be able to include or exclude eBooks. While most vendors offer some of these limits, they could be made more intuitive for users and accurate for libraries. Second, library instructors will need to continue to grapple with explaining the nature of different information sources to users, particularly as these sources evolve. Although once it may have been convenient for librarians and users to equate catalog searching with book-length items, and A&I searching with article-length treatments, no such dichotomy exists within a discovery product. Third, vendors should work with librarians to create clusters of resources that mirror the subject-specific databases discussed earlier, or to lead researchers to appropriate resources based upon search terms. One intriguing idea is the database recommender currently in place in Serials Solutions’ Summon (Figure 21.3), which employs the user’s search terms to find possible subject-specific databases for searching (http://www.serialssolutions.com/en/news/detail/summon-service-debuts-database-recommender). Another tactic for leading users to disciplinary-focused results could be to offer facets for journal subject classifications or subject-based electronic resource collections.
Finally, integrating reference collections into discovery is sorely needed. Microsoft Live Academic provides one model for how this could be done, by providing a definitions page for popular keywords with author, conferences, and journal facets (e.g., http://academic.research.microsoft.com/Keyword/18565/human-computer-interaction). Some libraries have chosen to use a federated search connection to add reference book results to their discovery tool (Paratext 2012). The paradigm of finding a Wikipedia entry at the top of search engine results could be a useful model to follow, by including relevant links to scholarly reference sources in the library’s collection at the top of search results.
Figure 21.3
Serial Solutions’ Summon Database Recommender Tool (June 2013). Courtesy of Drexel University Libraries
While the combination of records with different structures creates one cluster of problems in discovery tools, determining what content should be included in the discovery tool is equally challenging. Discovery tools will impact collection development and collection management practices in numerous ways, some philosophical and others practical. At JMU, a checklist has been developed to walk several stakeholder groups through the process of determining whether a given metadata source (e.g., a database) should be enabled in the discovery tool (Figure 21.4).
Some discovery tools have options to activate specific collections. Therefore, the first step in the checklist is determining whether the content of the resource is appropriate given the discovery tool’s purpose. The larger philosophical question that lingers is whether everything that could be activated within a discovery tool should be. For example, given the issues discussed earlier, should ER records be included? If the native interface for a resource is far superior for limiting and searching, should users also find that content within a discovery tool? Librarians will need to grapple with whether the discovery tool will become an all-encompassing index of the library’s collections or merely a showcase of select areas that the discovery tool can offer as a first step in the research process. Depending on the purpose of discovery implementation, the choices related to discovery tool setup will differ. If the library decides that the discovery tool is merely a first step in the research process, users should be directed to other library resources that are subject specific to continue with in-depth research. Ultimately, implementation decisions need to be made with the end user in mind. Depending on the library environment, a variety of internal stakeholders may also need to be consulted. In addition to the collection management librarian, subject librarians should look at new content additions in their discipline. For example, if MLA International Bibliography records are added, the English librarian may best know whether the records are of similar quality and utility to those in the native interface. For some disciplines, it may be very important that the full text be indexed (as mentioned earlier, for art); for other disciplines journal metadata may suffice.
Figure 21.4
Checklist: Enabling New Content in EDS
This content question relates to technical capabilities as well. If distinctive features of a source are not supported by the discovery tool (e.g., chemical structure searching or financial data viewing tools), should the records even be added to the tool? Other issues include the percentage of foreign-language items in the database under consideration and the proportion of records for which library-subscribed full-text is available. Additionally, databases that have records for books (e.g., Historical Abstracts) may duplicate catalog records and provide no clear path to get the book. These types of content decisions could be made on a case-by-case basis or could be made with an overarching policy. Each library will need to determine whether it is better to load as much content as possible into the discovery tool or whether a more selective approach to source inclusion is appropriate.
Even when the content is appropriate, there may be technical reasons why a certain collection should not be included, which is the reason for the length of step three in the checklist. All database content does not display equally within a discovery tool. As mentioned previously, records may vary based on content, which affects relevance ranking and discoverability. Database vendors other than the discovery tool vendor may place restrictions on how the information displays within the discovery tool. For example, when Alexander Street Press (ASP) initially offered metadata from their databases through discovery tools, libraries could either add all or none of ASP’s content, regardless of whether the institution had access to all of the databases (personal communication, EBSCOhost January 23, 2012). Thus, unless a library subscribed to all ASP collections, the end user would not have access to full text for many of the records that would display. If a collection causes problems in the discovery tool at JMU, the vendor is contacted and the records are suppressed or removed until the records meet the requirements. The extent to which a collection supports a discovery tool may become a significant factor in collections decisions.
It is inevitable that some content will never be available through a discovery tool. While a discovery tool has the potential to highlight a library’s hidden collections, it also may succeed at hiding valuable resources from the user. Evidence is mixed as to whether users believe the entire library’s collections are searched by discovery-type tools (Fagan et al. 2012, 91; OCLC 2011), and their perceptions could change over time. Users’ understanding of what is searched by any particular search box is limited (Vaughn and Callicott 2003, 14–15), although not completely naïve. When asked what users thought the discovery tool covered at JMU, seven of eight participants responded that they thought that it covered most resources (Fagan et al. 2012, 91). Because it is unlikely that discovery tools will ever search all library resources, it is important to clarify both what is included and what has been excluded from the discovery tool. For this reason, JMU keeps careful track of which collections are enabled or not (see step six in the checklist) and has also created a public frequently asked question with a bigger-picture view (Figure 21.5).
Figure 21.5
Diagram on the Library’s Public Web Site Explaining What the Discovery Tool, Quick Search, Contains. See also http://www.lib.jmu.edu/info/faqsitem.aspx?id=2139. Credit: David Gaines
There is also a need to highlight subject databases or niche collections that are either not covered by the discovery tool or that have more robust search limits in their native interface. A library will need to ensure that there are several escape hatches in place within a discovery tool interface that will allow the user to return to the library homepage. While users can and do use the back button to return to the library homepage, clear pathways returning to the library homepage and subject databases should be present. In a recent usability study, users were asked to use a specific database to locate a resource. Although most users were able to complete the task, many started their search by typing the database name into the discovery tool search box, which was unsuccessful (Fagan et al. 2012, 97). Given this tendency, there could be an argument by some libraries for including ER records in discovery tool indexes.
Another area that has not been covered as thoroughly in the literature is how collection policies will need to adjust in the age of discovery. Peggy Johnson indicated that collection policies “serve as plans for building and maintaining a collection, both locally held and accessed remotely” (Johnson 2009, 72–73). While institutional collection policies should cover approaches to collections in broad strokes rather than specifics, it seems that libraries should develop a general philosophy of access to collections, and one means of access is through discovery. Collection policies should indicate the type of content acquired and may include information on shared collection access. Often, they also include information related to ILL—another means of access—but few have indicated specifics related to search or discovery. Policies serve to inform users about what content is included in the collection as well as the scope of collection in order to provide guidance for what is accepted and what is not accepted as part of the collection (Anderson 1996.) While collection policies could remain mute on the discovery issue because policies deal with specific tools rather than content, library collection teams should carefully consider the effects that discovery tools are having on collection decisions. For example, while having a policy stating that collection decisions are made with no consideration of the discovery tool may be a disservice to users, so will having a policy that states that every collection decision is linked to the discovery tool. Clearly delineating the role of the discovery tool in collection building and management will allow institutions to articulate where they land on this decision spectrum. Given the prominent location of discovery tool software on many library homepages, it would behoove libraries to acknowledge its role as a first stop for most users, and therefore the tool should be addressed as a means of access in collection policies.
Library collections are changing in a variety of ways. As more information is available electronically and the amount of information available to users grows exponentially, access to library content in the easiest way possible for end users and should be of the highest importance. With the advent of discovery tools, general indexing and abstracting databases that do not offer full text may no longer be relevant. While specialty and subject-specific databases continue to see high use, general resource use has often decreased following implementation of a discovery tool (Way 2010; Fagan and Mandernach 2011, Figure 11). Therefore, libraries could rely on the discovery tool to provide access to general resources, which would allow more money to be funneled toward acquiring content and providing pathways to niche resources. However, full-text overlap analysis will be necessary to be sure access to important indexing and full text is not lost along with general search engine subscriptions (Fagan and Mandernach 2011, 17).
Finally, discovery tools may make an impact on other collection decisions. These broader decisions may evolve over time. All discovery tool vendors provide other databases on their platforms. An emerging choice for collection managers may be to wrestle with how they provide access to databases on multiple platforms. For example, if a library has a discovery tool with a particular vendor such as EBSCO, should the library favor the EBSCO platform for databases available on multiple platforms (e.g., PsycINFO) due to discovery? Also, given that records can display differently in the native interface than in the discovery tool, the collections committee may come to the conclusion that resources will be acquired only if they are able to meet the core display requirements within the discovery tool. Therefore display requirements on a third-party vendor may impact new resource acquisition decisions. Collection committees will need to wrestle with the role of discovery on acquisition of content. In the past these decisions were primarily made due to economic influences; now, discovery of content may play a more important role in selection.
Usage statistics are often a key factor in collection decisions, particularly as institutions shift from just-in-case to just-in-time collection strategies. Therefore it will be important to consider the impact that discovery tools have on statistics. Discovery tools may report statistics for all databases that are included in the knowledge base for each search performed, thereby increasing search and session reports for all resources. For example, in EDS, when a user searches for the phrase “medical malpractices” within a discovery system, it will be logged as a search both within the discovery tool and in all the databases that were activated within the discovery tool at that time, whether related to the topic or not. While federated search products had a similar, inflationary impact on usage statistics, each database was searched separately and the results were aggregated, therefore reporting individual search statistics for each database made slightly more sense. While COUNTER 4 protocols switch the focus of usage reporting from searches to downloads, discovery tools will continue to elevate usage.
Interlibrary loan (ILL) is also a tool many libraries use for collection development and management purposes, because content that is frequently requested is often acquired for the local collection, and discovery tools can dramatically increase requests for materials from other institutions due to heightened discovery of unowned content. And, when a discovery tool is coupled with a link resolver, requesting content through ILL is made even simpler. At The Ohio State University, when WorldCat Local replaced the library catalog as the default discovery interface on the library homepage, ILL saw an 81 percent increase in the number of requests when compared to the previous fiscal year (Brian Miller, personal communication). Decisions to set limits on undergraduate ILL requests might be one result, in order to relieve an overburdened ILL unit. Or, another solution might be not to activate a particular product or collection in discovery. For example, if a specific resource in discovery generates too many cancelled ILL requests, it might be helpful to exclude this resource from the tool.
Discovery tool vendors are aggressively developing discovery software to stay competitive in the information marketplace. In the short term, vendors will work to address specific problems with ingesting and indexing information from the currently available metadata. In the long term, as the semantic Web grows and information publishers and libraries adapt their metadata schema, the information sources to be ingested will themselves change, and vendors will be able to re-think the potential for integration.
Libraries should keep selecting discovery tool software carefully in order to encourage market developments that will serve end users. Usability tests and user studies will be important for understanding these complex tools. Measuring the impact of changes is also important for understanding changes in the usage of content and providing libraries feedback on collection decisions. All librarians will need to keep an open mind as they consider solutions to the inevitable challenges to come.
1 As of this writing, this feature is available only through Serials Solutions widgets. A company representative explained that this was purposeful, “so we can allow our customers to work with the functionality in a controlled environment. Discipline faceting is something that we want to incorporate in the UI (user interface) after we gain feedback on how librarians are working with it” (personal communication with Maryellen Sims, Serials Solutions).
Alexander, Harriet. 2001. “Searching the MLA International Bibliography.” Reference & User Services Quarterly 40 (3): 228.
Alling, Emily, and Rachael Naismith. 2007. “Protocol Analysis of a Federated Search Tool: Designing for Users.” Internet Reference Services Quarterly 12 (1/2): 195–201.
Anderson, Joanne S., ed. 1996. Guide for Written Collection Policy Statements. 2nd ed. Chicago: American Library Association.
Benjamin Jr., Ludy T., and Gary R. Vandenbos. 2006. “The Window on Psychology’s Literature.” American Psychologist 61 (9): 941–54.
Bland, Robert N., and Mark A. Stoffan. 2008. “Returning Classification to the Catalog.” Information Technology & Libraries 27 (3): 55–60.
Cervone, Frank. 2005. “What We’ve Learned from Doing Usability Testing on OpenURL Resolvers and Federated Search Engines.” Computers in Libraries 25 (9):10–14.
Coyle, Karen. 2012. “Linked Data Tools: Connecting on the Web.” Library Technology Reports 48 (4).
Drabenstott, Karen Markey, and Diane Vizine-Goetz. 2010. “Choosing among Existing Subject Search Approaches as Responses to User Queries.” In Using Subject Headings for Online Retrieval: Theory, Practice, and Potential, ed. Diane Vizine-Goetz (Library and Information Science, Volume 94), 266–300. Bingley, UK: Emerald Group Publishing Limited.
ERIC. 2012. Content Experts. Available at: http://eric.ed.gov/ERICWebPortal/resources/html/about/about_eric_experts.html. Accessed June 22, 2012.
Fagan, Jody Condit. 2002. “Use of an Academic Library Web Site Search Engine.” Reference & User Services Quarterly 41 (3): 244.
Fagan, Jody Condit, and Meris A. Mandernach. 2011. “Discovery by the Numbers: An Examination of the Impact of a Discovery Tool through Usage Statistics.” Paper presented at the Charleston Conference, Charleston, SC. Available at: http://www.katina.info/conference/archives/2011/2011_FullProgram_FINAL.pdf. Accessed October 25, 2013.
Fagan, Jody Condit, Meris Mandernach, Carl S. Nelson, Jonathan R. Paulo, and Grover Saunders. 2012. “Usability Test Results for a Discovery Tool in an Academic Library.” Information Technology & Libraries 31 (1): 83–112.
Gross, Tina, and Arlene G. Taylor. 2005. “What Have We Got to Lose? The Effect of Controlled Vocabulary on Keyword Searching Results.” College and Research Libraries 66 (3): 212–30.
Hartman, Robin. 2012. “Life in the Cloud: A WorldShare Management Services Case Study.” Journal of Web Librarianship 6 (3): 186–85.
Jay, Michael, Betsy Simpson, and Doug Smith. 2009. “CatQC and Shelf-Ready Material: Speeding Collections to Users While Preserving Data Quality.” Information Technology & Libraries 28 (1): 41–48.
Johnson, Peggy. 2009. Fundamentals of Collection Development and Management. Chicago: American Library Association: 66–87.
Jung, Seikyung, Jonathan L. Herlocker, Janet Webster, Margaret Mellinger, and Jeremy Frumkin. 2008. “LibraryFind: System Design and Usability Testing of Academic Metasearch System.” Journal of the American Society for Information Science & Technology 59 (3): 375–89.
Lancaster, F. W., and Julie M. Neway. 1982. “The Future of Indexing and Abstracting Services.” Journal of the American Society for Information Science 33 (3): 183–89.
Martin, Kristin E., and Kavita Mundle. 2010. “Notes on Operations: Cataloging E-Books and Vendor Records: A Case Study at the University of Illinois at Chicago.” Library Resources & Technical Services 54 (4): 227–37.
Music Library Association’s Emerging Technologies and Services Committee. 2012. “Music Discovery Requirements Draft 2.” Last modified February 9, 2012. Available at: http://personal.ecu.edu/newcomern/musicdiscoveryrequirementsfeb92012.pdf. Accessed July 7, 2012.
Music OCLC Users Group. 2012. “Public Services Information and Links.” Available at: http://www.musicoclcusers.org/reftools.html. Accessed July 7, 2012,
Nagy, Andrew. 2012. What’s New and Upcoming with the Summon Service—May 2012. Available at: http://www.serialssolutions.com/en/webinars/recorded/. Accessed October 25, 2013.
OCLC. 2011. “Some Findings from WorldCat Local Usability Tests Prepared for ALA Annual.” Available at: http://www.oclc.org/worldcatlocal/about/213941usf_some_findings_about_
worldcat_local.pdf. Accessed March 11, 2011.
Paratext. 2012. “Discovery Services and the Specialized Reference Collection: College of St. Scholastica Case Study.” Available at: http://www.paratext.com/pdf/scholasticaEDS.pdf. Accessed June 22, 2012.
Ponsford, Bennett Claire, and Wyoma vanDuinkerken. 2007. “User Expectations in the Time of Google: Usability Testing of Federated Searching.” Internet Reference Services Quarterly 12 (1/2): 159–78.
Randall, Sara. 2006. “Federated Searching and Usability Testing: Building the Perfect Beast.” Serials Review 32 (3): 181–82.
Sanchez, Elaine, Leslie Fatout, Aleene Howser, and Charles Vance. 2006. “Cleanup of NetLibrary Cataloging Records: A Methodical Front-End Process.” Technical Services Quarterly 23 (4): 51–71.
Serials Solutions. 2012a. “Summon Key Databases and Packages—Full-Text.” Available at: http://www.serialssolutions.com/en/resources/detail/summon-key-databases-and-packages-full-text. Accessed June 22, 2012.
Serials Solutions. 2012b.“Introducing Discipline Scoped Searching and Customizable Search Widgets for the Summon® Service. 19 January 2012.” Available at: http://www.serialssolutions.com/en/words/detail/Introducing-Discipline-Scoped-Searching-and-Customizable-Search-Widgets/. Accessed July 7, 2012.
Vaughn, Debbie, and Burton Callicott. 2003. “Broccoli Librarianship and Google-Bred Patrons, or What’s Wrong with Usability Testing?” College & Undergraduate Libraries 10 (2): 1–18.
Walker, Mary, and Deb Kulczak. 2007. “Shelf-Ready Books Using PromptCat and YBP: Issues to Consider (An Analysis of Errors at the University of Arkansas).” Library Collections, Acquisitions, & Technical Services 31 (2): 61–84.
Way, Doug. 2010. “The Impact of Web-Scale Discovery on the Use of a Library Collection.” Serials Review 36 (4): 214–20.
Weiner, Sharon A. 2009. “Tale of Two Databases: The History of Federally Funded Information Systems for Education and Medicine.” Government Information Quarterly 26 (3): 450–58.
Williams, Sarah C., and Anita K. Foster. 2011. “Promise Fulfilled? An EBSCO Discovery Service Usability Study.” Journal of Web Librarianship 5 (3): 179–98.
Wrubel, Laura, and Kari Schmidt. 2007. “Usability Testing of a Metasearch Interface: A Case Study.” College & Research Libraries 68 (4): 292–311.
Yi, Kwan, Jamshid Beheshti, Charles Cole, John E. Leide, and Andrew Large. 2006. “User Search Behavior of Domain-Specific Information Retrieval Systems: An Analysis of the Query Logs from PsycINFO and ABC-CLIO’s Historical Abstracts/America: History and Life.” Journal of the American Society for Information Science & Technology 57 (9): 1208–20.
Younger, Paula, and Kate Boddy. 2009. “When Is a Search Not a Search? A Comparison of Searching the AMED Complementary Health Database via EBSCOhost, OVID and DIALOG.” Health Information & Libraries Journal 26 (2): 126–35.
Without having read them yet, I suspect that many of the projects and programs described in chapters in this book would not be possible if libraries were not consortially or cooperatively minded. Just scanning the table of contents, topics such as floating collections and shared print storage suggest that libraries view consortia as one of the vehicles for accomplishing their collection management goals. Libraries have leveraged participation in consortia as an effective means to save money, save time, mitigate risk, and co-invest to extend resources and services that would be inefficient or impossible for a single library to accomplish alone.
In her article, “Assessing the Value of Academic Library Consortia,” Faye Chadwell provides a list of consortial membership benefits gathered by reviewing 48 library consortia Web sites (Chadwell 2011, 649). The benefits are listed as follows:
• shared purchasing of digital content;
• sharing library materials through interlibrary loan or courier services;
• shared catalogs;
• advocacy; and
• expanded access to content not held by a member library.
The list reflects at least a century of collaborative activity as libraries have, by necessity, always needed to be part of some type of resource sharing collective in order to provide researchers and learners with material that was never collected locally. The role of consortia, especially consortia that have a central staff and budget to support member operations and programs, for the most part have adapted successfully to the changing needs of their membership and to embrace the opportunities of collaboration brought about by networking and the digital environment.
The consortia that exist today would likely have no argument with the five benefits listed earlier, whether any individual consortium provides support for all of the listed services. But cooperative collections activities in particular have become much more nuanced and complicated in the digital age. In short, an entire list of consortial benefits around cooperative collections activities currently could include digitization, open access, licensing terms and rights, new business models, digital preservation, print storage, scholarly communication, and so forth. And in many cases, these new and emerging collaborations coexist alongside all of the traditional support services for shared print collections.
In his 2007 article, “The Cooperative Conundrum in the Digital Age,” which was in part a result of discussions from the 2005 Janus Conference on Research Library Collections, Dan Hazen reviews the wins and losses from long-held cooperative print collection efforts. He follows it though with an expression of the opportunity presented to collaborate on collections in the digital age. “We have the means to cooperatively create and structure more encompassing digital collections, and our economic constraints reinforce this approach. It’s time to act” (Hazen 2007, 109). The rest of this chapter addresses whether indeed we have risen to the challenge from Hazen to act, to bring library partners together not only to meet their collection goals but also to proactively shape their collections futures.
Library consortia have a rich and long history in the United States dating back to the 1890s (Alexander 1999, 2). For the most part, consortia were organized around geographically proximate libraries in order to accomplish common goals; goals that have changed over time from predominantly resource sharing to addressing the advent of automation and the explosion of digital content. Just as the services provided by consortia have changed, so has the landscape of organizations formed by libraries to support their collective goals.
According to the publicly available International Coalition of Library Consortia (ICOLC) Web site, there are currently 63 library consortia in the United States. It would seem libraries can take their pick of collaborative organizations in which to participate and to go to for collection services. And libraries do have multiple allegiances when it comes to participating in consortial collection services. As an example, a library in the Committee on Institutional Cooperation (CIC) consortium will acquire digitized commercial publisher or vendor content through a centralized purchasing program at the CIC; license databases via their statewide consortium; become an affiliate member of a distant consortium in order to take advantage of group licensing discounts; participate in the Center for Research Libraries Shared Purchase Program; and join national deals offered by large regional consortia such as Lyrasis.
Libraries are free to shop and compare when it comes to electronic licensing deals, since there are consortia who will allow any library to join a deal as long as it pays a service charge to support the operations of the consortium office. Consortial buying clubs are a phenomenon created when libraries that had no other form of collaborative relationship banded together to take advantage of discounts offered by vendors based on a number of participating libraries or aggregating a large spend commitment. Other consortia, like the Triangle Research Libraries Network, Orbis-Cascade Alliance, and California Digital Library (CDL) craft collections services that are specific to their membership and seek to accomplish goals beyond achieving savings on products with a much broader programmatic agenda.
In a 2007 survey of 92 libraries (from 18 states and 9 countries), respondents were asked to denote the determining factor for participating in a consortium. “Access to content” was the highest rated benefit (77%), followed by “lower annual price increases” (58%) and “predictable annual prices” (43%) (Clement 2007, 196). Because consortia are about as alike as snowflakes, educating the vendor community about the goals, strategies, and makeup of any given consortium can be challenging, especially for newcomers into the marketplace that had expected they would market their products direct to libraries. The fact that libraries participate in multiple consortia that may be negotiating for the same products only makes the picture fuzzier. Maintaining robust communication and a level of trust among libraries, consortium staff, and content providers is critical to achieving successful negotiations.
Efforts to determine if there are consortia that are more successful than the rest at keeping prices and inflation low on renewing products have proven inconclusive. In 2009, economist Ted Bergstrom used state open records laws to acquire publisher contracts from 36 institutions. His goal is, “As economists . . . we find that the academic journal industry presents a fascinating case study of such practices as price discrimination, bundled sales, and long-term contracting in an imperfectly competitive industry” (Bergstrom 2010). While Bergstrom reports that his team found a great deal of difference in what libraries of similar size pay to publishers, librarians more familiar with the details of these negotiations know that many factors, not documented, go into determining base pricing and inflation for consortial deals. In short, libraries are the final decision makers when it comes to selecting which consortia return savings and value to them in the electronic licensing arena.
Coordinated collection activities, some of which were referenced in the previous section of this chapter, are predominantly focused around licensed digital content. There are still very tangible benefits to libraries who work together to license databases, purchase eBooks, and enter into big publisher packages for e-journals. Libraries can demonstrate cost savings, gain access to additional content they would not have acquired individually, and save staff time to negotiate and process licenses. New players have entered the consortial licensing sphere with both the ARL and the CRL hiring staff to develop offerings for their memberships. Clearly, there continues to be demand enough from U.S. libraries to support dozens of consortia acting in this space. As long as each consortium has a clearly defined set of goals and the marketplace sustains the development of multiple and sometimes competing offers, this will continue to be the landscape for libraries seeking content partnerships.
Libraries also have an interest in converting their own paper and microformat collections into digital content. Consortia have successfully come together to get grants to digitize special collections material. As examples, the Greater Western Library Alliance (GWLA) Western Waters Digital Library Project draws on collections from 17 consortial members, and additional nonconsortium members, to build and offer a freely available research resource that would not have been possible before. The Lyrasis Mass Digitization Collaborative offers members the opportunity to have materials digitized for a low fee, taking advantage of bulk pricing for digitization services. There are many other examples of grant-funded digitization projects awarded to single institutions or a group of cooperating libraries. Libraries that work within a consortium framework for digitization projects realize lower costs for digitization, have the benefit of consortial staff to coordinate the project, and bring greater visibility to collections through shared platforms and portals for discovery.
Mass digitization projects, in partnership with large for-profit or not-for-profit entities, have also paved the way for libraries to digitize large numbers of materials. These materials, not necessarily drawn from special collections but rather from general collections, still represent a significant preservation challenge because of deteriorating paper quality and the likelihood that they will not draw the attention of traditional library funding partners such as Sloan, Institute of Museum and Library Services (IMLS), and Mellon. The now-defunct Microsoft Book Project, the Google Book Project, and the Internet Archive/Open Content Alliance can be considered forms of meta-consortia that provide investments, equipment, and platforms in which libraries can make their materials available digitally and often freely available to the world.
Print-based collection projects, sometimes referred to as cooperative collection development initiatives, are no longer simply carryovers of long-held agreements within consortia to distribute book-collecting responsibilities. Consortia are coordinating a range of activities that benefit both local libraries and the collective. The idea of the shared collection has taken on increased importance and visibility as libraries grapple with declining budgets for monographs, the high costs of storage in stacks, and low circulation of many materials. With robust resource sharing in place, libraries can reliably look to their partners not only to expand the number of titles available within a group collection, but also to deliver those titles in timely, efficient ways to users.
Consortia like OhioLink and the Orbis-Cascade Alliance have worked with book jobbers such as YBP to control the number of copies of single titles purchased for a shared collection and to examine gaps of titles that could be added to a collection. The SUNY research center libraries piloted a shared collection of university press titles that included purchases from eight university presses and housed among the four cooperating libraries. Through the services of a shared Japanese bibliographer for three campuses, the CIC libraries were not working from a shared collection model or framework. Rather, the bibliographer was working to build collections for each of the campuses that met the needs of local researchers, though the material would be available through the CIC resource sharing agreement.
In the consortial community currently, much attention is being paid both to DDA and shared print storage. There are entire chapters in this book dedicated to these topics but it would be an oversight not to acknowledge them as another facet of consortial collections activity. As with most cooperative collection projects, these two areas are being addressed in a variety of successful ways in the United States and globally, and the business models will continue to evolve.
Even though new products continue to come on the eResources market, weekly it seems, the process for dealing with new offerings, new companies, and the business of acquisitions is quite mature. EResources licensing has its roots in the late 1980s and early 1990s and there are infrastructures in consortial offices to manage and develop membership offers, negotiate license terms, and swap all of the technical information required to provide access to content. But it hardly means that this business is routine. Over the 30-year plus lifetime of eResources, a long list of issues related to licensing terms and business practices continues to be revisited and re-strategized, while new issues emerge as libraries shift their priorities and face fiscal pressures. Consortia can play a key role in bringing issues of importance to libraries to the vendor community. The ability to speak on behalf of a large group of libraries, or a group of libraries that represent a significant revenue stream to a vendor, can give a consortium the muscle to negotiate more aggressively.
The original Big Deal for publisher packages of e-journals is showing its age. Many consortia still hold licenses for these bundles of journal titles and many libraries can still justify the cost of the packages through usage and other metrics of value. A 2008 study of ROI at the University of Illinois reported that $4.38 in grant income is realized for each $1.00 invested by the university in the library (Luther 2008, 12). But for many libraries, the economics of the Big Deal has either become unsustainable or its simply that year-on-year justifications for inflation are no longer valid since the original terms of these deals were based on historic print spends from the 1990s. In the article, “Is the Big Deal Dying?,” 13 representatives from publishing, consortia, and libraries each give their opinion of the future of the big deal and possible alternatives (Boissy et al. 2012, 36). From the responses, it is clear that publishers and consortia alike have spent the past few years trying to reconceptualize a big-deal journals model that meets a new set of criteria. It is not that all libraries want to break apart their deals; rather they need some flexibility to drop titles that perform poorly, are no longer programmatically relevant, and have more options to manage their costs. Libraries and consortia have also become more aggressive about advocating for rights to deposit works by their institutional authors in local repositories. Several large STM publishers are currently running pilots of new big deals, but these have yet to seep more broadly into the rest of the community.
Under the broad umbrella of scholarly communication, consortia are now being asked to introduce language into licenses that allows for author’s rights, text mining, and international ILL. Each of these three issues has its own story and development timeline. When negotiating for author’s rights, consortia are recognizing the need to act on behalf of authors who sign away their rights to reuse their own created content at the time of publication. By turning over the copyright to publishers, faculty and researchers have also signed away the ability to make educational and scholarly uses of their content without seeking the publisher’s permission. The costs to gain permission to reuse this material often fall back to a library.
Text mining is the newcomer to the licensing world, a concept that works only in the world of digital content. While faculty working in the digital humanities have an established history of computational research across digitized textual material, content that is now housed on a publisher’s servers requires permissions to access. It is especially important that publishers can distinguish between systematic downloading that often signals piracy and legitimate uses of large blocks of content.
ILL has been at the center of many licensing tugs of war over the year. Early efforts by publishers to control the amount of ILL from e-journals was in an attempt to prevent widespread dissemination of copyrighted material made much easier by electronic distribution. More recently, a group of STM publishers sought to limit the ability of U.S. libraries to provide lending to the international community. Presumably, the confusion on the part of STM publishers was related to the difference between noncommercial lending between libraries and fee-based document delivery services. Regardless of the motivation, the need to reestablish the rights of libraries to make fair uses of material supported by Section 108 of Copyright Law has come back into focus in licensing negotiations.
As libraries continue to evolve in both their fiscal environments and the suite of services that they provide to researchers, faculty, and students, it is imperative that consortia are prepared to represent those interests at the negotiating table. While cost savings is a predominate way that consortia demonstrate value to their members, securing rights and permissions can be equally as important in order to maximize dissemination and use of scholarship.
The digital environment has created the opportunity for consortia to be less geographically based and to allow groups of dispersed but like-minded libraries to come together to accomplish mutual goals. While these organizations are not what might be considered under the traditional definition of consortia, each has a membership, members make investments in the organization, and the libraries involved have come together to accomplish a goal that could not succeed at a standalone institution. Writing on the topic of new forms of library collaboration, Paula Kaufman reinforces the need for thinking beyond traditional relationships to accomplish the new challenges faced by academic libraries, “There are compelling reasons to think that to be effective in the future the ways in which libraries work together must deepen, and the partners with whom they choose to work must broaden” (Kaufman 2012, 54). In many cases, the source of funds for these memberships and pledges of support come from the collections or materials budgets in libraries.
The HathiTrust Digital Library, founded by the CIC and CDL consortia, is housed at the University of Michigan and University of Indiana. Hathi was originally conceived as the storage mechanism for book files being returned from the Google Book scanning project as well as content from the Internet Archive, Microsoft, and locally digitized collections. Currently, the HathiTrust has 66 institutional members, some under the umbrella of a consortial membership. Members have the option of depositing content in Hathi or, if not depositing, “to participate in the long-term curation and management of the repository in return for enhanced services for accessing and using materials in the digital library” (http://www.hathitrust.org/). By investing together in a long-term content repository, the library community has more control over access and usage more in line with fair use rather than contractual language governing vendor agreements.
The open access movement continues to be a beacon of change in the entrenched scholarly communication environment. Although publishers are now trying to wrest and monetize open access options from grassroots movements, organizations such as SCOAP3 (Sponsoring Consortium for Open Access Publishing in Particle Physics) and arXiv are successful examples of consortia that impact collections activities at every level. arXiv is the open access repository for research articles, primarily in physics, math, computer science, and statistics currently housed at Cornell but founded by Los Alamos National Laboratory. arXiv recently restructured its business model to ensure its long-term viability with contributions from the libraries and research laboratories worldwide that represent arXiv’s heaviest institutional users. Despite the existence and heavy usage of arXiv, libraries had not typically been substituting freely available arXiv access for journal subscriptions in which the same articles are published. SCOAP3 is now working to facilitate open access publishing in high energy physics. The initiative is led by CERN (European Organization for Nuclear Physics) in Geneva and currently has pledges of support from universities and governments and labs in 30 countries. An outgrowth of arXiv, SCOAP3 is working on sustainable open access models for the peer-reviewed journals whose preprints are often deposited in arXiv.
As a consortium-employed author, it is in my organization’s best interest to always be looking for the next big opportunities to enrich and enhance our libraries’ collections and to accelerate the adoption of strategies to accomplish this collectively. Cost savings will continue to be an important metric for our success as libraries grapple with budgetary challenges. It will also be important to make strides in efficiencies for selecting, acquiring, and delivering content. Libraries have streamlined staff positions and collective action should reduce the speed of adoption for getting content to the users.
Libraries are facing requests for new kinds of content for which norms for selecting, pricing, and licensing must be established. Datasets, textbooks, and researcher profile systems are examples of new areas to tackle as a result of user demand. In each of these cases, it will require multiple layers of collaboration to manage the integration of these new forms of content into the library space. Datasets will need the support of information technology to ensure usability and long-term curation of the material. Textbooks have typically been in the purview of the bookstore and the individual student, so the transition to libraries for handling the negotiations for price and delivery must take into account preexisting revenue streams. And researcher profile systems will involve discussions among the Offices of Research, provosts, and the libraries. Consortia may also get involved in negotiations for deals that will benefit across institutions.
Libraries have also taken on the role of publisher, many now running operations with their own imprints for books and journals. As a provider of content, libraries are in a position to create business models for more favorable pricing and to give wider dissemination to content through open access. Because most of these presses are not operating at a large scale, consortia and libraries should seek to give support to such efforts. Through these projects and programs there will be a shift, however big or small, in the scholarly communication lifecycle that makes scholarship more accessible and affordable.
Partnerships among libraries, consortia, and commercial information providers should take on new dimensions, whether they are longstanding relationships or first-time meetings. Using the power of the purse, libraries and consortia should provide incentives for publishers to create content for disciplines under-represented in digital form. As large-scale shared print initiatives continue to develop, libraries and publishers should consider how these collections could be converted to digital form without the concomitant price tag that often goes along with format migration for the same content.
The long-term and ongoing support for consortial activity in the area of collections indicates that libraries find value in working together to save money, extend access to content, co-invest in digitization, and share in the exploration of new models for creating, managing, and sharing content. The challenge faced by both libraries and consortia is to rightsize the activities and investments made at the local and consortial levels to meet the needs of users in efficient and cost-effective ways.
In a 2005 paper delivered at the Janus Conference, Mark Sandler predicted that “core resources that serve 80 percent or more of users will be selected and served up centrally; with local efforts focused on truly local needs” (Sandler 2006, 241). Whether this should be the goal of cooperative collection activities or is just a natural evolution as a result of the availability of digital content and coordinated effort is yet to play out. Perhaps it is a little of both. Collaboration can be slow and frustrating and requires libraries to yield a certain amount of autonomy and decision making, but the results can and should surpass anything a library would have been able to accomplish on its own.
Alexander, Adrian W. 1999. “Toward ‘The Perfection of Work’: Library Consortia in the Digital Age.” Journal of Library Administration 28 (2): 1–14.
Bergstrom, Ted C. 2010. “Big Deal Contract Project.” Available at: http://www.econ.ucsb.edu/~tedb/Journals/BundleContracts.html. Accessed November 1, 2012.
Boissy, Robert W. et al. 2012. “Is the ‘Big Deal’ Dying?” Serials Review 38 (2012): 36–45.
Chadwell, Faye A. 2011. “Assessing the Value of Academic Library Consortia.” Journal of Library Administration 51 (7–8): 645–61.
Clement, Susanne. 2007. “Skills for Effective Participation in Consortia: Preparing for Collaborating and Collaboration.” Collection Management 23 (1–2): 191–204.
Hazen, Dan. 2007. “The Cooperative Conundrum in the Digital Age.” Journal of Library Administration 46 (2) 101–18.
Kaufman, Paula. 2012. “Let’s Get Cozy: Evolving Collaborations in the 21st Century.” Journal of Library Administration 52 (1): 53–69.
Luther, Judy. 2008. “University Investment in the Library: What’s the Return?” Library Connect. Available at: http://libraryconnectarchive.elsevier.com/whitepapers/0108/lcwp010801.html. Accessed November 1, 2012.
Sandler, Mark. 2006. “Collection Development in the Age Day of Google.” Library Resources & Technical Services 50 (4): 239–43.
In the Oxford English Dictionary, one of the definitions for floating is “having little, or comparatively no attachment.” Another meaning is “not fixed or settled in a definite state or place.” If one looks up the same word in Thinkmap’s Visual Thesaurus, one can follow a trajectory through unfixed and free to available and lendable. When librarians decide to implement floating collections, they travel this same path as they make their materials available in new venues for patrons to borrow. What follows is a brief description and discussion about the journey two academic libraries took along the road to a floating collection—a collection in which materials borrowed from one library by patrons of another remain at the second library rather than being returned to the home library.
Denison University and Kenyon College are two small liberal arts schools located 30 miles apart. They are members of the Five Colleges of Ohio (Ohio5) consortium along with Oberlin College, Ohio Wesleyan University, and the College of Wooster. Four of the schools (Denison, Kenyon, Ohio Wesleyan, and Wooster) have shared an online catalog since 1996. All five college libraries are also members of OhioLINK, a state consortium of 88 academic libraries, including both public and private institutions, ranging from research universities to community colleges, as well as the State Library of Ohio. A statewide union catalog allows direct patron-initiated borrowing from any member library and delivery usually occurs within three to five days. OhioLINK libraries participate in a group membership in the Center for Research Libraries (CRL) and CRL-cataloged books may also be requested through the OhioLINK central catalog, though delivery time may be slightly longer than for materials housed in member libraries. The OhioLINK central catalog provides access to more than 45 million items, including 13 million unique items. With renewals, students may borrow materials for up to 21 weeks, and faculty for 42 weeks. An additional union catalog interface also permits OhioLINK patrons to request and borrow materials from SearchOhio, a consortium of 21 public library systems. Some 9 million items are available through the SearchOhio collection. These consortial arrangements make an extensive number and wide range of materials available to Denison and Kenyon patrons within a relatively short time frame. This existing network of mechanisms for patron access was considered as part of the context within a floating collection might be implemented.
The Ohio5 schools have a number of consortial committees and the libraries have worked together on projects both large and small. During a consortial grant project 10 years ago, discussions about cooperative collection development among librarians at the Five Colleges included musings about the possibilities of floating collections among the libraries.1 Materials would be shelved at the library to which they were returned, regardless of which of the five libraries actually owned the items. The same proposal has also been floated at the OhioLINK consortium level, but to date has not gained sufficient traction to move forward. One of the stumbling blocks identified during the Five Colleges’ conversations was delivery time. Without the ability to provide next-day delivery, the proposal lacked the carrot necessary to bring everyone on board—a dramatic and tangible improvement in service for our patrons. After all, the libraries and their patrons already had statewide consortium borrowing and delivery within a few days. What the collection development librarians believed was that next-day delivery, combined with a floating collection, would help erase some more of the boundaries that continued to hinder cooperative collection development. With those two pieces in place, both librarians and patrons would begin to view the unified collection as a single unit rather than five individual collections. Unfortunately, at the time, the financial costs related to an improved delivery service were deemed too high when compared to the possible benefits.
Among the Ohio5 schools, Denison and Kenyon are located most closely to one another, and that small geographic distance has made even closer cooperation between these two libraries possible. In 2004 Denison and Kenyon received another grant from the Andrew W. Mellon Foundation, this one to plan a redesign of library technical services work across the two colleges. The goal of the project was to improve access to information resources and create value-added services for their patrons (Andreadis et al. 2007). As one part of the Denison–Kenyon technical services redesign process, a daily delivery service between the two schools was initiated in order to transport cataloged and processed materials from one library to the other. Although the delivery service was not intended specifically for the purpose of facilitating floating collections, it became clear that it could serve in that capacity in addition to fulfilling its role for the technical services staff and their needs. For these two schools in the Ohio5 consortium, one of the main obstacles to a workable floating collection had been eliminated. Certainly many libraries have successfully employed floating collections without next-day delivery, but in this instance, considering the proximity of Kenyon to Denison and the preexistence of OhioLINK, 24-hour delivery was viewed as a necessity to the project’s success. In hindsight, as beneficial and convenient as it is, next-day delivery may not have been as crucial as it was once seen to be.
The two libraries did not conduct detailed surveys of other implementations of floating collections before moving forward, but it did seem clear that the idea had been undertaken more commonly in public libraries than in academic ones. This fact was perhaps daunting and exciting at the same time. By initiating a floating collection between two academic libraries, Denison and Kenyon would be moving into uncharted territory. In the end, the decision by the Kenyon and Denison library directors to move forward with a floating collection between the two schools was not based on an analysis of data. Rather, it was based on an instinctual and anecdotal opinion that the daily delivery of materials provided in the technical services redesign process created an opportunity to provide better service for their patrons by offering them immediate access to materials they have used and might need again. This rationale for implementing a floating collection is related but not identical to the reason often cited by public libraries—to reduce the time that materials spend in transit (Hilyard 2012, 16). The project also appealed to the love of experimentation that both directors possessed.
Although there seemed to be many advantages to the floating collection proposal, there was also a certain amount of apprehension about it. Five years later, it is difficult to remember the angst of that time clearly. However, the concerns were very real. Many of these same issues have been faced by other libraries undertaking this change in practice and procedure (Cress 2004, 49). One of the primary concerns was about a potential imbalance in borrowing. For example, would patrons at Denison request many more materials than patrons at Kenyon, resulting in an overabundance of materials at one location and empty shelves at another? A related issue was whether large numbers of titles in particular subject areas would be requested from one library, leaving patrons researching those topics at the other college with fewer resources immediately available. A third concern was about potential shelving issues. Neither library had an abundance of available stack space. What would happen if large numbers of books were requested in particularly tight areas of the stacks? Would difficult and time-consuming shifts of parts of the collections be required on a regular basis? Another issue was possible confusion for shelvers and other staff at both institutions. Prior to the redesign of technical services, spine-label formatting and barcode placement were not uniform at the two schools. How much training would be required for shelvers and for circulation staff? And, as other librarians who have implemented floating collection programs are aware, none of these questions can be answered in advance. It is impossible to accurately predict how much and which parts of the collection will be affected. It is possible, however, to train staff so they are aware of the technical and physical processing issues. It is also possible to maintain awareness of potential issues and to monitor implementation for potential hiccups.
For Denison and Kenyon a critical question that required investigation was whether the integrated library system (ILS) they shared with two other colleges had the capability to support a floating collection, and whether it could be implemented by only two of the four libraries using the ILS. Since Ohio Wesleyan and the College of Wooster, the other two schools sharing Denison and Kenyon’s catalog, had decided not to participate in the technical services redesign grant and were not part of the daily delivery route, they also opted not to participate in the floating collection project. In 2005, when the technical services daily delivery service began, Denison and Kenyon did not have the necessary software within their shared ILS to make a floating collection possible for only two of the four schools, although it was available from the vendor. The decision to move ahead with the project was made and that software was acquired in 2006. Floating became a reality in the summer of 2007.
To begin the process of floating collection implementation, an initial meeting of systems, circulation, and stacks management staff of both colleges occurred early in 2007. Since the academic year for both Kenyon and Denison runs from August to May, without a full academic summer semester, it was decided that beginning implementation during the summer months would be the least disruptive. Instituting new procedures at a nonpeak period of lending and borrowing activity would allow staff time to adjust. Check-in issues, shelving concerns, and training could be resolved without the constraints of the school-year endeavors. When checking in materials, circulation staff would need to be aware of whether the materials were from the floating collections partner library and thus meant to stay for shelving in that library, or if they would need to be returned to their home institution via the courier service, since borrowed materials come from many different libraries.
Both Kenyon and Denison have a number of collections (archives, multimedia, periodicals, reference, special collections, etc.) with limited, restricted, or no circulation. It was decided at the outset that only the main monograph collections would be part of the floating collection project, but at both Denison and Kenyon collections with restrictive circulation policies represent a large percentage of each overall collection. Isolating them was not an issue, however, as these collections were also easily distinguished by item and location codes. A single combination of item and location codes at each institution identified the materials that would be part of the floating collection. Among the other circulating collections, government documents and folios (or oversize materials) were excluded, at least initially. With those choices made, implementation was not overly complicated. As part of a global update, a field indicating owning library was added to the item records of all materials slated to be part of the floating collection. Once the records were updated, the software was turned on that enabled floating as part of the circulation process. With the home library clearly listed in each record, floating books could easily be identified and returned to the owning library if necessary. For staff members familiar with barcodes, materials are also easily distinguished by the differing initial character string of each institution’s barcodes. As a change to ongoing workflow, cataloging staff would need to add the owning location field when new records that fit the parameters of the floating collection were created. That field would automatically be added to batch-loaded records, which account for the bulk of items in the proposed floating collections. It was eventually decided to create an item type for materials in the general collection that would not be part of the floating collection. This item type has been used rarely, but having it available eliminated one additional cause of concern.
Neither Kenyon nor Denison tracked the floating collection at regular intervals, but periodically they created lists of what Kenyon materials are at Denison and what Denison materials are at Kenyon. Each report showed that the number of floating items at one institution was very similar to the number at the other library. The overall number of floating items has also remained relatively stable during the five years since floating was initiated. Currently, about 11,300 Kenyon items are at Denison and 11,500 Denison items are at Kenyon. This works out to about 3 percent of the materials made available for the floating collection.2 In the end there has not been an appreciable imbalance of borrowing, nor has any particular subject area been more affected than any other. This is clearly different from the experiences at a number of public libraries with floating collections, where regular rebalancing of collection materials has been necessary (Canty et al. 2012, 68). To date, neither Kenyon nor Denison has undertaken any rebalancing projects. Perhaps this variance is due to the status of both Denison and Kenyon as undergraduate, liberal arts colleges, with similar collections and patrons. It could also be attributed to the inherent differences between academic and public library collections.
As expected, one of the real benefits to users has been next-day delivery. At the reference desk, the librarian can reassure patrons that the requested item will be available to them by a definite time depending on when the request was made—either the next day, if the request is made early enough in the day, or the day after, if the request is made in the late afternoon or evening. The floating collection has not been advertised per se at either school, but librarians do mention the delivery time during instruction sessions and in one-on-one interactions. Many patrons are also aware of the service, either from personal experience or through word of mouth.
The floating collection has also assisted with ongoing consortial collection development initiatives designed to reduce unnecessary duplication. Perhaps neither librarians nor patrons think of Denison and Kenyon’s collections as a single entity rather than separate, but it is an easier sell to both librarians and faculty if one can say that the item does not need to be acquired because it can be sitting on our shelves tomorrow.
As with most projects, there have been a few unintended or unanticipated consequences. One must be more vigilant during the weeding process. Occasionally, items have been mistakenly pulled as duplicates when one copy actually belongs to Denison and the other to Kenyon. At other times, materials that fit the weeding criteria do not actually belong to the library doing the weeding, but these occasions are not so problematic as to overshadow the many benefits of this project. And, some of the common reported challenges in floating collections—shortages and overflows; lack of shelf space; and insufficient communication about redistribution (Hilyard 2012, 14)—have simply not been issues for either Denison or Kenyon.
The idea of the floating collection was initially billed as a pilot project, but after five years that label no longer applies. Anecdotally, it seemed clear early on that the project was working and formal assessment has never been deemed necessary. This decision is bolstered by the fact that no major issues developed and that the program is taken for granted by library staff. These are also signs that the pre-implementation anxiety was unnecessary and that the floating collection has been a valuable addition to the way that Denison and Kenyon provide access for patrons to materials.
Because of the different experiences reported by public library implementations, some questions remain. It would be interesting to know if other academic libraries implementing floating collections would also escape the public library overflow or shortage issues, as Kenyon and Denison have done. Would the type of academic library (research institution vs. community college) have any effect? What would be the results of floating among different types of academic libraries, as opposed to floating between two very similar institutions? The answers to these questions will have to wait for the implementation of more academic floating collections, but if the Denison–Kenyon experiment is at all indicative, this is a collection management strategy worth exploring.
1 It is a three-year grant from the Andrew W. Mellon Foundation to foster cooperative collection development. The grant ran from 2001 to 2003. Additional information about the project may be found at: http://www.ohio5.org/web/default.html.
2 Denison University acquires approximately 4,800 volumes each year with a circulating monograph collection of 290,000 volumes. Kenyon College acquires approximately 4,000 volumes per year with a circulating monograph collection of around 340,000 volumes. In the past two years, Denison has circulated an average of 36,000 volumes from its collection and Kenyon 33,000 volumes.
Andreadis, Debra K., Christopher D. Barth, Lynn Scott Cochrane, and Karen E. Greever. 2007. “Cooperative Work Redesign in Library Technical Services at Denison University and Kenyon College.” In Library Workflow Redesign: Six Case Studies, ed. Marilyn Mitchell, 39–49. Washington, DC: Council on Library and Information Resources.
Canty, Adrienne Brown, Louise C. Frolek, Richard P. Thornley, Colleen J. Andriats, Linda K. Bombak, Christalene R. Lay, and Michael Dell. 2012. “Floating Collections at Edmonton Public Library.” Evidence Based Library and Information Practice 7 (1): 65–69.
Cress, Ann. 2004. “The Latest Wave.” Library Journal (October 1, 2004): 48–50.
“floating, adj.” OED Online. Oxford University Press. Available at: [http://www.oed.com/view/Entry/71742?rskey=U8Lg4o&result=2]. Accessed December 5, 2012.
“floating, adj.” Visual Thesaurus. Thinkmap. Available at: [http://www.visualthesaurus.com]. Accessed October 5, 2012.
Hilyard, Nann Blaine. 2012. “Take the Plunge, Implementing Floating Collections in Your Library System.” Public Libraries 51 (3): 13–20.
To float a public library collection is to watch collection management Darwinism in action. The fittest items in the collection survive and float; the non-floaters scream “weed me.” Popular items spin throughout the entire library system, never returning to a specific home branch, but catching hold after hold, traveling from one branch to the next. Selections made just in case are mercilessly left standing in place, a testimony to misspent collection development funds. Floating is dramatic, immediate, and for better or worse, serves to put collection issues squarely in everyone’s lap from shelvers to the director.
Simply defined, floating collections are collections in which there is no owning location. Items may be started at a location that has a heavy interest in mystery, for example, but once that item leaves the branch at which it started, either from a patron request, or from being returned to a different branch, it stays there. It is not routed back to the starting location.
Over the past few years, floating collections have gained tremendous popularity in public libraries for several reasons. With reduced staffing and budget challenges making headlines across the country, administrators have looked to floating collections to provide instantaneous relief by cutting costs in delivery, labor (the items are handled half as often), and collection development budgets.
Because items no longer spend time traveling back to an owning location, hot items are more immediately available, causing a bump in circulation, and a savings on the collection development purchasing end. Patrons will universally exclaim that there are more new items available, when nothing could be further from the truth. The reality is that collection development has not spent any extra funds on materials whatsoever. Because delivery time is halved, the number of popular items needed can actually be reduced.. Because patrons perceive the collection as abundant and fresh, they spend more time browsing, often increasing circulation.
Much work goes in to preparing a collection to float, primarily in terms of weeding materials that have not earned their keep in terms of circulation. Again, depending on how rigorous or lax the weeding has been in the past (and there is often great variety within departments and age levels at the same branch or throughout the same library system), this is either a mind-bending all-out effort or a mere extension of good weeding practices. Once this pre-floating weeding is accomplished, the system will often get an increase in circulation simply because the stacks look better, shelvers find room for hot items languishing in the back room, and patrons can more easily find what they are looking for. It is a more pleasant browsing experience even before the floating materials reach the door.
But what is important to remember is that for every positive outcome, there can be major and permanent collection handicaps that impact the public librarian’s ability to do his or her job well and to serve patrons, and while some of that impact can be lessened, the collection is always off once floating starts. This chapter addresses best practices that librarians have developed to get the most of their floating collection while attempting to ameliorate the negative effects that floating has on collections and collection management.
A recurring theme with floating is that one solution or method or practice does not by any means fit all. Systems with far-flung branches often decide to float everything to cut down as much as possible on delivery and fuel costs. Systems with a main library with a well-respected and highly trafficked research collection often decide to float branch materials and popular new materials, leaving main library’s collection out of floating altogether. Sometimes funding plays a role. Some systems can only float those items bought with communal money, such as state funds, while being restricted in terms of where items purchased with local monies can go. In those cases, systems will frequently purchase popular movies and books and float those, keeping the rest of the locally purchased collections traditionally shelved and owned by a particular branch.
And remember the following maxim, because it is frequently forgotten in the what-ifs of planning to float. There is a direct correlation between circulation and floating. Popular items float more, no surprise. But what usually does come as a surprise is how little low-circulating collections actually float. Children’s materials and the less-popular Deweys in adult nonfiction actually take a long time to float and refresh, while movies at a popular branch will float in and out almost immediately, creating an entire new selection.
Many systems float movies, music CDs, and best sellers, but do not float the regular collection. Others float everything except children’s materials and adult nonfiction. And many systems float absolutely everything, save the odd special collection such as local history.
Determining how much and what to float is really a pretty simple decision. The more that is floated, the more money that is saved. Librarians who approach floating as a way to help a serious budget problem will want to float as much as possible, and deal with whatever collection disruptions and imbalances arise later. It is important to consider though, the hidden costs of floating the entire collection. Because the collection disruption is more intense—some branches are inundated, some lose a significant portion of their more popular materials—the time and labor costs involved in trying to even out the collection is not insignificant, and almost always involves higher paid staff. So while a savings is realized in delivery costs and at the circulation and shipping desks, branch managers and librarians spend a lot of time trying to contend with the uneven distribution of a floating collection. While the goal is never to try and restore a collection to the pre-float level, customers at the branches which lose materials will not be happy with a lesser selection, and librarians will have to try and adjust for that. Another hidden cost that the collection development department and librarians must work closely to control is the possibility of newer and highly desirable materials being weeded by the librarians at the inundated branches who are desperate to create more space for the materials that continue to float in. And finally, staff morale can be very challenged at both extremes of the floating spectrums. Librarians in branches with too much stuff feel as though they are working twice as hard as everyone else in the system just to beat back the tide. And they are right. And librarians in the branches with a net loss work twice as hard to provide something with which to please their patrons.
It is possible to have the best of both worlds. Many librarians have realized that floating brings wonderfully refreshed collections, which are highly desirable and help circulation. Librarians in that situation get the best of both worlds by floating parts of the collection, which help their budgets, while keeping large sections of their collections intact and balanced. There are as many scenarios as there are floating libraries. The advantage of only floating part of the collection initially is that the decision can be made piecemeal to float the rest, allowing for time to evaluate how disrupted the collection becomes. Clearly, a financially stressed organization (and how many libraries can claim not to be?) cannot be choosy, and floating as much as possible will help to ease that financial stress.
Whether the decision is made to float the entire collection, parts of it, or only popular materials, the more thorough the preparation of the collection prior to floating, the less negative the impact of floating will be. Weeding the collection, making sure that all branches are caught up on shelving (and stay caught up), measuring pre-float factors like book drops and hold shelves, knowing traffic patterns in the area of service, and communicating tirelessly with staff will go a very long way to making sure the float is as uneventful as possible. Needless to say, the degree to which these collection management and communication practices are in place as a matter of course in the library’s daily life will determine how much time it will take to prepare to float.
A library with great communication in place will have little trouble with the rest of the preparation needed. A library where staff communication is a special event will have to gear up the communication piece before the tasks needed to get the collection floating can be farmed out and accomplished. Staff need to understand why floating is being put in to place, what parts of the collection are being floated, and how it will affect them and their patrons. And it is crucial that every staff member, from the shelvers (who play a critical role) to the branch managers and administrators understand floating. Librarians are sometimes guilty of communicating solely to other librarians and other staff members who work service desks, leaving circulation staff, delivery drivers and shippers, and shelvers out of the conversation. Floating is so granular, that it is impossible to overemphasize the necessity of bringing all employees into this conversation early and often. Staff will have many questions, and not a few fears about floating. Many of them focus around not knowing their collection; librarians worry that as the float happens, they will no longer be able to know what is on their shelves. This concern dissipates almost as soon as floating begins. It turns out that librarians do not give themselves enough credit. They hit the ground running after floating begins, and prove to be much more flexible and have better memory and sourcing abilities than they thought possible. The comfort zone is definitely gone, particularly if the entire collection floats, but librarians rise to the occasion. One way to help ease those staff concerns is to invite librarians from floating library systems to address the staff early on in the floating conversation. This goes a long way to dispelling concerns.
Weeding is an essential preparatory step to readying the collection to float, and again, if a collection has a history of being regularly and vigorously weeded, this is a simple step. But for many library systems, system-wide, regular weeding is unheard of. Weeding is something to do when it’s slow. The problem with that approach is obvious to anyone who works in today’s public library. It is never slow! Maybe back in the day when our mentor librarians were starting out, waiting for a slow period to weed was an effective tactic. With staff and budget cuts, most libraries are at staffing minimums. Add aggressive programming goals and the computer help needed by many patrons, and many library staff are lucky to get daily tasks accomplished. So what to do? Simply put, weeding needs to become a new daily task.
But to clear the decks and get floating going, most systems find themselves in need of a major weeding effort. The goal is pretty straightforward. If the shelves are full with very little wiggle room, at least 15–20 percent of each part of the collection that is designated to float needs to go in order to make room for the items that will float in. Fifteen to twenty percent of the collection is a lot with which to contend. But weeding aggressively will freshen the collection and prevent a huge workload later on. The average system needs to look hard at anything over three years old that has not circulated in a year. Getting that out will raise circulation instantly simply because patrons can find the more desirable items more quickly. It will also create room on the shelves for materials to come in and out. And librarians must make sure that every section that is going to float is weeded to this degree. If only nonfiction shelf sitters are weeded, fiction and music CDs will still be overcrowded when floating begins.
Librarians need to take a hard look at their backroom and see what and how much is waiting to be shelved. If all items are not back on the shelves within a day or two of being returned, not only is the library losing valuable circulation, but it will also be a catastrophe waiting to happen when floating begins, because more high circulating items will accrue in the backroom, particularly at branches with high call-in and drop-off rates. If there are branches in the system where shelvers struggle to catch up and stay caught up, that needs to be addressed and resolved before floating begins. Failure to do so will mean that the most desirable items brought in from the float will languish in the back room, defeating the purpose of refreshing collections for patrons.
The most common concern that librarians have about floating is that a smaller branch will be completely swamped by incoming materials. There are actually a couple of ways to predict what branches will feel an impact, and what branches will not. The first is simply location. If a branch is located near busy commuter routes, more items are dropped off there. But with floating, they stay there, and if the branch is not a top circulating branch, items may not leave fast enough to continue to absorb the drop-offs. Another excellent predictor is to analyze the morning book drop. What is the percentage of items that are owned by that particular branch, and what is the percentage of items that are being dropped off from other branches?
Looking at the hold shelf can also be very helpful. Staff is often convinced for some reason that their hold shelves will explode with floating. That’s just not true. Hold shelves either stay the same or shrink a bit as better maintained collections mean customers more often can find what they need. But it is a good idea to record approximately how many shelves are being used, pre-float, in order to reassure staff once floating has begun. Staff will often maintain that they are a huge drop-off. Running actual metrics on a book drop and a hold shelf and then comparing them across the system will help staff understand where they are in the big picture, and will help them be a little more realistic about what to expect from floating.
One positive aspect about the effects of floating is that they are immediate. By the end of the first day, or possibly the second, it will be very evident which branches are hardest hit. It will take a few weeks to determine which branches are steadily losing materials with the float. Bear in mind, the vast majority of branches will not experience a noticeable difference at all. Floating works great for medium-sized, medium circulating libraries in low- to medium-traffic areas. It is the branches on the extreme end of the spectrum that catch the mega drop-offs or that feel the drain of resources. When this happens, don’t make the mistake some systems make. The ostrich method, for example, is a popular but disastrous choice. Those directing the float maintain that the collection imbalances are not as drastic as librarians think they are, which is laughable, since the librarians are dealing with the fallout each and every day, and know precisely how drastic their particular situation is. Librarians are told that “imbalances happen” and are asked, essentially, to deal with it. Not good. This approach wastes resources, crushes morale, and most of all is a disservice to the customer who either finds a mess of overabundant resources or a scarcity thereof.
A better approach is to try and set up a central clearinghouse of some kind for librarians to communicate their needs. For those libraries with centralized selection, the collection development department is the logical choice. These librarians know all the branch collections, not just one. They also know who is truly experiencing hardship with floating, as opposed to others who may just need time to adjust to the change. Partnering between haves and have-nots can be very effective, in no small part because of the accountability engendered by having only a few people involved in rebalancing and exchanging materials. If the person in charge of the exchange at the sending branch sends materials that are old or are in poor condition, the receiver knows to whom he or she should address his or her concerns. Moreover, communication can flow on a more granular level, so that the sender is not sending just cookbooks, but slow-cookery cookbooks, or five-ingredient cookbooks, or whatever is popular at the receiving branch. The sender has more than enough to share, and the receiver ends up with a tailor-made collection for his or her patrons.
Beware of vendors claiming to be able to do rebalancing automatically. There is currently no vendor program that can rebalance a floating collection. Baker & Taylor’s Collection HQ can swap old shelf sitters in to a new branch, for example, but that is precisely what floating does organically. Collection HQ is very promising, but has, of this writing, not caught up with the swift moving, daily needs of a floating collection. Other vendors try to rebalance by measuring linear feet of books and other items. A superb approach if one is leveraging inventory across several carpet stores, but a pretty blunt hammer to take to a library collection across several branches. A branch that needs medical books on diabetes can receive cookbooks. Or a branch needing cookbooks on Crock-pot cookery can receive books on Tuscan cookery. The collection may be balanced numerically with this method, but the gestalt of a well-balanced collection tailored to a particular community’s needs cannot be achieved with the linear feet approach.
There is no magic bullet for rebalancing. There is no magic bullet for rebalancing. There is no magic bullet for rebalancing. Once librarians have repeated that to themselves often enough and internalized that mantra, they can stop looking outward for solutions to rebalancing and begin studying what other systems have done, and what they might effectively adapt for their own system. A workgroup composed of internal experienced librarians, circulation, and shelving staff will be best equipped for finding those rebalancing answers within their own organization.
Perhaps the best news about floating is that it is invisible to the customers. Staff have sometimes worried that their patrons will be disappointed somehow by floating, and the opposite is true. Patrons will claim that the library has bought more DVDs and best sellers, when in actuality, the system has bought the same number as always, or fewer. Why do patrons perceive it this way? Simple. Materials that used to be on trucks and in delivery bins simply pop right back out to the shelves. The patrons are not wrong; there is more to peruse because the hot merchandise is not on a truck or in a bin.
Workflow after floating has begun must include weeding as a daily task. Librarians will be stunned by the condition of items that arrive from branches that were not as fastidious about weeding. Shelvers and their supervisors will want to make sure that the shelving is kept caught up so that the hot and desirable materials are in front of the patrons as much as possible, and no time is lost in the backroom awaiting shelving. Merchandising is a much more dynamic process with a floating collection as a new browsing collection appears several times a day in a busy branch. Librarians will want to capitalize on that browsing collection to expand the selections that they want to point out to their patrons.
Floating is fast becoming a widely accepted best practice in many library systems across the United States and Canada. This popularity is due largely to two outcomes that floating produces: saving money and refreshing the collection, and increasing circulation. Floating lowers the cost of delivery because items are not returned to an owning location. For the same reason, it saves time and labor costs at circulation and shipping. Collection development can get more mileage out of popular titles because items do not spend down time on a delivery truck, but go out immediately to a new patron from wherever they have landed. Patrons love browsing the refreshed collections, and that is reflected in circulation.
But floating brings challenges and problems that become a permanent workflow issue. Floating materials will accrue unevenly at branches with high call-in and drop-off rates. Other branches where the book return is difficult to use, or that are farther off highly traveled routes, will lose more books to the float than they will accrue. This imbalance will become a constant issue for librarians at both extremes of the floating continuum as they work harder to have a balanced collection to offer their patrons.
Many library systems have opted for the best of both worlds by floating only a portion of their collection, often popular best sellers, movies, and music. Others include adult fiction. Many with large research collections at the main branch float everything except the main collection. No matter what the choice, frequent and two-way communication with all staff is critical to a successful float with the fewest possible negative outcomes.
Preparing for floating by having a well-weeded collection that is shelved in a timely fashion lessens the severity of collection issues created by floating. Impact can be predicted pretty accurately by monitoring traffic, and looking at the book drops and hold shelves. Once floating has launched, the imbalance in the collection has to be addressed. There is no single method to address this, but some kind of central oversight is ideal to maximize opportunities for getting the most out of materials, and because the busier branches will need a hand with rebalancing. There is no magic bullet for rebalancing, and librarians must stay open to the realization that they may have to change their approach to it from time to time as the collection demands will change. And remember, the fewer parts of the collection that float, the less rebalancing will rear its ugly head!
Once floating is underway, patrons will find new materials, which will enjoy a new life in a new setting. Librarians will also see the collection begin to change over time as materials float in and out, and will be able to add even more titles to the repertoire with which they help patrons. Browsing collections, particularly of new materials, movies, and music, thrive. Best of all, savings of delivery, labor, and material costs mean more money is available for administrators to infuse into other critical services. And if the library’s budget situation is not in desperate straits, a mixed floating or nonfloating collection is often just the ticket to wake up and revitalize a traditionally housed collection.
Bartlett, Wendy K. 2013. Floating Collections: A Collection Development Model for Long-Term Success. Santa Barbara, CA: Libraries Unlimited.
Cress, Ann. 2004. “The Latest Wave.” Library Journal 129 (16): 48–50.
Hilyard, Nann Blaine, ed. 2012. “Take the Plunge! Implementing Floating Collections in Your Library System.” Public Libraries 51 (3): 13–20.
As I began this chapter in the summer of 2012, a drama was playing out on Bryant Park.1 It was a real-world, if not reality, show presided over by iconic lions, and it involved, for those viewers in the academic library collections business,2 a by-now familiar cast of characters, plot elements, and dialogue, complete with tirades from important actors, production numbers for protesters, and a panoply of Brechtian commentary on the action from the press, bloggers, and miscellaneous fulminators and apologists. It potentially pitted the interests of some library users against those of others, rested on mission-shaping questions of asset allocation, and surfaced decisions informed by technological and resource-based change.
The script concerned plans announced by its president, staff, and board to send more of the New York Public Library’s (NYPL) research collection to, from the point of view of Manhattanites, the desolate, remote, and savage plains of New Jersey and a facility operated, but not a collection held jointly, by NYPL and Columbia and Princeton universities. The collective protagonist Administration would then redevelop the space vacated by books for other traditional library functions taken into the leonine 42nd Street building from two buildings NYPL proposed to sell in order to support its research functions. This drama is emblematic of the tensions surrounding print collections on college and university campuses and particularly concerns itself with disputes over the allocation and functions of library space, disputes that not only reveal the needs, work practices, preferences, and turfs of different populations, but also a particular and powerful presence with which books, physical books, have become invested. Moreover, this very mythology of presence implicates the ethical status printed books enjoy as bearers of knowledge and identity and the workflows, interests, and cultures that surround them. The place of physical books in the future of libraries also evokes for many readers a larger set of tensions induced by the massive cultural shifts underway among the institutions of knowledge creation, publication, dissemination, use, and preservation, not to mention entertainment, communication, and community formation, in the wake of the 20-year commercial exploitation of the World Wide Web.
Libraries are what they collect, are they not? To think of a library is to think of what’s on the shelf. The collection so conditions the connotations of library that from both popular and scholarly perspectives a library is institutionally as well as etymologically nonsensical without those shelves of books. Indeed, the commonsense connotation of library, especially of academic library, is stacks, endless and hushed stacks of books, the sounds of pages turning at remote desks hidden away among the ranges or at reading tables in well-proportioned, imperially high and preferably coffered-ceilinged rooms, the silence punctuated only by the occasional sneeze, whispered piece of advice, or, these days, the ring of a rogue phone.
Today, as the NYPL example demonstrates, common sense is tense with change in the service and materials interests of various groups with their several traditions of building use and their respective work practices and schedules: witness the complaints about lack of quiet space or the merry sounds of barely suppressed laughter among students working together, the conflicts about electronic versus print materials, open-stack versus remote high-density shelving, owned materials versus leased or borrowed, computers and cafes versus books. Academic libraries are adjusting with varying degrees of urgency and with emphases that depend on their local circumstances and mission to new information and education environments, and they are doing so in many aspects of their work by collaborating more closely than heretofore, deepening their dependence on other libraries and indeed restructuring their work strategically through these partnerships.3 The shift for many libraries from retention on campus shelves to retention in off-site facilities is vexed enough, as is the case at NYPL; divesting of print volumes, especially of monographs, in favor of electronic or partnered, remote access falls for many readers and researchers on a continuum from worrisome to crazy. In a fundamental sense, academic librarians are rethinking the role of the local, open-stack collection—the very thing whose definition, composition, and service have constituted the idea of library—as libraries adjust to the several pressures and opportunities confronting them and as issues of the local versus shared approach to collections and materials management come to the fore. As libraries move beyond resource sharing to shared print4 or, better, as resource sharing becomes shared print and as the materials to which libraries give access are increasingly electronic, librarians are asking questions about what in fact constitutes a local collection these days.
Since the chapters in this book propose to account for and contribute to the current rethinking of collections, my treatment of shared print begins with the traditions of collection development and management as they have been codified by textbooks and, presumably, as evidenced by their bibliographies, in the practices of local libraries.5 As librarians have come to understand them, the activities and practices of collection development and management proceed from the assumption that the gathering, organization, and preservation of a group of materials are specific to an institutional context. For libraries in higher education, the mission and curriculum of the parent institution, the pedagogies its faculty employ, the kind of work it asks its students to complete, the programs and degrees it offers, the degree to which it emphasizes research, and the students it enrolls determine the kinds, provenance, and formats of materials the library collects, that is, owns and places on a shelf or a server or otherwise gives access to. These same institutional circumstances inform the depth and breadth of a library’s collecting efforts, the criteria it applies to decisions to retain items, and the position it takes among other collections and in partnerships.
Whether understood as a prescriptive or self-study document, the collection development policy written by library staff and vetted with their constituents codifies an understanding of the library’s place in the institution and in most cases explicitly states how the library identifies and procures materials, budgets for its collecting and preservation activities, gains knowledge of the needs of its constituents, analyses and makes decisions about the collections, staffs its collecting activities, houses materials, and relies on means other than purchase or subscription to provide materials to its users. These policy elements in turn imply, even if they do not explicitly name, a host of such other aspects of the library as user infrastructure (signage, circulation rules, communication lines, advisory services, space allocation), interpretation and promotion of collections, information systems for discovery of and access to materials, the quality and content of data about materials, staff roles and expertise, the creation and dissemination of knowledge, the legal and commercial relationships entailed by the processes of knowledge creation and dissemination, and, bringing the policy back to its reason for being, the mechanisms or criteria employed to assess the success of the library’s contribution to the fulfillment of its parent institution’s mission.
The library in its materials-provision dimension may function in a complex of such extra-institutional organizations as publishers, vendors, information systems, other libraries, and consortia, but its existence depends on and serves above all else the interests of the students and faculty, alumni, and citizens who constitute its parent institution’s community. Elaborated and justified as the reasons for its collecting role have been by generations of how-to and best-practices advice and by reflection on how the tenets and traditions of collection development and management adjust to changing circumstances and contexts; the library’s status as a local enterprise charged with the gathering and guardianship of (print) materials has become a cornerstone of librarians’ sense of their professional identity.
The body of thought and experience that defines this collection development tradition incorporates, nonetheless, and contrary to the primacy of the local, the sense of a library’s being a node in a network, one among many beasts inhabiting the ecosystem of the information jungle. The history and literature of cooperative collecting and interlibrary materials provision dates from the inception of modern librarianship in the United States in the later 19th century (Johnson, citing Melvil Dewey, 2009, 264; Burgett, Phillips, and Haar, 2004) and ranges from the passive, that is, allowing readers to know what is available elsewhere, to the more active stance of using knowledge of other libraries’ holdings to inform local decisions about selection and deselection, to the yet more active, that is, agreeing to build complementary holdings of expensive items or agreeing to gather and house centrally or among partners segments of such bodies of material as newspapers or foreign governmental or commercial publications. Over the years, printed and online information systems and delivery strategies, in tandem with the sheer number of collectable publications and the space and fiscal limits that shape any given library’s ability to amass material, have taken the cooperative impulses constitutive of modern librarianship and, in effect, made gathering or otherwise enabling materials access a collective activity.
Again, however, libraries are what they collect, are they not? Local library staff are proud of what they do for their students and faculty, proud to have anticipated reader interests by aggregating material, proud to be able to pull the rabbit of a new or oddball old book out of the hat with a trip to the shelf or the click of a mouse. Local library staff are proud to take care of their own and suspicious of the ability of other libraries’ staff to do so; they are wary of entrusting their readers’ fate to “that library over there” where, as everyone knows, staff couldn’t muster a fully functioning brain or efficient workflow among them. Cooperation on collections may be in the DNA of libraries, but independence is as well. Librarians are, like other members of the species, territorial, hierarchical, and tribal beings, which means that institutional independence, whether grounded in mission, tradition, financial resources, governance and jurisdiction, or size and reputation, conflicts with common interests. At least for print materials, and in spite of their capacity for sharing, librarians will cite geography, delivery time, reader preferences and habits, and concerns about other libraries’ commitments to collection management as reasons to gather and retain collections on campus. As the title of this chapter suggests, though, and as the changing information ecology requires, the local is sorely pressed these days to remain so.
A manifold of developments in higher education, publishing, the economy, and communication are prompting libraries to reconsider their several roles—collector, service and workspace provider, node in the network of collecting and cultural heritage institutions, site for learning, builder of academic communities, link in the chain of scholarly communication, and publisher. Per Richard Lanham, the library in its preservation aspect “has always operated with a digital, not a fixed print, logic. Books, the physical books themselves, were incidental to the real library mission, which was the dispersion of knowledge” (Fitzpatrick, 2011, 126, quoting Lanham). Put another way, staff who work in buildings called libraries are in the education business, not the library business, which means that the library and its collection are not the heart of the college, rather, the student–faculty and other learning relationships are. In an information-abundant world, libraries are entertaining, or are being asked to entertain by technological, fiscal, and social change, a series of cultural shifts that question the traditions governing the practices and relationships of gathering and organizing materials. The current age of digital reproduction and delivery severely stresses the concept of local collections as the aggregate or collective becomes the new local. At the very least from the point of view of the new information environment, and controversial though radical sharing of print collections is for many, an individualized collecting effort in a loosely federated, interlibrary-lending support system is insufficient for meeting the needs and expectations of a campus’s user base of students and scholars. A library materials environment that emphasizes collective effort (Association of Research Libraries 2012), therefore, proceeds from several trends or assumptions, not the least of which is that convenience of access trumps all for readers (Connaway, Dickey, and Radford 2011). Libraries’ readers and researchers want access to as much material as possible as quickly as possible, and the proliferation of information published on the open Web encourages them in this desire. Digitization greatly expands access to and the potential uses of information or text at substantially less cost than that of housing print (Courant and Nielsen 2012). Electronic delivery of journal content is now generally accepted (Schonfeld and Housewright 2009b), while print-on-demand shortens delivery time of remotely accessed text.
The maturation of these trends and circumstances has several consequences for the traditional practices of collection development and management and for the work of staff associated with them. If libraries are what they collect, then such changes to collection programs as those advocated by Atkinson (2005) and Lewis (2013) lead to a different model of library identity, one constituted as a collective, multisubjective entity whose identity is based, for the most part, not on collections, on things, but on relationships. As libraries consider different balances among the items in their service portfolio and as the value of their several assets of space, staff, collections, partnerships and affiliations, commercial relationships changes (Dempsey 2012; Lavoie and Dempsey 2010), the hard truth for many in this new information environment is that the value of the print collection as an asset for study and scholarship is changing relative to the value of other assets. On many campuses the space occupied by less-used print materials housed on open shelves has become more important for other purposes.6
Advocates of shared print make their argument from several propositions. It saves costs of housing low-use older materials in expensive central campus real estate, allows for the reinvestment of that space, and, in the case of new publications, reduces the rate of unnecessarily duplicative collection growth. Especially for librarians concerned about maintaining the record of publication, large-scale collaborative approaches to managing print collections reduce the risk of loss of scarce and unique or even all copies through consortial collection assessments that systematize deaccessioning and retention. Moreover, shared print potentially increases preservation capacity by placing low-use materials in more secure environments, allowing the library community to concentrate preservation efforts on fewer copies, and encouraging greater access through digitization programs that run parallel to materials storage initiatives. Spending less on the maintenance of print collections also enables libraries to shift resources to other services and materials; through the kinds of institutional collaboration required for shared print and its electronic counterparts, libraries can strengthen the foundation for support of scholarship along other inter-institutional dimensions.
Major recent reports from OCLC Research (Lavoie, Malpas, and Shipengrover 2012; Malpas 2009, 2011; Payne 2007) have benchmarked the developing situation with respect to shared print and shared collections management. Other policy- and data-centered papers (Lavoie and Dempsey 2009; Lavoie and Schonfeld 2006; Lavoie, Connaway, and Dempsey 2005; Reilly 2004; Reilly and DesRosiers 2003; Schonfeld and Housewright 2009c), together with planning documents and reports from many consortia and organizations, support the argument for shared print and inform collective strategies that supersede libraries’ familiar individual roles for gathering and maintaining the raw materials of scholarship.7 Particularly as academic libraries and their parent institutions have responded to the technological and fiscal challenges of the past five years, shared print has become an important means for adjusting to local circumstances as new consortia have entered the picture and especially as existing partnerships or consortia based on materials housing, resource sharing, or eResource purchasing have built shared print onto their inter-institutional trust networks. The shared print movement has grown in variety and in the number of program implementations; it has been enlivened by the development of consulting practices and a community of interest, a rapidly diversifying multiplication of conference sessions and Webinars, and the establishment of an informal discussion group that meets at each ALA Conference under the auspices of the CRL (Kieft and Reilly 2009). Shared print has arrived at a point, in other words, where thinking about and incentives for it are sufficiently mature and abundant, and models for shared print agreements sufficiently numerous and well developed that groups of libraries may draw on a body of knowledge and experience to develop their own approach.8
In the past 5–10 years, the evolving practice of shared print librarianship has concerned itself largely with journal collections in regional configurations or state systems. In many cases, these journal programs rely on a decentralized housing model that promises long-term but not perpetual preservation of materials in access-oriented light archives with libraries absorbing their own costs of processing or donation and with relatively small amounts of money changing hands (Kieft and Payne 2012).9 Such consortia as the Western Regional Storage Trust (WEST) (Stambaugh 2010) have thus adopted a pragmatic, lightweight approach, doing what they can relatively quickly with materials in place and setting a retention period at once long enough to satisfy conservatives and not so long as to impair the ability of partners to adjust archiving commitments. By the same token, keeping the lights on in the archive, relying on existing consortial resource-sharing or ILL protocols for access, and specifying exit requirements, these consortia have reduced the likelihood that an already complex planning process will bog down in persuading members to deposit items in a dark archive (and all that entails for designating service copies), or in creating new workflows and funding streams. In other words, they assemble components ready to hand rather than build something completely new.
In that such prominent examples of shared housing facilities10 as the Five Colleges of Massachusetts, Washington Research Libraries Consortium (WRLC), and the Research Collections and Preservation Consortium (ReCAP) were designed for geographically proximate institutions, such distributed journal programs as WEST’s (100+ members in 17 states), Association of Southeastern Research Libraries’ (ASERL; 40 members in 11 states), and Committee on Institutional Cooperation’s (CIC: 13 members in 9 states) expand the collective housing model to very large territories. Existing journal archiving projects also exhibit some tendency to consolidate and achieve extra-regional reach, for example, the members of the Consortium of Academic and Research Libraries in Illinois (CARLI) joining en masse in 2011 as affiliates of the Five College Depository for access to back files of journals collected there, or the partnership of ASERL and WRLC for journal archives.11 In addition to geographical expansion, the past few years have witnessed the development of new organizational relationships for preserving and serving print collections, for example, that between the Linda Hall Library and CRL for scientific journals.12
Vital to all shared print programs, whether distributed or centralized and whatever the terms of or parties to their founding, are the emerging tools and services that enable individual and consortial collection management decisions based on an understanding of the composition, status, and use of titles in collections. For journal archiving projects, ASERL and CIC are using local methods for comparing holdings, assigning archiving responsibilities, and calling for volumes to complete archives. WEST ingests member holdings and uses Print Archives Preservation Registry (PAPR) (see later in this chapter) to process them into assignments for archiving and to identify members that have volumes to fill in holdings gaps. Chief among the new tools available is Sustainable Collections Services’ (SCS) actionable collection intelligence analysis program for monographs. SCS has made rapid strides in the past couple of years in helping libraries and groups of libraries, including several of the monographic projects mentioned later, understand their collections and the potential for joint management of them with a variety of comparisons using holdings and circulation data supplied by libraries in conjunction with holdings data extracted from the WorldCat database. OCLC also promises a new version of its long-available and clumsy WorldCat Collection Analysis tool in 2013.13
In addition to being able to analyze holdings and use data, shared print agreements depend on specifying and disclosing the retention status and, ideally, condition of items in the shared collection so that partners as well as other libraries and consortia can deselect local materials in favor of the shared copies as they deem advisable. To meet this need, OCLC conducted a Print Archives Disclosure Pilot beginning in 2010. In 2012, it released a final report with draft metadata standards for the MARC 583 field, a preservation data dictionary patterned after the existing Preservation and Digitization Action term list, and resource-sharing test scenarios (OCLC 2012). In addition to this report and the practices it recommends, OCLC has established a staffed shared print management program for its members.14 Beyond the print archiving data standards that OCLC is building into WorldCat, CRL has developed the searchable PAPR knowledgebase to support “archiving and management of serial collections by providing comprehensive information about titles, holdings, and archiving terms and conditions of major print archiving programs.”15 CRL’s system thus not only includes retention and preservation data but also information about the consortial arrangements under which materials are kept. These two means for recording and disclosing archiving commitments contribute important decision-making tools for libraries as they shape local and regional collections under shared print agreements; they therefore serve as important building blocks of the developing community of interest and infrastructure of tools for shared print.
If work on journals has dominated the contemporary shared print agenda, projects for other classes of materials and in additional configurations of libraries expand the scale and scope of shared print activity, develop different cost-sharing options, and consider alternative means for sharing responsibility for the retention or purchase of print materials. Projects within and beyond consortia working on journals have taken, for example, a domain-based approach. With funding from the Andrew W. Mellon Foundation, CRL, in partnership with relevant national libraries and major collections, has undertaken collection preservation and access programs for agriculture17 and law18 materials. The area of U.S. government documents is ripe for shared print treatment in that a network of libraries is in place, collections are large and highly duplicative at the regional level, and digital versions exist or could be produced under such arrangements as those of the CIC with Google.19 As discussions continue about the future of the Federal Depository Library Program and its retention rules,20 a shared-print step in the direction of a different retention model for historical documents has been taken by ASERL with the establishment of a Collaborative Federal Depository Program,21 which creates comprehensive collections of federal agency publications in Centers of Excellence throughout ASERL’s region.
As journal sharing and other projects gain momentum, consortia are also turning their attention to monographs, which present the greatest challenges to shared print for a complex of reasons, not least because of their sheer number and the trivial amount of shelf-space gained per disposition decision (Kieft and Payne 2011). Local as opposed to collective sensibilities about books, the difficulty of easily gathering and using data about holdings and circulation, and the lack of business models for large-scale retention and serving of monographs, together with a corollary uncertainty about how many copies are needed to serve foreseeable demand, are significant impediments to program development. Sorting out roles among libraries for preserving copies and the shadow cast by e-text use experience, sales models for eBooks, and lack of full-text access to digitized orphan and in-copyright books also contribute their fair share of discouragement to the prospects for shared monograph collections.
Important though these financial, legal, and governance issues are, even more important are the affective dimensions of monographic shared print. Many librarians and faculty feel a sense of loss in what they see as the implicit devaluing of books in any program that draws down local, open-shelf print in favor of off-site housing or partnered access; they feel that (mere) administrators are forcing them to change their work practices and preferences, or even violating them, and are slighting their discipline by reducing its physical representation on the open shelves. Reading, teaching, and learning practices and text-use preferences are acutely in question in the monographic case, as is the place that print has come to occupy in culture generally and more specifically in the elaboration of library collections and services.
In addition to the controversy that surrounds books and reading, then, the monographic case exaggerates all the challenges that obtain in journal or domain-based shared print agreements. No recipe exists for making monographic shared print work, whether for older or new materials, print or electronic books, but several projects are bringing the requisite ingredients coherently together. Two stand out at this point, namely, the IMLS-funded Maine Shared Collections Strategy (MSCS),22 which involves two public library systems and six public and private colleges and universities, and the Midwest Collaborative for Library Services’ Michigan Shared Print Initiative (MCLS MI-SPI),23 which involves nine publicly funded universities, including one ARL. MSCS’ project began in 2010 and MI-SPI’s in 2011; collection analysis and related consulting by SCS inform them. Both projects focus on identification and retention of a body of materials for 15 years, and both agree to retain those volumes in open or storage stacks under customary inventory control methods and the exercise of local preservation practices. Both factor into retention decisions parameters for acquisition date, recorded use, and the availability of the title elsewhere in the country. Neither places onerous burdens on participants, with the Michigan project’s MOU emphasizing local library autonomy more often. Neither makes withdrawing from the agreement or deaccessioning retained volumes particularly difficult, and both rely on participant good faith and flexibility in maintaining the agreement as circumstances change.
The projects diverge, however, in the matter of identifying copies for retention. The Maine project focuses on unique and scarcely held titles and does not foresee, although it does not preclude, deselecting nonretained copies during its three-year course; moreover, as of May 2013, MSCS has yet to decide how to approach the roughly 2 million title-holdings with more than the (up to) two copies they want to retain. The Michigan project starts, conversely, by identifying titles with more than two copies that meet its deselection criteria. In other words, MI-SPI assumes from the beginning that libraries may divest of any copies beyond the two designated as retained in order to enable space reallocation, and it excluded from consideration copies uniquely held.
Both projects address older materials, but Michigan starts with the commonly held and unused and Maine with the scarce or unique. They are thus well down the road to retention commitments and deselection-based space management. Other consortia are at earlier stages. The six Los Angeles–based California state universities are in receipt of a collection analysis from SCS as an aspect of the work of the system-wide Libraries of the Future Taskforce (LOFT); they expect in 2013 to come to conclusions about shared print based on it.24 Elsewhere, SCS is supplying an analysis and working on retention scenario development with 12 members of the ConnectNY group,25and in-process monographic collection analyses undertaken by the Greater Western Library Alliance (GWLA)26and the Council of Plains and Pacific University Libraries (COPPUL)27 are contending with effectively using large OCLC-generated holdings data sets. Beyond these projects that are operational or are at the data gathering and review stages, several others are incipient. Beginning in the summer of 2012, Five Colleges of Massachusetts have served as the convener of discussions that may lead, with the encouragement of a grant from the Andrew W. Mellon Foundation received in the late winter of 2013, to the creation of a Northeast Regional Depository (NERD) (Five Colleges of Massachusetts 2012). They are launching a feasibility study to gauge interest among libraries in New England and are undertaking a research agenda that seeks to answer such questions about monograph archiving as how to qualify the differences among or uniqueness of copies, how many copies to keep, how to coordinate retention at different locations, how light the archive can be, what the right delivery time is, and how NERD would fit into the national or international picture.28 Discussion is also beginning in California among representatives of the several segments of the academic library community about whether a regional approach to monographs might develop based on OCLC’s mega-regions model, existing resource-sharing consortia, and collection studies undertaken by individual systems and groups.
The projects and initiatives discussed here begin to manifest a typical sequence, set of components, or decision points from which best practices are emerging. On the whole, it seems that where there is a history of resource sharing, reinforced by shared or linked library information systems, eResource purchasing, and rapid materials delivery, libraries find greater potential for collaboration on or federation of collections new and old. Conversely, in states or regions that lack a well-developed user-initiated system for resource-sharing or where, unlike the case of Minitex and the University of Minnesota, an ARL library in the state does not see its mission as including regional leadership for collections initiatives, colleges cite such absences to explain why they cannot think seriously about collaborative collections activities.29 In cases where a consortium chooses to build a shared print program as an extension of their resource-sharing experience, partners start with a collection composition and use study so that members may understand the number of titles, items, and holdings overlaps in the partnership, the partnership’s holdings with respect to those in specific other partnerships or geographical groups or in digitized collections, and an approximation of aggregate demand based on circulation history over the number of copies in the partnership. This analysis typically reveals a host of cataloging and record-matching anomalies that partners need to address in order to create greater confidence in the integrity of the data on which their retention commitments will rest. As a project develops toward recording and displaying retention commitments, it may also engage in systems development with the vendors of local and regional catalogs and OCLC so that the requisite data is adequately displayed as it moves among them.
In the case of journals, retention and archiving decisions are based on risk analysis (WEST), title (CIC), or nomination (ASERL). Only the first of these is practical for monographs in one of two ways, first, by assuming that uniquely held or scarce titles are at high risk and retaining them is therefore important for maintaining diversity in the knowledge ecosystem or, second, by assuming that for commonly held titles the risk of reducing the number of copies is low because the retained copies will be able to meet foreseeable demand. Whichever their focus, though, current projects appear to assume that (up to) two copies are sufficient when circulation is very low over a substantial period of time.30 Thus far, Maine’s is the first project for monographs that is registering a large number of retention commitments using newly created OCLC holdings symbols for shared print, although in the spring of 2013 ReCAP also decided to declare 5 million of the volumes held there as shared.31
Partners record their commitments in an MOU that defines the relationships among them, a governance structure, and decision-making protocols, exit clauses, and retention and agreement-review periods. In addition to relying on the relatively lightweight policy requirements in such MOUs as those of MSCS and MI-SPI that specify a 15-year retention commitment and do not demand item validation at the time of retention commitment, shared print agreements have tried to keep participation costs low through asking libraries to fund internally the costs of deselection and shipping of materials to archivers or through distribution of archiving responsibilities among member libraries that have already paid or in any case would have paid at least some of the costs of facilities construction and of processing and validating archived publications. In the WEST and CIC journal projects, for example, relatively small membership fees pay some of the costs for collection analysis services, for supporting the creation of archives, or for ongoing compensation to the archiving site for space use. The Five Colleges of Massachusetts charge an affiliate membership fee to libraries that want to rely on access to the journals they have archived, and HathiTrust foresees instituting a member fee to support the long-term storage costs that libraries which house monographs for its prospective Distributed Print Monographs Archive will incur. These relatively low-cost methods for participant support of journal projects have worked as projects are achieving early stability, but, given the cost of buildings, projects are well advised in the early stages to consider options for declaring materials already housed in storage as the base shared collection as ReCAP is doing and as WEST did with journals already held in Orbis Cascade’s Distributed Print Repository and in the University of California system.
A major condition of possibility for large-scale collaboration on older monographs is the availability of a digital library that allows readers to perform many of the text-use functions that have in the past required printed books, for, if nothing else, the use of digitized text eliminates the need for some shipping of materials among partners. The replacement of print with e-text is controversial on several grounds,32 but shared print agreements, as well as individual libraries, find it attractive to depend for public domain texts on digitized versions as they deaccession them or retain as a group a single copy. Shared print partnerships can certainly retain or deaccession large numbers of monographs without resort to digital surrogates on the assumption that less-used titles will continue to experience low use and can be delivered as needed to readers. The availability of e-text, however, just as certainly conditions the perceived role of print books and the demand for them. Such statements as “It’s all on the Web, so why do we need libraries?” betray at the very least an incomplete understanding of what’s on the Web and what libraries do, but in some quarters librarians and readers have to make an argument for retaining print at all these days. In that the demand for print will decline and the decline itself will affect the willingness of institutions to fund its retention; the rationales for print retention and business models for serving print have to account for and include electronic collections.
The approval by the members of HathiTrust in 2011 of a proposal to plan a Distributed Print Monographs Archive, in which “a group of print storage repositories will assume formal curatorial responsibilities for volumes that have a corresponding digital representation in HathiTrust’s collection,” is extremely important in this regard if only because OCLC Research has determined that approximately 33 percent and growing of any given academic library’s print collection matches digitized copies in HathiTrust and 31 percent of Hathi’s digitized texts are in the public domain (Malpas 2011). Members have not yet acted on this proposal; given, however, the distribution of Hathi’s membership across the country and the size and importance of their print collections, given, too, the possibility that Hathi’s millions of books could become a generally accessible digital library under an access agreement between rights holders and libraries, a Hathi-based archive of print would lend invaluable support to a national alignment of collections of print monographs.
In summary, these projects for print titles are working on aspects of an ideal or facilitator state for monographic shared print in which,
• interoperable library information systems and data enable (1) accurate and easily actionable holdings and materials-use comparisons for judging aggregate traffic and demand, (2) consistent and clear communication about item condition and archiving status, and (3) user-initiated borrowing within and potentially across current groups and systems;
• a materials housing program combines archiving off-site and in place in both small and large repositories; in other words, a program creates a network of housing nodes based on natural or current alignments, a several-to-many cooperative;
• collectively funded and governed arrangements allow libraries to retain and serve scarce and unique copies and to serve as libraries of last resort, coupled with an understanding of how many partners or how large a collective collection needs to be in order to meet demand;
• an archiving collaborative balances current access with long-term preservation of copies and is nuanced enough as to allow for the changing needs of individual libraries; and
• linking digitized texts from the catalog, a digitization program, or including digitized texts in discovery services provides access without physical delivery.
The most important task before shared print programs, especially those for monographs and as suggested by the foregoing, is the development of business models that will allow some libraries to retain and serve the shared copies. Shared print projects may well predicate themselves on the existence of information and delivery systems that enable readers to easily discover, locate, request, and receive items from partners, but, as the Maine and Michigan projects make clear and as the NERD grant and incipient discussions in California are assuming, the success of shared print will depend on those libraries that will commit to playing the archiving role. As OCLC’s several analyses of WorldCat holdings and SCS’s more local studies show, all libraries have unusual or unique items that could be worth retaining; a relatively small number of libraries, however, can collaborate on archiving them for shared print agreements. In the past, the library community has relied on an informal or tacit understanding of the roles of college, research, public, and special libraries in preserving materials. Those roles and the relationships they define are probably still in place, but for shared print they must achieve a level of formality they have not heretofore enjoyed. As shared print formalizes libraries’ collective roles, archiving libraries will need support from the library community on whose behalf they play the archival role, whether that support comes in the form of donation of materials under last- or scarce-copy policies, a fee for ingest, transaction fees for borrowing, or an annual membership charge like CRL’s that funds the repository function. Shared print partnerships will have to contribute to the cost of building and maintaining the archive of retained copies in ways that give the host library more than the satisfaction of being a good citizen, and those costs are going to look like new money to libraries that divest of print in favor of depending on shared access.
The major hurdle to leap, then, in terms of the business of shared print under the roof of an increasingly digital library is less that of placing a price on the collective responsibility for maintaining the shared collection than it is of creating value from maintaining large collections of low-use items. Having made the argument locally that a library need not retain low-use materials on campus, the shared print librarian must turn around to argue for his or her library’s helping fund a collective effort to do just that. Addressing the issues of inventory size and control will help that argument, but the argument for bearing the costs of collaborating on building and maintaining print archives, of recording and disclosing useful data about them, and potentially of creating digital versions of those texts has to be grounded in libraries’ collective stewardship role for the record of publication and the way in which that record enables study, scholarship, and teaching. Shared print librarianship has much to do to develop business models for an environment where the number of copies is decreasing and the remaining copies concentrate in fewer libraries; the cost of serving those copies should decrease as the demand for print declines, but that decline will put all the more pressure on the shared print partnership to justify the costs of maintaining the archive.
At this point, it appears that shared print and the conditions required by shared print agreements will not result from designing a systematic, directed, or planned national-level program but from an additive, ad hoc series of local and regional projects proceeding on the same general precepts and values with respect to library collections. The virtue of this approach is, of course, its reliance on existing partnerships and it’s not requiring the level of organization that a national approach would; that a national program will emerge from a series of local and regional programs conscious of each other is, moreover, consonant with the mosaic of jurisdictions, consortia, library systems, and partnerships that constitute the U.S. library community. In fact, the presence among local and regional projects of such funders as the Andrew W. Mellon Foundation and the IMLS, such research groups as OCLC and Ithaka S + R, such consultants as Samuel Demas, Lizanne Payne, and SCS, and such (overlapping) membership organizations as ARL, CLIR, CRL, and OCLC may well create a sufficiently integrated environment for communication and joint activity that the sum of local and regional projects will become a reasonably coherent national program in less time and with less effort than would be the case if libraries attempted to create an overarching national structure.
From all reports, marriage is a relatively easy status to attain but difficult, even arduous, to maintain successfully. In the case of libraries and their parent institutions, however, the contrary may be closer to the truth, for, unlike human beings, they never enter relationships willy-nilly as a Las Vegas lark, or for lack of anything better to do, and in no known cases at the wrong end of a shotgun in anticipation of the arrival of ill-timed offspring. Rather, like royal houses and empires, they enter into marriages grounded in equal parts guardedness and calculation of mutual territorial or dynastic advantage. In spite of the often-cited impediments to collections collaboration (Burgett et al. 2004, ch. 2), libraries have flirted since the beginning with professions of common interest, gone steady through a century of resource sharing, and shacked up in shared collections and facilities. The ways in which libraries and the colleges and universities that house them have heretofore in the world of print gained individual status among their kin have militated against the kinds of deep cooperation, even willed dependency, that economic and cultural forces abroad today are encouraging. Since love is not the question, except perhaps for readers meeting in the stacks or cafe, the relationships articulated by the Janus Agenda (Atkinson 2005), a host of MOU pre-nups, and the much-heralded 2CUL banns of Cornell and Columbia Universities are reconceiving funding and institutional prestige on a very different, committed ground.
The topics that animate shared print discussions and the goals that define the projects mentioned here at once mimic long-standing arrangements and patterns and require new approaches to them. Although the sense of what a library is remains local for many users and librarians, an appealing, if not universally agreeable, vision for shared print has evolved. The projects invoked here make the case for shared print in its many dimensions and suggest the contours of a collective future for print collections as programs take shape in many parts of the country and among a variety of partnerships. In manifesting a tendency to geographical expansion and toward interregional cooperation, the evidence suggests an emergent national structure and an understanding of the scale at which shared print can achieve critical mass. In the near future, one size, one collection, could well fit all. Even if, and especially if, one thinks “we are all weird,” as collection analyses tend to indicate in terms of holdings of unique items, the universal library of digitized and collectively archived print text will level many of the distinctions among libraries in higher education, digital divides notwithstanding.
The collaborative futures adumbrated by the new civil unions among libraries affect all aspects of their relationships to each other and the roles they play in the world. Those suggested by OCLC Research’s proposed reexamination of the library service bundle (Dempsey 2012; Lavoie and Dempsey 2010), by Lewis for a new collections regime in 2020 (2013), and by the present author and Payne (2012) for an “ideal state” of the print collections system in the 2020s are now becoming or soon will become concrete for a library community in which the tasks of collections staff will orient toward the collective or network of users and materials more than to the local campus. The research and modeling published in OCLC’s “Print Collections at Mega-Scale” (Lavoie, Malpas, and Shipengrover 2012) provide a framework that enables libraries to understand the regional and national implications of print holdings management and opens the way to establishing a more coherent program than libraries now have for collective decision making, developing the bibliographic data needed for shared collections management, creating local and regional collections policies that look beyond the present to the long term, and communicating a narrative about the value of collections collaboration for library constituencies.
Not all the lions guarding NYPL’s imposing Fifth Avenue entrance may soon lie down with the lambs of ReCAP in central Jersey, and libraries and their consortia will continue to experience conflict as the largely local print library gives way to the largely distributed print and electronic library. On the way from resource sharing to shared print, librarians are finding their way through many and dynamic uncertainties as they flesh out the ways and means for developing the shared collections inventory, a process involving substantial risk taking, experimentation with processes and relationships, research and data modeling projects, and cooperation not only among librarians but also the consortial groups, vendors, funding agencies, and service providers with whom they work. Shared print is the inevitable evolutionary outcome, however, of century-old programs of cooperative collection building and interlibrary lending as readers and researchers take advantage of new means for accessing and using the materials they need for their work. Shared print, together with such collective enterprises as CLOCKSS, HathiTrust, Digital Public Library of America, and the Digital Preservation Network, exemplifies the traditions of library cooperation as cooperatives become a true network, a system of mutual dependencies and potentials that makes sense of the endlessly proliferating similitude and differences among libraries. Although not easily and not until the library community and the organizations that fund it develop information and data systems and a sense of common purpose that dissolves some of the boundaries among current jurisdictions, shared print enables libraries collectively to ensure the availability of print publications in a systematic and orderly way. In a shared print environment, the collective becomes the local and the local the collective.
1 Thanks to Victoria Steele, Brooke Russell Astor Director of Collections Strategy at the New York Public Library, for sharing her knowledge of the plans at NYPL. This chapter is especially indebted to Constance Malpas, Lizanne Payne, and Emily Stambaugh; colleagues in the Maine Shared Collections Strategy; and the many librarians and consortial staff who have participated in the Print Archiving Community Forum at ALA these past four years. The chapter discusses a number of recent projects; it does not cover, let alone do bibliographic justice to, the many other projects, influences, and published or presented work that have laid the foundation and provide context for them.
2 The chapter concerns academic library collection practices and the reconfiguration they are undergoing. Except in the Maine Shared Collections Strategy and in the Minnesota Library Access Center (MLAC), public libraries have not played a role in shared print agreements, but see the “Conclusions” drawn by Lavoie, Malpas, and Shipengrover (2012) for discussion of the importance of their holdings in the collective collections picture.
3 See the case made for libraries becoming a collective institution in Council on Library and Information Resources (2008, “Introduction”); Association of Research Libraries (2012); Demas and Lougee (2011); Gherman (2007); Kenney (2009); and Kenney (2012).
4 According to Kieft and Payne (2012), “A ‘shared print’ agreement (also called ‘print archives’ or ‘shared collection management’) is a formal program in which multiple libraries coordinate long-term retention of print materials and related services by one or more participants to support preservation and allow space recovery among campus collections. A ‘shared print agreement’ is not the same as a shared storage facility. Rather, it is characterized by an explicit commitment to retain materials for a specified time period (or indefinitely) in potentially multiple locations by multiple partners.”
5 See Johnson (2009) and Evans and Saponaro (2012), textbook summae of the field that treat an array of historical, institutional, organizational, environmental, staffing, business, and procedural topics.
6 Today, the primary texts and practices of libraries’ collecting role are being rewritten, even though to many librarians, faculty, and students that rewriting is controversial. The case for onsite, open-shelf access has been made most strongly by Andrew Abbott, University of Chicago, whose views on the scholar’s need for very quick access to a large and heterogeneous corpus of sources were instrumental in the planning for the University’s Mansueto Library (Abbott 2006, 2008). Conflicts over the downsizing or off-site housing of large portions of the print collection at Ohio State University (Chronicle of Higher Education, September 1, 2006, May 13, 2009), California State Polytechnic University, Pomona (Chronicle of Higher Education, July 10, 2007), Syracuse University (Chronicle of Higher Education, November 11, 2009), and Augustana College (Chronicle of Higher Education, July 25, 2011) reveal, as does the conflict at NYPL, assumptions about the role of the library and how the location of library print collections affects the work practices of readers. See also John M. Budd’s discussion of physical collections and the role of their materiality in scholarly communication in his chapter, “The Collection(s)” (2012). In addition to these cases for maintaining open-stack, on-campus collections of printed works, especially of monographs, such writers as Carr (2010) make a case for print based on issues of cognition and the effects on learning of using networked devices and electronic text; Jabr (2013) summarizes the scientific literature on page and screen reading.
7 Interlibrary lending and resource-sharing programs have always positioned libraries as each other’s off-site collection-housing facilities through a de facto distributed or decentralized model of archiving. Moreover, the establishment of CRL in 1949, the creation of such materials partnerships as those of the Triangle Research Libraries in the 1930s, and the implementation of the Farmington Plan and PL480 or of the Federal Depository Library System prefigure, as collections cooperatives, current shared print initiatives. Shared print takes such antecedents to higher levels of intensity and a difference in kind, however, by addressing not well-defined groups of specialized materials but garden-variety, general collections volumes.
8 See, for example, the Web site of the Print Archiving Community Forum http://www.crl.edu/archiving-preservation/print-archives/forum for reports on best practices and consortial activities; the record of such preconferences as “Shared Print Archiving: Building the Collective Collection, and a Print Safety Net” held at the Charleston Conference in 2011 at http://www.crl.edu/node/7488, “Shared Print Monographs: Making It Work” at http://sustainablecollections.com/weed-feed/2013/4/3/alcts-shared-print-monographs-preconference-detailed-program.html at ALA Annual in 2013, or the 2012 ALCTS “Local Collections, Collective Context: Managing Print Collections in the Age of Collaboration” at http://www.ala.org/alcts/events/ala/ac/12/collaborationage. See also issue 37 (3–4) (2012) of Collection Management at http://www.tandfonline.com/toc/wcol20/37/3-4, which, in addition to reports on specific projects, includes an especially useful essay on collection plans by Demas and Miller (2012). Rick Lugg and Ruth Fischer of Sustainable Collections Services (http://sustainablecollections.com/), which has emerged as an important member of the shared print community with its analysis service, offer a “Why Now” case for deselection and a “Knowledge-base” on matters related to print collections on their Web site as well as news about their projects and conference sessions. Samuel Demas began editing a column in Against the Grain, “Curating Collective Collections,” in 2011, and Rick Lugg maintains a blog titled “Sample and Hold” (http://sampleandhold-r2.blogspot.com/); both constitute a compendium of commentary on current issues in print collections management.
9 The CIC journal project mentioned in the following paragraphs operates with a centralized model in which members support a facility operated by Indiana University. It is also important to note the role that grants from the Andrew W. Mellon Foundation have played in the development of PAPR for CRL and WEST and from Mellon and IMLS in other projects involving shared print over the years.
10 See Clement (2012) for a literature review of the development of regional shared print initiatives based in the history of cooperative collection development. In addition to Payne (2007), see Murray-Rust (2005) and Kohl (2003) for historical and synoptic treatment of collaborative materials housing facilities. New purpose-built materials housing facilities continue to appear at individual institutions, for example, Colby College (http://www.colby.edu/colby.mag/issues/63/article/1396/new-storage-facility-to-be-boon-to-colby-libraries/) and the University of Pennsylvania (http://www.library.upenn.edu/libra/), or on behalf of consortia, as is the case with the Florida Academic Repository in Gainesville (see mention in http://lj.libraryjournal.com/2013/04/opinion/editorial/kudos-for-print-archiving/), even as such consortia as WRLC and ReCAP add modules to their facilities. The libraries of the University of California system have discussed building a third Regional Library Facility (RLF), and to delay the arrival of the day when they must do so they have established a nonduplication policy for materials housed in the two existing RLFs. The Orbis Cascade Alliance has yet to be able to fund a long-discussed shared housing project. OhioLink members are deduplicating their five regional housing facilities in an effort to avoid building more of them. It is increasingly unlikely that new collective housing facilities will allow housing of multiple copies, especially of journals.
11 See http://www.aserl.org/wp-content/uploads/2013/02/ASERL-WRLC_Shared_Archive.
12 See http://www.crl.edu/news/8113.
13 See http://www.oclc.org/collection-analysis.en.html and the OCLC presentation linked on the Print Archiving Community Forum under the January 28, 2013, posts from ALA Midwinter, 2013, at http://www.crl.edu/archiving-preservation/print-archives/forum.
14 See http://www.oclc.org/services/projects/shared-print-management.en.html.
15 See http://www.crl.edu/archiving-preservation/print-archives/papr for a full description of the PAPR registry.
16 As several consortia are preparing to manage older, less-used monographs collectively, some are seeking ways to slow the unnecessarily duplicative growth of print collections by reaching agreements on the acquisition of new books and to maintain resource-sharing capacity by developing collectively owned eBooks. Space constraints do not allow consideration of these important dimensions of shared print, nor do they allow consideration of the effects on the collective collection inventory of DDA practices.
17 See http://www.crl.edu/node/7371.
18 See http://www.crl.edu/node/7372.
19 See www.cic.net/Libraries/Library/CICDocumentsFramework.sflb.ashx.
20 See Schonfeld and Housewright (2009a) for an important study of a potential future for the FDLP and Ragains (2010) for a succinct journalistic treatment of the debate about paper-copy retention and digital publication in the FDLP.
21 See http://www.aserl.org/programs/gov-doc/.
22 See http://www.maineinfonet.net/mscs/ for thorough documentation of the grant through its project plan, meeting notes and decisions, conference presentations, and so forth.
23 See http://mlc.lib.mi.us/cms/sitem.cfm/library_tools/mi-spi/ for project documentation, their MOU, and work with SCS.
24 Unpublished report.
25 See http://www.connectny.org/. Fall 2012 Update.
26 See Goals 2 and 3 of the GWLA strategic plan (http://www.gwla.org/strategic2012/Strategic-Plan).
27 See http://www.coppul.ca/projects/SPAN.html for the COPPUL program description; see also their presentation linked on the Print Archiving Community Forum under the January 28, 2013, posts from ALA Midwinter, 2013, at http://www.crl.edu/archiving-preservation/print-archives/forum.
28 Unpublished grant proposal.
29 Unpublished interview and online survey results from 2011 and 2012 by the present author of print collections planning and campus attitudes toward shared collections practices at liberal arts college libraries.
30 Some consortia have set “unnecessary duplicate thresholds” as guidelines for purchase or, by implication, retention of copies, for example, CARLI’s five copies for 140+ members in Illinois (http://www.carli.illinois.edu/products-services/collections-management/I-ShareNum-CopiesRe and the Orbis Cascade Alliance’s three copies for three dozen members) (see link for November 3, 2010, under Threshold Project at http://www.orbiscascade.org/index/cdmc-current-work). These numbers, along with the retention policies of the several single-copy depositories and Ithaka’s work on optimal copies (see, e.g., http://www.crl.edu/sites/default/files/attachments/pages/CRL%20webinar%202012%2001%2010%20Optimal%20Copies.pdf) are suggestive of movement toward, if not a consensus on, the smaller number of copies needed to satisfy collective demand or to archive the record of publication.
31 See presentation, “Shared Print Collections in North America: Going Main Stream and Picking up Steam,” by Lizanne Payne, consultant to ReCAP, at the MSCS meeting of May 23, 2013, linked at http://www.maineinfonet.net/mscs/about/presentations/.
32 In addition to reader preferences and practices, dislike of screen reading, device and software deficiencies, lack of shareability, and so forth, scholars make the case against the use of digitized copies based on the artifactual evidence multiple copies of the same printed book can offer for several kinds of history and the inability of one digitized copy to represent variations in production.
Abbott, Andrew. 2006. “The University Library.” Internal report. Available at: http://www.lib.uchicago.edu/e/about/abbott-report.html. Accessed June 1, 2013.
Abbott, Andrew. 2008. “The Traditional Future: A Computation Theory of Library Research.” College and Research Libraries 69 (6): 524–45.
Association of Research Libraries. 2012. “21st-Century Collections: Calibration of Investment and Collaborative Action.” Issue Brief. Available at: http://www.arl.org/storage/documents/publications/issue-brief-21st-century-collections-2012.pdf; and http://www.arl.org/bm~doc/21stctfreport_11may12.pdf. Accessed June 1, 2013.
Atkinson, Ross. 2005. “Introduction for the Break-Out Sessions: Six Key Challenges for the Future of Collection Development.” White paper. Available at: http://dspace.library.cornell.edu/bitstream/1813/2608/1/Atkinson_Talk.pdf. Accessed June 1, 2013.
Budd, John M. 2012. The Changing Academic Library: Operations, Cultures, Environments. 2nd ed. Chicago: ACRL, Publications in Librarianship.
Burgett, James, Linda Lucille Phillips, and John M. Haar. 2004. Collaborative Collection Development: A Practical Guide for Your Library. Chicago: ALA Editions.
Carr, Nicholas. 2010. The Shallows: What the Internet Is Doing to Our Brains. New York: Norton.
Clement, Susanne K. 2012. “From Collaborative Purchasing towards Collaborative Discarding: The Evolution of the Shared Print Repository.” Collection Management 37 (3–4): 153–67.
Connaway, Lynn Silipigni, Timothy J. Dickey, and Marie L. Radford. 2011. If It Is Too Inconvenient, I’m Not Going after It: Convenience as a Critical Factor in Information-Seeking Behavior. Dublin, OH: OCLC. Available at: http://www.oclc.org/content/dam/research/publications/library/2011/connaway-lisr.pdf. Accessed June 1, 2013.
Council on Library and Information Resources, 2008. No Brief Candle: Reconceiving Research Libraries for the 21st Century. Washington, DC: The Council. Available at: http://www.clir.org/pubs/reports/pub142/reports/pub142/pub142.pdf. Accessed June 1, 2013.
Council on Library and Information Resources. 2010. The Idea of Order: Transforming Research Collections for 21st Century Scholarship. Washington, DC: The Council. Available at: http://clir.org/pubs/abstract/pub147abst.html. Accessed June 1, 2013.
Courant, Paul, and Matthew “Buzzy” Nielsen. 2010. “On the Cost of Keeping a Book.” In The Idea of Order: Transforming Research Collections for 21st Century Scholarship, 81–106. Washington, DC: Council on Library and Information Resources.
Demas, Samuel, and Mary E. Miller. 2012. “Rethinking Collection Management Plans: Shaping Collective Collections for the 21st Century.” Collection Management 37 (3–4): 168–87.
Demas, Samuel, and Wendy Pradt Lougee. 2011. “Shared Print Archiving and Formation of a National Trust: Will Your Campus Participate in a Collective Pint Preservation Strategy?” Library Issues 31 (6).
Dempsey, Lorcan. 2012, December 10. “Thirteen Ways of Looking at Libraries, Discovery, and the Catalog: Scale, Workflow, Attention.” Educause Review. Available at: http://www.educause.edu/ero/article/thirteen-ways-looking-libraries-discovery-and-catalog-scale-workflow-attention. Accessed June 1, 2013.
Evans, G. Edward, and Margaret Zarnosky Saponaro. 2012. Developing Library and Information Center Collections. 6th ed. Santa Barbara, CA: Libraries Unlimited.
Fitzpatrick, Kathleen. 2011. Planned Obsolescence: Publishing, Technology, and the Future of the Academy. New York: New York University Press.
Five Colleges of Massachusetts. 2012. “Conference: Could the Bunker Be a NERD?” July 21, 2011, Mount Holyoke College. Unpublished Report. Available at: http://legacy.fivecolleges.edu/sites/libraries_council/documents/NERDConf
Summary.pdf. Accessed June 1, 2013.
Gherman, Paul. 2007. “The North Atlantic Storage Trust: Maximizing Space, Preserving Collections.” Portal: Libraries and the Academy 7 (3): 273–75.
Jabr, Ferris. 2013, April 11. “The Reading Brain in the Digital Age: The Science of Paper versus Screens.” Scientific American. Available at: http://www.scientificamerican.com/article.cfm?id=reading-paper-screens. Accessed June 1, 2013.
Johnson, Peggy. 2009. Fundamentals of Collection Development and Management. 2nd ed. Chicago: American Library Association.
Kenney, Anne R. 2009. “Approaching an Entity Crisis: Reconceiving Research Libraries in a Multi-Institutional Context.” Presentation, Distinguished Seminar Series, OCLC Research. Available at: http://www.oclc.org/research/dss/ppt/dss_kenney.pdf. Accessed June 1, 2013.
Kenney, Anne R. 2012. “From Product to Process: Collection Development in the 21st Century.” Presentation, 2012 International Roundtable for Library and Information Science, Kanazawa Institute of Technology/Council on Library and Information Resources.” Available at: http://staffweb.library.cornell.edu/system/files/Tsinghua-Nov%202012_AnneKenney.pdf. Accessed June 1, 2013.
Kieft, Robert H., and Bernard F. Reilly. 2009. “Regional and National Cooperation on Legacy Print Collections.” Collaborative Librarianship 1(3): 106–8.
Kieft, Robert H., and Lizanne Payne. 2011. “A Nation-Wide Planning Framework for Large-Scale Collaboration on Legacy Print Monograph Collections.” Collaborative Librarianship 2 (4): 229–33.
Kieft, Robert H., and Lizanne Payne. 2012. “Collective Collection, Collective Action.” Collection Management 37 (3–4): 137–52.
Kohl, David F. 2003. “Paper and Digital Repositories in the United States.” Library Quarterly 13 (3–4): 241–50.
Lavoie, Brian, Constance Malpas, and JD Shipengrover. 2012. Print Management at ‘Mega-Scale’: A Regional Perspective on Print Book Collections in North America. Dublin, OH: OCLC. Available at: http://www.oclc.org/content/dam/research/publications/library/2012/2012-05.pdf. Accessed June 1, 2013.
Lavoie, Brian, and Lorcan Dempsey. 2009. “Beyond 1923: Characteristics of Potentially In-Copyright Print Books in Library Collections.” D-Lib Magazine 15 (11–12). Available at: www.dlib.org/dlib/november09/lavoie/11lavoie.html. Accessed June 1, 2013.
Lavoie, Brian, and Lorcan Dempsey. 2010. “Rethinking the Boundaries of the Academic Library.” Next Space 17 (December 15, 16–17). Available at: http://www.oclc.org/publications/nextspace/issues/issue017.en.html. Accessed June 1, 2013.
Lavoie, Brian, Lynn Silipigni Connaway, and Lorcan Dempsey. 2005. “Anatomy of Aggregate Collections: The Example of Google Print for Libraries.” D-Lib Magazine 11 (9). Available at: http://www.dlib.org/dlib/september05/lavoie/09lavoie.html. Accessed June 1, 2013.
Lavoie, Brian, and Roger Schonfeld. 2006. “Books without Boundaries: A Brief Tour of the System-Wide Print Book Collection.” JEP: The Journal of Electronic Publishing 9(6): dx.doi.org/10.3998/3336451.0009.208.
Lewis, David W. 2013. “From Stacks to the Web: The Transformation of Academic Library Collecting.” College and Research Libraries 74 (2): 159–77. Available at: http://crl.acrl.org/search?author1=David+W.+Lewis&sortspec=date&submit=Submit. Accessed June 1, 2013.
Malpas, Constance. 2009. Shared Print Policy Review Report. Dublin, OH: OCLC. Available at: www.oclc.org/research/publications/library/2009/2009–03.pdf. Accessed June 1, 2013.
Malpas, Constance. 2011. Cloud-Sourcing Research Collections: Managing Print in the Mass-Digitized Library Environment. Dublin, OH: OCLC. Available at: http://www.oclc.org/research/publications/library/2011/2011-01.pdf. Accessed June 1, 2013.
Murray-Rust, Catherine. 2005. “From Failure to Success: Creating Shared Print Repositories.” Paper presented at the 12th ACRL National Conference. Available at: http://www.ala.org/acrl/sites/ala.org.acrl/files/content/conferences/pdf/murray
rust05.pdf. Accessed June 1, 2013.
OCLC. 2012. “OCLC Print Archives Disclosure Pilot Final Report April 2012.” Available at: http://www.oclc.org/content/dam/oclc/productworks/OCLCPrintArchives
DisclosurePilotFinalReport.pdf. Accessed June 1, 2013.
Payne, Lizanne. 2007. Library Storage Facilities and the Future of Print Collections in North America. Dublin, OH: OCLC. Available at: www.oclc.org/programs/publications/reports/2007–01.pdf. Accessed June 1, 2013.
Ragains, Patrick. 2010. “Fixing the Federal Depository Library Program.” American Libraries 41 (5): 36–38.
Reilly, Bernard F. 2004. “Preserving America’s Print Resources: Toward a National Strategic Effort Report on the Planning Day Discussions.” Focus on Global Resources 24 (1). Available at: www.crl.edu/sites/default/files/attachments/events/PAPRreport.pdf. Accessed June 1, 2013.
Reilly, Bernard F., and Barbara DesRosiers. 2003. Developing Print Repositories: Models for Shared Preservation and Access. Washington, DC: Council on Library and Information Resources. Available at: www.clir.org/pubs/reports/pub117/contents.html. Accessed June 1, 2013.
Schonfeld, Roger, and Ross Housewright. 2009a. Documents for a Digital Democracy: A Model for the Federal Depository Library Program in the 21st Century. New York: Ithaka S+R. Available at: http://www.sr.ithaka.org/research-publications/documents-digital-democracy. Accessed June 1, 2013.
Schonfeld, Roger, and Ross Housewright. 2009b. U.S Faculty Survey 2009. New York; Ithaka S + R. Available at: http://www.sr.ithaka.org/research-publications/us-faculty-survey-2009. See also the 2012 survey results. Available at: http://sr.ithaka.org/research-publications/us-faculty-survey-2012. Accessed June 1, 2013.
Schonfeld, Roger, and Ross Housewright. 2009c. What to Withdraw? Print Collections Management in the Wake of Digitization. New York: Ithaka S+R. Available at: http://www.ithaka.org/ithaka-s-r/research/what-to-withdraw. Accessed June 1, 2013.
Stambaugh, Emily. 2010. “Heading West: Circling the Wagons to Ensure Preservation and Access.” Against the Grain 22 (5): 18–22.
Conventional wisdom tells us that special collections will play an increasingly important role in academic and public libraries. The idea is that as the content of general library collections grows more and more similar through eBooks, e-journals, research databases, and consortial sharing, the rare and unique content found in special collections will grow in stature. Eventually, special collections will define and individualize the content and personalities of libraries.
Essentially, this idea makes sense. The personalities of special collections will become increasingly distinct as most general collections grow ever more standardized. At the moment, libraries are in a multimedia hybrid state. However, as general library collections are digitized or acquired electronically and access grows increasingly virtual, special collection libraries and departments will remain vital physical locations for library users. They will be a key part of the on-site service mission of many libraries.
Digital access is an essential growth area in special collections, but artifacts held in these collections often demand hands-on research. This can occur when the nature of the object does not allow for a suitable facsimile or scholars are studying the artifact as a material object and a facsimile simply will not do. A recent rise in the study of material culture across many disciplines has benefited special collection libraries, as scholars study books and other historical objects not only for their contents, but also as physical artifacts that manifest important evidence of the cultures that produced them. This has brought an increasing number of researchers to special collection libraries.
This rise in interest in special collections has been a part of a larger shift in the field toward greater access to rare and unique materials. This has been done through efforts to catalog once-hidden collections, better outreach to users, and the adoption of user-friendly policies. Perhaps most significantly, special collections are finding their way into more classrooms across a number of disciplines as librarians teach with these materials, whether in an academic setting or through programming for the general public.
With all of this in mind, the future of special collection libraries does seem to be one of great promise and increased prominence. In order to actualize this future, however, special collections need to be built with care. In addition to describing and promoting current collections, special collection librarians must build their collections in ways that establish their libraries as vital centers of programming, teaching, and research.
What follows are thoughts regarding the role collection development can play in ensuring the vitality of special collection libraries. From seeking out utility players and unique objects to building a library of record in a fringe or local field of study, this chapter explores how shrewd collection development will help guarantee that the conventional wisdom regarding special collections will actually come to fruition.
The future of special collection libraries depends not only on current collections but also on how these collections are grown. In a time where acquisition budgets might be small and the demand for material high, librarians must critically assess potential acquisitions, asking: What role will the item play in the library? Who will use it and how? Will it be used for teaching, and will it add to an existing research collection? Does the item have immediate research value? Might this item be worth digitizing and adding to our digital library? Might it be used in an upcoming exhibition? Does my library have a custodial responsibility to preserve this item? In short, how does this item expand our library’s mission?
Occasionally, an item expands a library’s mission in many ways. It fits the bill for research, teaching, exhibitions, and digitization. Moreover, it might be used for these purposes across a number of disciplines. If an artifact has all of this going for it, it is what we might call a utility player, to lift a term from the world of sports. That is, a player who is able to play several different positions for a team.
As an example, the Cary Graphic Arts Collection at Rochester Institute of Technology has in its collection a copy of a book printed ca. 1470 titled Incipit epistola lugubris et mesta simul et consolatoria de infelice expugnacione ac misera irrupcone et invasione insule Euboye dicte Nigropontus or “a mournful and miserable and at the same time comforting letter of the unfortunate assault and lamentable breaking into and invasion of the island Euboea called Negroponte.” This brief report describes the Ottoman Empire’s siege of the island of Euboea from Venice in July 1470. The primary reason this book is in the Cary Collection is that it might possibly be the first instance of printed news and thus, in a sense, a very early newspaper. The overarching mission of the Cary Collection is to document the history of graphic communication and so the book is a strong fit. The Cary Collection has a custodial responsibility to preserve the book, while providing access to it for research and teaching.
Epistola lugubris is ideal to show a class of journalism students who visit the library to learn more about the history of printed news. But the book has much more to offer. Due to the copy-specific information that has been added to it as it has traveled through the centuries, the book is a strong example of a utility player. For example, like many books printed in the first decades of printing in Europe, Epistola lugubris is a hybrid of manuscript and print. Most of what appears on the page is printed, but manuscript embellishment in red ink has been added to emphasize all the capital letters that appear on the page, thus bringing the reader to proper nouns and the beginnings of sentences.
The book’s second paragraph begins with a large space that has been left for someone to hand draw an initial letter “R” before the sentence beginning “—evere[n]dissimo i[n] Cristo patri ac clarissimo et spie[n]tissimo” If the “R” were added to the head of the sentence it would read “reverendissimo,” Latin for “most reverend.” Again, this hybrid of print and manuscript reflects the influence of manuscripts on early books. In other copies of Epistola lugubris, someone has added a manuscript “R” to the “—everendissimo”; the various previous owners of the Cary Collection’s copy chose to leave it blank.
More evidence demonstrating the influence of manuscripts on this work includes the lack of title page, table of contents, and pagination or foliation. These are bibliographic elements that became standard in books as printing continued to develop. For students of the history of the book, Epistola lugubris provides many examples of the printed book in its developing stages. After all, books printed before 1501 are books in their infancy, as indicated by the term given them—incunabula—which is Latin for swaddling clothing or cradle.
The Cary Collection’s copy of Epistola lugubris also features evidence of reader response, the kind of information that is crucial to the study of the history of reading. An early reader has marked many passages with brackets that look almost like grotesque faces. On two occasions, the reader has also drawn a hand pointing to a passage of some significance. Manuscript pointing hands, what bibliographers call manicules, are fairly common in early European books. Early readers often wrote in their texts, just as modern readers do today. Whereas we might use a highlighter or underline passages, a 15th-century reader might highlight important passages by drawing a manicule or some other mark. Manicules were so common in manuscripts that they became popular ornaments in wood and metal types, as well as in digital fonts (check your Wingdings). In this way, the discussion of Epistola lugubris could extend to the influence that manuscripts and printed books had on each other.
Another angle from which to approach the Cary Collection’s copy of Epistola lugubris is from the point of view of the conservation and preservation of historical artifacts. The previous owner of this book displayed it with bands holding the pages open to those showing the manicule and other examples of reader response. It is clear that this book was displayed in this fashion for too long a period in an environment in which the climate was not controlled. The result is that the pages darkened, except for two light strips along the outside margins where the bands had once held the book open. The book must have been in direct light or in a harsh environment, because it now carries with it these signs of its mistreatment. While there is not really a positive side to damaging an artifact, at least this book can now be used to demonstrate the damaging effects of displaying a book improperly. Hence, this might be appropriate to share with a class of museum studies students or maybe students engaged in book conservation.
Epistola lugubris is a utility player for the Cary Collection: an artifact that meets the demands of the library’s teaching, research, and custodial missions, while extending across a variety of academic disciplines. This kind of artifact is crucial to the future of special collection libraries. They are precisely what librarians and curators are hoping to find when they open bookseller and auction catalogs.
Accordingly, these are also precisely the kind of items for which antiquarian booksellers are looking. At a time when the growth of bookselling Web sites such as abebooks.com and alibris.com has made it easier than ever to find specific editions of books, utility players have become standout items for booksellers searching for something special to offer their customers. Booksellers and special collection librarians are also on the lookout for another very important class of items: those that are unique.
The notion of rare has changed considerably in the digital age. In the opening of this chapter it was noted how most general collections are becoming more homogenous as libraries have more resources in common and are sharing more resources. A similar trend can be found in special collections, although to a much lesser extent, due to increased numbers of digitized artifacts.
One question that a librarian might ask before acquiring an item is “Do my users already have access to it?” In the past a special collections librarian would have rarely asked this question. Depending on the mission, scope, and resources of a library, this still might be a question rarely asked. But for librarians who need to think very carefully about how to allocate limited acquisition funds, this question is certainly a consideration. Never before have rare and unique materials been so accessible, and this access is only going to increase. In other words, expanding digital access has rendered these objects less rare, at least in a virtual way.
For example, a researcher wanting to view the 1611 first edition of the King James Bible can see a PDF facsimile through Early English Books Online (if their institution is a subscriber) or a free high-quality digitized version from the University of Pennsylvania Libraries’ Schoenberg Center for Electronic Text & Image.1 If a researcher just wants to read the text of the King James Bible translation, many different Web sites can accommodate.
Does this mean that a library should not acquire a 1611 King James Bible? Of course not. For some libraries this book is an essential item to have in the collection. But for most libraries the need to acquire this particular item may be diminished. There is already access to it. It is very important to note that for those of us who study artifacts as material objects, facsimiles will not do. We need to work with the artifact itself. But a great many researchers simply need to see the text, study the image, or hear the sound. Increasingly this can be accomplished digitally, and thus, from an access point of view, the actual object becomes, in a sense, less rare.
Rare may have become less rare, but unique is always unique. A great portion of modern history has been built on information found in printed books. Our understanding of earlier history is dependent upon analysis of manuscripts and archaeological artifacts—artifacts that continued to be produced after the invention of printing but are sometimes not studied to the extent of printed books. There is so much of the historical record waiting to be discovered in manuscripts and other realia (i.e., three-dimensional historical artifacts) that have not yet been examined. These entirely unique items belong in special collection libraries.
Manuscripts have always been a staple of special collections, though they are sometimes collected less frequently than printed books. At a time when printed books are less rare, manuscripts may rise in prominence. They contain information that might not exist in other media. They hold historical evidence that might have been overlooked as a consequence of the focus on print. Manuscripts are sometimes more difficult to use due to paleographic barriers that stand between the reader and the text, but these barriers cannot be broken if manuscripts are not preserved, described, and made accessible to readers.
Realia is even less frequently represented in special collection libraries. Traditionally, libraries tended to consist of books, not things. Yet things often contextualize the cultures that produced traditional library holdings; therefore, they can be used to enhance existing collections in exciting ways. The Cary Graphic Arts Collection, for example, houses not only a library of books and manuscripts, but also examples of the equipment used to create the books and manuscripts. From a 15th-century woodcut, to the matrices used to cast Bruce Rogers’ Centaur type, to an Albion printing press that belonged to the American type designer Frederic Goudy, these materials not only make great teaching tools but also are themselves worthy of study. Moreover, under the right circumstance these materials could all be used again, replicating historical processes and, in turn, creating new examples.
The idea of unique extends to printed books, too. The most obvious case is when just one copy of a printed book survives. But there are other important ways that a printed book can be unique due to the distinctive cultural information it carries with it.
The notion of unique is not limited to manuscripts, realia, or single surviving copies of printed books. As books and other artifacts pass through centuries and decades, they absorb new information that later serves as evidence of the cultures that used them. For example, a book may carry with it important clues of its provenance indicating who used it and how. Association copies, or books formerly owned by someone associated with the book or by someone of note, are perhaps the most common example. For instance, 19th-century editions of Shakespeare’s works are exceedingly common, but consider the 1832 copy of Dramatic Works of William Shakespeare owned and annotated by George Eliot and George Henry Lewes that survives in the Folger Shakespeare Library.2 Such a copy clearly has a lot more to offer researchers than a clean copy of the same edition.
It is not just famous owners. Readers of all sorts leave historical evidence in the form of new bindings, bookplates, owners’ marks, manuscript marginalia reacting to the text, and damage, both from use and censorship. Rather than acquiring simply another copy of an edition, librarians and curators are increasingly seeking out artifacts that have unique, copy-specific information.
Surviving copies of John Foxe’s 16th-century protestant martyrology Actes and Monuments or Book of Martyrs sometime contain rich examples of reader response. A copy of the 1570 edition found now at The Ohio State University displays evidence of how one early reader inflicted intentional damage to the book in reaction to the often gruesome woodcut illustrations of executions.3 On woodcut after woodcut, this reader blotted with ink the faces of executioners and other men depicted as villains. Finally, when the reader reached the most notorious image in the book, a woodcut illustration of Bishop Bonner scourging the bare bottom of a prisoner, the reader defaces Bonner by stabbing out most of his face, creating a hole in the page. Thus, this copy of the Book of Martyrs manifests distinct evidence of how one reader reacted to the book’s inflammatory contents. Its contents are invaluable to researchers of the Protestant Reformation.
This sort of unique information is not only found in books (printed, manuscript, or otherwise), but also on most historical objects ranging from tools to machines to currency. Unique artifacts and artifacts containing unique information are of increasing interest to librarians and booksellers. For booksellers such items are more easily placed in new homes and often at higher prices. Consequently, prices may become obstacles to acquiring unique items and utility players, but special collection librarians must think big and work out ways to bring such items into their libraries.
In his Taste et Technique in Book Collecting, John Carter writes “be less afraid of paying a stiff price than of letting slip some book you know to be rare and which is important to you. You cannot tell when, at what price, or even whether, you will see another” (1949, 136). In short, whenever possible, price should not get in the way of an essential acquisition.
Okay, this probably sounds like an impossible way to approach acquisitions, particularly in a time of limited resources. Perhaps it is unrealistic, but the core of the idea is true. Libraries should not let big items slip away. A big item probably falls into the categories discussed earlier: a utility player, a unique item, or an artifact with immediate research value. Passing on such an item because of price means this item might never make it through your library’s doors. Yet, librarians might pass on one major item because of its cost, and then spend the same amount on a handful or more other items. Although they might feel like they were thriftier with their budget, have they really gotten more value for their money? The question is whether the research value found in the handful of items equals the research value of the one big item. Has the library lost out on a critical opportunity?
Obviously there is value in purchasing certain types of materials while they are still affordable; many times it is a good way of building for the future. But this should not be done at the expense of items that belong in your collection but are simply more expensive. There is often a reason why big items are more expensive. Librarians should not be afraid to spend significant parts of their acquisitions budgets on a few major purchases. Say that if in a year you have bought only two new items for your collection, you might feel like you have not been doing your job completely or that you might have missed out on other potential acquisitions. But the greater question is whether you have truly enriched your collection.
Such important items might not even be purchased through the usual acquisitions budget. If budgets cannot get you what your collection requires, then creative solutions need to be found. High-end items might be enticing enough to whet the appetite of donors and friends of the library. Fundraising for such an item can be done prior to the purchase, but also afterward, through an adopt-a-book event. Looking beyond your local community, you may find a foundation or two to which you could apply for support.
Beyond budget challenges, librarians might also fear their administrators will look askance at the purchase of just a few items rather than what they might view as a more healthy number of items. This should not be seen as a source of potential conflict, but as an opportunity for education and outreach. Whenever possible, major acquisitions should be accompanied by outreach.
Big items should not arrive at a special collections library without some amount of outreach. If the item was purchased because of its research value, then librarians should alert those scholars for whom the item will be helpful. If the item was purchased because of its relevance to various classes, then the librarian should reach out to the appropriate teachers. Joel Silver, the curator of the Lilly Library at Indiana University, recently wrote: “While previous generations of rare book librarians tended to focus on the building of their collections above all else, today it is access to and communication about the collections that occupy a greater portion of librarians’ time” (Galbraith and Smith 2012, viii). This observation extends to all special collection librarians. The time of collection development independent of outreach has passed. In prior years the primary focus on collection building was extremely important, but it also led to major backlogs in processing and cataloging material. This, consequently, led to the growth of hidden collections. Hidden collections are effectively nonexistent collections: if your materials cannot be found in your online catalog, they do not exist to most users.
Over the past decade we have seen a movement to make hidden collections in special collection and archives visible. Building on this momentum and thinking beyond just backlog cataloging, librarians should view outreach as an essential part of collection development. Spending significant portions of a budget on materials demands the provision of support for getting these materials into the right hands. Time and resources needed to promote new acquisitions on places like Web sites, and appropriate listservs need to be a part of the acquisitions workflow. A portion of the acquisitions budget could be earmarked for the cost of promotional material or a welcome event. In 2008 when the British Library acquired a remarkable copy of Everard Digby’s A short introduction for to learne to swimme, better known was the Art of Swimming, that had been hand colored and filled with humorous manuscript annotations, the library held an event to introduce this exciting new acquisition to its community.4
New acquisitions should also be defined broadly to include newly discovered or rediscovered items already in a library collection. Although it seems strange that a new item might be found in a library that has housed it for decades or longer, those in the field can testify to exciting discoveries within their own walls. These important discoveries require outreach to potential users. There are also times where known artifacts should be reintroduced to users through outreach. Researchers may not always know the extent of resources available to them.
In addition to simply doing right by the artifacts, bringing attention to acquisitions could also be a matter of survival. Your administrators need to know about the libraries and collections they are funding. They have to understand the mission of the organization and how new acquisitions support and expand that mission. Successful outreach regarding collection development will help build a library’s reputation beyond its walls, letting the greater community know the library’s value.
Eventually outreach supporting a library’s holdings should serve to define its collection as a library of record in one or more fields. This sometimes means being creative—thinking beyond traditional collections and exploring the fringe.
The nature of what content constitutes special collections or archives has broadened. What was once a fairly traditional definition of what was archive-worthy has rightly expanded to include what might be considered fringe materials. The term fringe is borrowed from an excellent seminar held at the 2012 Annual Preconference of the Rare Books and Manuscripts Section of the Association of College and Research Libraries titled, “From Dungeons to Dragons: Collecting, Processing, and Accessing Fringe Formats.”5 The seminar examined the challenges of managing popular culture collections such as the Edwin and Terry Murray Collection of Role-Playing Games at Duke University, the UT Videogame Archive at the University of Texas at Austin, and the Eaton Collection of Science Fiction & Fantasy at University of California, Riverside. Each of these archives houses collections that might be considered nontraditional, though each represents a dynamic field of study.
In this way, they are not fringe collections at all, but important material that reflects the societies that created them. While role-playing and video games might seem trivial to some, they carry important cultural information. What will future generations learn about our present generation through the games we play? A great deal, if someone has the foresight to preserve our games for later study and use.
Foresight is key. Most collection development in special collections is guided by existing collections and the needs of local communities. Thus, most of the items that librarians acquire are predictable choices. They are often cultural touchstones. Exploring the fringe takes librarians in a different direction. It requires some degree of adventure and risk. Librarians must raise their cultural antennae and tune into the sort of artifacts that might be flying under the radar at the moment, but may have value further down the road and thus deserve to be preserved in an archive. Preservation is important. Often these types of artifacts are ephemeral and in danger of being lost.
Not only does building a fringe collection preserve what are often ephemeral materials for later study and use, but it also establishes a library of record for unconventional topics. In turn, the libraries that house them carve out a niche for themselves that distinguishes them from their peers. Unconventional libraries also make for interesting human-interest stories. When the University of California, Santa Cruz sought an archivist for its Grateful Dead Archive, the story was picked up widely by the mainstream news media and even landed on the desk of The Daily Show’s Jon Stewart.6 Mockery was made, though lighthearted, and in the process millions of Americans were introduced to the Grateful Dead Archive.
Librarians do not always need to travel out to the fringes to find unconventional special collections. Traditionally, many simply looked to their communities to identify materials that reflected their own local personalities and histories. The New Bedford Whaling Museum in New Bedford, Massachusetts, for example, holds an amazing assortment of whaling artifacts, including the world’s largest collection of American whaling logs and scrimshaw. These are natural fits. In the 19th century, New Bedford was a major center of the whaling industry in America. Thus, the artifacts preserved within the museum and its library document a significant and fascinating part of the city’s history.
Thinking locally, however, does not always mean collecting a topic directly related to local history or culture. The Providence Public Library preserves the second largest collection of whaling logs, together with a group of manuscripts, scrimshaw, and other whaling artifacts. Providence was never a major whaling city and so these materials do not have the same sort of historical connection that they would in New Bedford. The Providence Public Library’s focus on whaling and maritime history is built on a collection donated in 1956 by a Providence businessman named Paul C. Nicholson. In this way, a local benefactor’s interest in whaling led his public library to become a research center for whaling history.
Building a special collection into a research-level collection in a subject or several subjects will define the library to its local community and beyond. Returning to the conventional wisdom regarding the increased prominence of special collections, there is no better way of defining the character of a library than establishing it as a library of record for exciting areas of study based upon its rare or unique holdings. The library may not make it onto The Daily Show, but it will attract the attention of teachers, students, and scholars, who will in turn come to that library for primary research material and teaching resources.
Primary research with special collections will continue to be conducted and taught as a methodology. Some of it will take place online, but a great deal of it will continue to take place in the reading rooms of special collection libraries and archives. Special collection libraries must continue to meet the needs of their users through smart collection building. This includes embracing utility players, seeking out unique items and other items with copy-specific information, being willing to find ways to acquire expensive materials so as not to pass up on artifacts that are vital to a specific collection, and establishing new collections based on material that might be considered unconventional or on the fringe. For all of these new acquisitions (and for those already in the archives), outreach is necessary to ensure they are visible to those who need them.
The most exciting part of all of this is that just about any library can build a special collection.7 General research collections are very expensive to build and difficult to maintain. They are developed through commercial networks of vendors. Special collections certainly require funding, but they are usually built through the generosity of donors, whether individuals, corporations, foundations, or grant-funding agencies. They are established through creative and cooperative relationships among people with shared intellectual and artistic interests. This means that any library can help bring conventional wisdom regarding special collections to fruition. If we have all successfully done our share to make it so, more and more libraries and archives will continue to transform into vital intellectual centers.
That is something to think about.
1 http://sceti.library.upenn.edu/sceti/printedbooksNew/index.cfm?TextID=kjbible.
2 http://shakespeare.folger.edu/cgi-bin/Pwebrecon.cgi?BBID=116816 and http://www.folger.edu/html/exhibitions/thys_boke/ordinarybooks.asp.
3 John Foxe, The First [-Second] Volume of the Ecclesiasticall History Contaynyng the Actes and Monumentes of Thynges Passed in Every Kynges Tyme in This Realme, Especially in the Church of England . . . (London: John Daye, 1570), 782. BR 1600 .F6 1570 vol. 2.
4 Everard Digby, A Short Introduction for to Learne to Swimme. Gathered Out of Master Digbies Booke of the Art of Swimming. And Translated into English for the Better Instruction of Those Who Vnderstand Not the Latine Tongue (London: James Roberts for Edward White, 1595). https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=lis-rarebooks;375bd8fe.0808.
5 http://www.rbms.info/conferences/preconfdocs/2012/2012docs.shtml.
6 http://www.thedailyshow.com/watch/wed-november-11-2009/want-ads---grateful-dead-archivist.
7 I need to thank my brother James Galbraith, the wiser of the two Galbraith librarians, for driving this vital point home for me.
Carter, John. 1949. Taste et Technique in Book-Collecting: A Study of Recent Developments in Great Britain and the United States. Cambridge: University Press.
Galbraith, Steven K., and Geoffrey D. Smith. 2012. Rare Book Librarianship: An Introduction and Guide. Santa Barbara, CA: Libraries Unlimited.
While libraries and cultural heritage collections have been struck by disaster for centuries—from the fire at the Alexandria Library to the Florence Floods of 1966, the onslaught of disasters affecting our written history and artistic heritage seem to have greatly intensified over the past 20–25 years, especially in the United States.
Beginning with Hurricane Hugo and the Loma Prieta Earthquake, both of which happened in 1989, and extending through the wide swath of devastation caused by Hurricane Sandy in late October 2012, our print and increasingly digital collections have faced many natural and manmade disasters, which affected large geographic regions.
Slowly but surely, the library and cultural heritage communities are creating institutional disaster plans. Many organizations are benefitting from working with local first responders such as police and fire departments, as well as emergency management agencies, in the development of their disaster plans. Once an institution has moved forward to complete its institutional plan, it can then turn to its community for additional support—and to provide additional support to its community. One way to combat large, city-wide, regional, statewide, or multistate disaster events, and the widespread destruction in their wake, is through collaborative disaster planning among multiple libraries or across cultural institution types, working with local emergency managers.
Beyond Hugo and Loma Prieta and their effect on libraries and museums, the 1990s saw flooding and wildfires, and the first decade of the 21st century included disasters causing tremendous amounts of damage to cultural heritage collections.
• Tropical Storm Allison, in June 2001, caused flooding in the greater Houston metropolitan area, which resulted in water damage to collections ranging from theater costumes to master film reels, from photographic collections to rare books and special collections materials.
• The terrorist attacks of September 11, 2001, in addition to a terrible human toll, destroyed materials from books and archival records to outdoor sculptures.
• In August 2005, Hurricane Katrina caused widespread damage to public libraries, municipal archives, museums, and historical homes in Louisiana, Mississippi, and Alabama.
• In August 2011, Hurricane Irene created a wide swath of damage to cultural institutions on the East Coast, from North Carolina to Vermont.
• October 2012 saw damages to library and museum collections in New York, New Jersey, and other East Coast locations due to Superstorm Sandy.
Unlike many manmade emergencies caused by building construction accidents and system malfunctions, these were large-scale disasters affecting multiple institutions or even whole regions of the United States. Although individual–institution disaster plans offered some assistance in these events, collaborative disaster planning among groups of institutions has proven to be even more effective. And, collaborative disaster planning is not just for catastrophic events, which actually happen very infrequently; this type of disaster planning approach can help when there is a disaster just at your institution. For example, if a pipe bursts over collections at your library, other institutions can help provide supplies to pack out wet materials, and can even provide trained people to assist, if previous collaborative work has included disaster response training.
Many national and regional organizations began to support the concept of collaborative disaster planning in the 1990s and 2000s.
Heritage Preservation, the leading preservation advocacy organization in the United States, led the charge by establishing the Heritage Emergency National Task Force in 1995. A collaborative effort of 42 cultural heritage organizations and associations, federal agencies, and national service organizations, the Task Force was a key collaborative in preparations for and recovery from incidents such as Hurricane Katrina and Superstorm Sandy, in part by offering a way in which affected institutions can communicate with organizations that could provide recovery information and assistance.1
The American Institute for Conservation of Historic and Artistic Works (AIC) developed its Collections Emergency Response Team in 2005, partially in response to Hurricane Katrina. This group provides assistance ranging from a disaster response telephone and email hotline to matching conservation professionals and specialists with organizations needing assistance to recover specific types of artifacts.2
Regional Conservation Centers such as the Conservation Center for Art & Historic Artifacts (CCAHA), Northeast Document Conservation Center (NEDCC), Midwest Art Conservation Center (MACC) and regional library networks such as LYRASIS (the merger of SOLINET, PALINET, NELINET, and members of BCR) and Amigos Library Services provide disaster assistance training and information as part of their preservation field services operations, funded in part by the National Endowment for the Humanities’ Division of Preservation and Access. WESTPAS—the Western States and Territories Preservation Assistance Service—is another key group that primarily focuses on disaster assistance for many states west of the Mississippi River.3
On a local and regional basis, other approaches began to appear:
• Mutual aid agreements, where individual organizations signed memoranda of agreement, which pledge staff resources, supplies, and other support between member institutions in the event of a localized emergency or more widespread disaster.
• Multi-institutional circuit-riding recovery assistance where an expert within a state or region travels to assist organizations with disaster planning or recovery activities. Examples of this approach exist in Massachusetts, Iowa, and Missouri, as well as many other states.
• Disaster Supply Caches, where multiple organizations work together to develop stockpiles of disaster supplies. These caches save individual institutions’ money through joint purchase and management of supplies. A further explanation of this disaster planning and recovery approach is described later.
• Libraries and cultural organizations that have a sister institution in another region of the country, which can provide communication support if your organization has to evacuate, and have even in some cases provided data backup support when there is an area-wide disaster.
These national, regional, and local multi-institutional disaster networks provide a wide variety of benefits to their members.
Ensuring that your library has an institutional disaster plan, which covers the specifics of your collection formats, collection storage, and building design, is the cornerstone preservation policy on which every library needs to build. Writing and practicing a plan, at least once a year, can protect your collections.
When disasters strike, especially those with the potential to damage a large number of buildings, such as hurricanes, wildfires, and tornadoes, a plan to protect multiple library buildings or branches, as well as neighboring cultural heritage institutions (e.g., museums, historical societies, and archives) should be in place and ready for activation.
The benefits of cooperative or collaborative disaster planning are many. An institution can save money, better utilize its resources, gain information and training, and deal with insurance and technical issues.
When disaster strikes, gaining access to your collections within the first 48 to 72 hours after they are initially damaged is important in allowing you to recover materials before water damage or mold causes further problems for the items.
Working on a local or regional basis and coordinating with other cultural heritage institutions can make recovery easier. Whether allowing you to offer services through reciprocal borrowing, shared service delivery, or other means, collaborative activity has meant speedier business resumption by libraries and other cultural heritage organizations. And in some cases, working with partners to find sites for relocation during long-term recovery can also help to leverage resources, resume service sooner, and deal with technical and insurance issues. Providing emergency preparedness information and disaster response and recovery best practices to a large group of organizations at the same time via group education and information puts participating collaborative members on a level playing field.
Educational efforts for a disaster collaborative can build familiarity in many ways. Planning needs to start with collaboration—getting to know your neighboring cultural heritage institutions—and first responders—in your community, and developing from there.
After developing their own disaster plans, organizations may want to consider sharing their disaster documents and floor plans, and storing them in a safe place in case their partner institution needs access because originals are lost or damaged.
Networking with peers on disaster plan development, demonstration and practice of wet material recovery, and tabletop disaster scenarios where group members assume recovery roles and work through a disaster timeline can be helpful in making sure all members of a disaster collaborative have baseline preparedness, response, and recovery skills.
Initial training, just as with initial disaster plan development, is an important activity. Just as important can be a novel opportunity to practice and update the plan. Pick a day each year for your group to learn about changes in each other’s building collections and policies. One suggested time of the year to practice disaster plans on a group basis is MayDay, the date of a national campaign to promote emergency preparedness.4 And, consider bringing first responders, emergency managers, and commercial disaster recovery vendors in for group meetings to demonstrate their services and answer questions (this approach also works with group insurance representatives).
In this era of tight library budgets, saving precious library funds through coordinated disaster planning may allow those monies to be spent on other important programs and services.
An initial area where some collaborative disaster planning groups have saved money is through receiving group training and consulting. This might include an outside expert visiting each of the collaborative sites to assess disaster risks, and then providing training specifically designed to address the highest-ranked risks in the region.
Consortial contracting is another potential area of fiscal savings. Developing a coordinated contract with commercial disaster recovery vendors, which will make cost containment easier for you and capacity and coverage issues easier for the vendor, is one beneficial approach. Additionally, consortial or group insurance contracts (which many county, municipal, or higher education institutions already have) can help to spread costs over multiple organizations and may bring down prices on premiums.
Another tact which has shown demonstrable costs savings is the development of centralized disaster supply caches. Rather than individual cultural heritage institutions spending a large amount of money to develop kits of disaster recovery supplies just for itself, that may never be used, the development of a larger, shared cache of disaster supplies which can be accessed by any member of a collaborative disaster planning group is a proven money-saving approach for cooperative disaster planning. Examples of these caches are housed in locations from southern California to Pittsburgh/Western Pennsylvania. The caches are monitored and managed by collaborative groups, and available to members 24/7/365.
Collaborative disaster planning and recovery efforts can provide for efficient utilization of human resources as well as financial resources. Certainly the training mentioned earlier, if offered on a collaborative basis, can provide a greater number of trained responders from the cultural heritage community on a local, state, or regional basis who can group together to assist an individual institution or cultural organizations in a city or area struck by a disaster. In some cases, the responders from the library and museum communities may have specializations such as dealing with computer and technical equipment, or with difficult formats ranging from audiovisual materials to paintings and works of art on paper. Creating emergency response teams to aid other institutions requires specialized training and coordination, which is different from or in addition to writing and practicing disaster plans or conducting risk assessments. Ideally, the preparation for joint response would include local Community Emergency Response Team (CERT) training, a program developed by the Federal Emergency Management Agency and made available through many county Emergency Management Agencies (EMAs).
The other resource-based benefit which a number of disaster collaboratives have achieved is through touring each other’s facilities to gain familiarity with other institutions in their consortia. Through these tours, you may be able to determine your partner institutions’ collection strengths, staff strengths, and building weaknesses. Including facilities staff on these tours can be helpful in determining systems strengths or weaknesses, and having first responders involved in this process can ensure that they are familiar with your institution’s building, floor plans, and contents before they respond to an emergency.
A number of insurance companies and customers within the cultural heritage industries have noted exponential growth in the levels of premiums during the 2000s. The 9/11 and Hurricane Katrina are key factors in these rate increases, and it is expected that Hurricane Sandy may cause price increases as well.
There are a number of things institutions and collaborative groups can do to potentially reduce premiums. Identifying high-probability risks—both inside your building and in the external environment (are there railroad lines, wooded or grassy areas, or industrial plants nearby?) is one way to determine disaster risk and probability levels. Then, determine what risks can be addressed. The Heritage Preservation’s Risk Evaluation and Planning Program (REPP) initiative pairs representatives of individual libraries, museums, or other cultural heritage institutions with a local first responder to assess and rate the possibility of occurrence of disasters related to specific targeted risks due to the institution’s building and geographic location. REPP began as a pilot program in 2008–2009; the tools to complete this assessment remain online.5
Additionally, having a current and complete collection inventory, and considering a collection valuation to determine the worth of your holdings are important steps. Finally, working within a disaster planning group may allow you to take advantage of consortial insurance policies and increase your buying power.
In addition to becoming familiar with each other’s collections, staff, and buildings, work collaboratively to review your disaster planning partner institutions’ safety systems—such as smoke, heat, and fire detection, fire suppression, and building security systems. Individual libraries can note improvements they might want to make as a result of seeing other systems within their cooperative group.
The rapid growth of electronic records, electronic library resources, and digitized materials is another facet of cultural heritage collection stewardship that must be considered in disaster planning. Redundant, geographically separated sites for storage of digital collections and electronic records or resources have also grown in importance since the time of Hurricane Katrina. Shared, distributed digital preservation services such as the MetaArchive allow multiple copies of digital collections to be distributed and preserved among cooperating libraries.6
Finally, consider utilizing the concept of a computer hot site, where replica systems and services to what you currently have can be made available to allow resumption of service.
As mentioned earlier, reciprocal borrowing, where another library system can allow your patrons to borrow materials with their current library card or ID, has been a popular method for damaged libraries to resume service since the mid-1990s.
Another large issue to deal with in today’s library environment is fulfillment of ILL requests. If your library is closed due to power outages, damage, or other concerns, you may not be able to locate and distribute ILL materials as usual. Working with collaborative disaster recovery partners and your ILL Service Provider may be a good way to keep up your lending and borrowing practices at a time of crisis.
A final area to consider may be thought of as too much of a good thing. Many libraries damaged by flood and fire have noted the generosity of other libraries and even patrons, in donating materials for these organizations to put back on their shelves. Having a strong collection policy that outlines what an institution is, or is not willing to accept as a donation is important in bringing your collection and services back to full strength. However, what if the materials you receive as a donation after a disaster are outside your library’s normal collection development scope? An example is the Academic Research Library that received romance novels to fill its shelvesl after its collection was badly damaged by flooding. One solution is cooperative donation processing centers, which have been utilized to ensure that libraries are replenishing their shelves with appropriate materials.
One of the most important features of collaborative disaster networks, as they have matured, is the establishment of relationships between cultural heritage staff members and representatives of the emergency management and public safety communities.
It is extremely important to develop these relationships for a number of reasons. Local emergency managers and responders (fire, police, city or county emergency managers) are in charge during an event, and will take control of a fire, crime, or emergency situation until the immediate danger is controlled. Any response activities by a library, museum, or historical society must be coordinated and integrated with public safety and emergency management representatives. Before an emergency strikes, these partner organizations can assist cultural institutions with risk assessments, planning, and pre-disaster mitigation issues (such as suggesting how to secure or barricade windows if high winds are imminent). And, in recent years, more and more libraries are being designated as providers of essential services for their communities (e.g., computer and telecommunications centers for both responders and disaster victims), which means they can get recovery assistance sooner.
These relationships start at the local level, but there are also many important cooperative disaster planning relationships being built at the top national administrative levels. The Federal Emergency Management Agency (FEMA) was a cofounding member of the Heritage Emergency National Task Force in 1994. Through its Office of Environmental Planning and Historic Preservation, FEMA has a group of staff specialists who can focus on cultural heritage recovery when disaster strikes.
FEMA, as well as state and local emergency managers, interacts with cultural heritage staff on a local or statewide basis through the Alliance for Response initiative. This program, with active groups in more than 20 cities and regions, provides a way to build bridges between the emergency response and cultural heritage communities before a disaster happens. Following initial Forum meetings where the groups first meet together, continuing programs of education and information exchange have provided invaluable linkages in disasters such as Hurricane Sandy and the Colorado Wildfires of 2012.7
One of the products of the Heritage Emergency National Task Force and the Alliance for Response Initiative is a poster on working with emergency responders before, during, and after an emergency. Including steps such as identifying emergency responders, communicating and building a relationship with them, and providing them with the information they need to know, this publication can assist cultural heritage institutions in disaster preparedness outreach to the most important allies they can have in an emergency or disaster situation.8
Finally, a program which has been implemented in several states (Massachusetts, Rhode Island, and Connecticut as of late 2012) is the Coordinated Statewide Emergency Preparedness program, or COSTEP. A “planning tool designed to bring together cultural resource institutions with emergency management agencies and first responders,” it provides “a blueprint for preparing for area-wide disasters and building alliances with federal, state, and local emergency management agencies” COSTEP “guides states through the process of planning for a disaster and fosters collaboration among a wide range of agencies” (COSTEP n.d.).
With all of the benefits and resources which collaborative disaster planning and recovery networks provide, libraries and cultural heritage organizations should definitely explore working together with other local cultural entities, emergency managers, and first responders. These resource groups can become allies that can help to save buildings, collections, and, most importantly, the lives of staff and patrons.
1 For information on the Heritage Emergency National Task Force, consult the Task Force’s Web site http://www.heritagepreservation.org/programs/taskfer.htm.
2 To learn more about the AIC Collections Emergency Response Team, consult its Web site http://www.conservation-us.org/publications-resources/disaster-response-recovery/aic-cert#.UnAFZBBGbl8.
3 More about these regional conservation and preservation centers can be found at the Web site http://www.rap-arcc.org/rap-members.
4 For additional information on the MayDay campaign, consult the Web site http://www.heritagepreservation.org/mayday/.
5 To learn more about Heritage Preservation’s Risk Evaluation and Planning Program, consult the Web site http://www.heritagepreservation.org/REPP/.
6 More about the MetaArchive Cooperative can be found at the Web site: http://www.metaarchive.org/.
7 To learn more about these events, consult the Web site http://www.heritagepreservation.org/AFR/index.html.
8 This poster, “Working with Emergency Responders—Tips for Cultural Institutions,” can be found at Web site: http://www.heritagepreservation.org/lessons/HPR_Emergency_Poster8.5x11.pdf.
“COSTEP: Coordinated Statewide Emergency Preparedness.” n.d. Available at: http://www.nedcc.org/free-resources/costep. Accessed May 17, 2013.
L. Suzanne Kellerman INTRODUCTION This chapter discusses digitization projects as a collection development strategy, and according to the Penn State University Libraries’ Digital Collections Review Team’s Digital Toolkit, a digital project can best be defined as “a series of collaborative activities that transforms analog materials into a digital format. These activities include selection, organization and planning, preparation, description and access, production, digital preservation and assessment, evaluation, and publicizing the digital resource. A digital project may involve creating new content, organizing access to existing electronic content, or digitizing existing historic collections.”1 Possible items for digitization could include bound text, unbound sheets such as maps or works of art, still and moving images, data, and three-dimensional objects. The size of any digital project may range from several sheets, as in a single mining map, to upwards of thousands of pages or more, as could be found in a large archival manuscript collection, or it may be a single sound recording or a collection of oral histories found on reel-to-reel tapes. The goal of digitization, and in this case a digitization project, is to improve sustainable access to the materials, to preserve materials by reducing the handling of originals, and ultimately for the digitized materials to be searchable on the Internet. For decades now libraries have been in the forefront of using computer-based technologies to describe, manage, and enhance access to collections. Advancements in digital technology tools since the mid-1990s have revolutionized our thinking about collections—giving historic hidden collections new life and users the ability to discover them anew (Lynch 2009). In recent years libraries have turned to mass digitization partnerships with Google and other consortia such as the Open Content Alliance to cost-effectively digitize large portions of their collections—digitization that would be too costly for an institution to take on by itself (Conway 2010). Moreover, with each passing year new digitization vendors and service bureaus have emerged giving libraries a much improved competitive marketplace to purchase digitization services. Regardless of the digitization approach, whether produced in-house, outsourced to a vendor, or partnering with a mass digitization agent, libraries and institutions of all sizes are either contemplating, planning for, or actively being engaged in some sort of digitization initiative. But before jumping on the proverbial digital bandwagon, there are many factors one needs to consider and decisions to make before scanning begins. In this chapter I will bring to light, through a series of questions, key decision points in the process that you will need to ask yourself before embarking on a digitization project. With 20 years of digital project experience, I will share resources, discoveries, and lessons learned along the way to successfully convert materials to digital form as a collection building strategy. SELECTION Choosing the right collection is the key to a successful digital project. Making good selection decisions at the outset of the project is time worth taking. The goal is to build an online product that is useful and useable. It is critical to keep the audience in mind, give them what they will use, and build a product that will bring something new to the collection by adding value. Remember digital projects are labor intensive and are a significant institutional investment. Once the content goes online, it requires perpetual care, continuing management, and periodic evaluation. System upgrades, the addition of new content, strategies for long-term preservation of the digital assets, content and system migrations, and the general housekeeping of the online site to keep it fresh and aesthetically appealing are all necessary ongoing activities for any digital project. When selecting materials for a digital project, consider the following questions: • What are the goals of the project? Does the project support your institution’s mission or strategic planning goals and objectives? • Who is the intended audience? Is there a specific user group targeted? Students, general users, researchers, scholars? • How do you envision the final product? Are there value-added features to bring to the project such as incorporating teachers’ guides to support classroom use, providing active links to similar collections, inserting geographic access reference points, adding blogging or other interactive publishing options, enhancing search capabilities or even adding crowdsourcing options for OCR (optical character recognition) text correction? • Are the items in the collection scannable, or do they require preservation stabilization? • Will the entire collection, a subset of the collection, or a sampling of the collection be scanned? • Have you conducted a thorough Internet search of your targeted collection to determine if it has or has not already been digitized? • Are there potential collaborations or resource-sharing partnerships with other organizations or institutions that have similar collection strengths? COPYRIGHT Copyright is the bane of most digital projects. It can derail a digital project instantly unless the necessary steps are taken to determine whether you have the legal right to make digital surrogates and disseminate the images. Whether the collections are published works or unpublished works hidden in your archives, carefully weigh the following questions as part of the selection- and decision-making process. • Do you own or hold the legal rights to the content you want to digitize? • Is there a donor’s deed of gift document that you may need to review first to determine your rights? • Is the content in the public domain, that is, pre-1923 imprint? • Is it an orphan work, that is, works whose copyright holders cannot be identified or found? • If you do not hold the rights, are you willing to seek the necessary permissions? To help you make knowledgeable decisions regarding copyright issues and your rights to legally preserve published and unpublished works, arm yourself with the facts. Following are selected resources available to assist you navigate the legal implications of your digital project idea. USEFUL LINKS The American Library Association’s digital slider (slide rule) for works published in the United States—http://www.librarycopyright.net/digitalslider/. Introduced in 2008, this tool allows users to select different date ranges and conditions to determine copyright status and permissions needed. Columbia University Copyright Advisory Office—http://copyright.columbia.edu/copyright/. This site offers a wide array of copyright information for libraries, including the fundamentals of copyright, fair use, copyright ownership, and permissions. Stanford Copyright and Fair Use Center’s Charts and Tools—http://fairuse.stanford.edu/charts_tools/. This Web page offers a variety of charts, tutorials, flowcharts, and tools to help you navigate your way through the legal issues of copyright. Copyright Terms and the Public Domain in the United States by Peter Hirtle, Senior Policy Advisor, Cornell University—http://copyright.cornell.edu/resources/publicdomain.cfm. This tool offers useful charts to determine if a work is in the public domain for published works, unpublished works, works first published outside the United States, sound recordings, and architectural works. The Society of American Archivist’s Selected Copyright Resources—http://www2.archivists.org/groups/intellectual-property-working-group/selected-copyright-resources. This Web resource provides a wealth of information on general introductory topics on copyright, fair use, orphan works, public domain, digitization, copyright law, and workshop information. For further information on public domain works, see The Public Domain Review’s Guide to Finding Interesting Public Domain Works Online–http://publicdomainreview.org/guide-to-finding-interesting-public-domain-works-online/. COPYRIGHT AND NEWSPAPER DIGITIZATION When it comes to digitizing newspapers, there are two important and distinct rights issues to resolve before moving forward with the project. First, you must determine whether the newspaper you plan to digitize is in the public domain dating prior to 1923 and second, if the newspaper has been microfilmed, you must determine who owns the microfilm images regardless of the date. Housing a microfilmed newspaper in your library collection does not mean that you own the rights to digitize the microfilm. Permissions must be sought and granted from the rights holder of the microfilm before the holdings can be digitized. To determine ownership of the microfilm images you will need to check the microfilm box labels for a microproducer’s name and address and view the reel of film. At the beginning of the reel, after the blank leader film and before the actual filmed contents, are a series of targets. Among these targets is the ownership target or copyright statement target listing the name of the contents owner or rights holder. As an example, if you are considering digitizing historic newspapers that were dated pre-1923 and you found that by looking at the ownership target on the reel that it was microfilmed by the United States Newspaper Program, National Endowment for the Humanities in 1990, and the microfilm was held at a public institution, you would need permission from the public institution holding the microfilm. If on the other hand the newspapers you plan to digitize were dated post-1922, meaning the holdings are not in the public domain, permission must be sought from the newspaper publisher if he or she is still in business or from whoever holds the rights to the paper. If the contents have been microfilmed, you must also seek permission from the microfilm rights holder. To consider digitizing newspapers dated post-1922, the publisher would need to give you permission to use the content, and the microfilm producer would have to grant permission to digitize the microfilm. In some cases the publisher may own both the content and the microfilm rights; in other cases the publisher may own the rights to the content, while the microfilm rights belong to a commercial company. Also with newspapers be careful to consider whether the paper contains service-produced news articles such as those from the Associated Press (AP) and Reuters, or articles produced by syndicated columnists. The publisher may not own this content and therefore cannot give you permission to redistribute it in digital form online. It is always best to seek guidance and permissions from the newspaper publisher first to determine what can and cannot be digitized. PROJECT PLANNING AND ORGANIZATION Now that you have carefully weighed the legality to digitize the collection you have in mind, and have determined that the collection has not already been digitized by another entity, the next step in planning your digital project is to determine whether you have the skill set, manpower, infrastructure, and funding to carry out the project. Whether you have many staff to call on for assistance or you are the only one in your organization assigned to carry out the project, your answers to the following questions will define the direction of your project and how the project will be managed. • Who should be involved in the project? Possible stakeholders could include collection curators and staff, administrators, faculty or other subject experts, systems technicians, catalogers and metadata specialists, preservation professionals, scanning staff, Web developers, and financial officers. • Is this a collaborative effort with another institution? If so, clearly define the roles and responsibilities for each party. Prepare a formal agreement outlining the agreed-to terms and meet regularly to discuss project goals, timelines, progress, and any changes needed to keep the project on track. • Do you have the technical skills and knowledge needed to carry out the project? Will staff need to be trained? For instance, who will be responsible for inventorying the collection to determine its completeness, number of items or pages, condition, and structure? It is imperative to know your collection at the box-level, folder-level, and even page-level. Count and record the number of pages in the collection to determine its overall size. Fill missing gaps to ensure the completeness of the collection to be digitized. • Will the collection require preservation stabilization of the originals prior to scanning? Will the materials need to be cleaned, unfolded, flattened and/or sleeved in Mylar sheets prior to scanning? Will the collection require additional or special housing after scanning now that it is unfolded? • Will the project be produced in-house or outsourced to a vendor? What types of materials can you accommodate in-house? If outsourced, which vendors will you contact? Consider contacting peer institutions with known digitization programs to build a list of reputable vendors. As a financial steward of your institution’s money, request project price quotes from several vendors to determine the best price for the desired service. • What equipment and software will you need, whether produced in-house or outsourced? Will you need to purchase new hardware and software such as a dedicated computer and scanner for digitization, scanning software, content management platform software, image editing software, calibration software, or a color output printer? • Will your current system architecture or infrastructure support the project? Where will your final product live? Does your institution have an existing locally hosted digital content management system you can use, such as CONTENTdm? Is there adequate long-term server storage space for Web delivery of the content and for the long-term preservation of the digital files you produce? Does your institution have a disaster recovery plan for electronic media? How will the project be sustained over time? Does your institution have a data migration plan in place? • Is the collection cataloged in its entirety or partially? Is there a finding aid or an index to the collection that can be converted as part of the project to enhance access to the collection? • How and who will gather the descriptive, administrative, technical, structural, and preservation metadata to create access points for the online collection? It is worth noting that catalog records may be created at various points of the workflow, depending on the origin of the metadata and project needs. • What are your Web site requirements? Who will design and periodically review and upgrade the project Web site? • How will others know what you digitized? What is your marketing plan? • Is there adequate funding for the project? How much will it really cost? What are the direct costs and indirect costs? Are there internal or external grants available? Is there funding available from alumni or friends? While these questions may appear daunting, it is important to take the necessary time at the beginning of the project, whether several weeks to several months, to seriously contemplate the various aspects of your project plan and its implementation. Also at this time consider establishing realistic time-specific project workflow goals and objectives. Setting measureable outcomes will help to keep the project on track and on budget. One way to master the digital project planning process is to develop a conceptual framework that will help stakeholders understand how the process works. One resource that I found useful in developing and implementing the digital project workflow at the Penn State University Libraries was the Northeast Document Conservation Center’s Handbook for Digital Projects: A Management Tool for Preservation and Access. This primer on managing digital projects covers the gamut of topics from the rationale for digitization and preservation, project management, selection, copyright, vendor relations, and end users’ needs to a series of case studies on digital production best practices for various format types. PRODUCTION DIGITIZATION STANDARDS AND BEST PRACTICES The next step in the digital project progression is the capture process. Using either digital photography or digital recording devices, original source materials are captured in such a way as to not alter the information—meaning that the digital version represents the same information as the original analog version. While digital access to a collection can preserve originals through reduced handling, generally speaking, digitization is not yet a preservation medium; it is only a means of copying original materials that are stored on a stable medium. Digitization is all about improved sustainable access—the idea being that collections will be scanned only once and not rescanned. Moreover, once the content has been scanned, management of the master image files and maintenance of the image database is essential; it is necessary to know what you have and where it is stored. Due to the relatively short life cycle of digital information, preservation of your digital content is an ongoing process. It is critical to have a backup plan for items mounted on local servers, but it is also important to have a preservation plan in place for the archival files. The Association of Library Collections and Technical Services (ALCTS), Preservation and Reformatting Section (PARS), Working Group on Defining Digital Preservation developed and issued the following definition of digital preservation: “Digital preservation combines policies, strategies and actions to ensure access to content that is born digital or converted to digital form regardless of the challenges of file corruption, media failure and technological change. The goal of digital preservation is the most accurate rendering possible of authenticated content over time” (ALCTS PARS Working Group on Defining Digital Preservation 2009). To ensure the long-term accessibility of your digital assets consider the following questions: • Does your digital project plan include a process to manage descriptive, administrative, technical, structural, and preservation metadata for digitized collections over time? • Are processes in place to create and store backup copies on a stable medium for all master and derivative files produced for the project now and into the future? • Who is responsible for maintaining the intellectual content of the collection over time? • Who is responsible for maintaining the technical integrity of the collection over time? • What process is in place to regularly check the files for corruption by use of a checksum to verify the integrity of the data? Best practices for digital image production, digital preservation and quality control for capture, file management and maintenance and metadata standards abound. Following is a listing of some of the many resources available to assist you to follow current digitization project best practices. USEFUL LINKS Digital Project Resources American Library Association’s Association for Library Collections and Technical Services Preservation and Reformatting Section’s Minimum Digitization Capture Recommendations—http://connect.ala.org/node/185648. British Columbia Digital Library Digital Library Construction Tools: How-To Manuals—www.bcdlib.tc.ca/tools-manuals.html. Cornell University’s Moving Theory into Practice, Digital Imaging Tutorial—http://www.library.cornell.edu/preservation/tutorial/. Digital Library Federation’s Digital Library and Standards and Practices—http://old.diglib.org/standards.htm. Howard Besser, “Introduction to Imaging: Issues in Constructing an Image Database.” Getty Research Institute (2004). http://www.getty.edu/research/publications/electronic_publications/ introimages/index.html. Indiana University’s Digital Library Program Use of Digital Imaging Standards and Best Practices—http://www.dlib.indiana.edu/education/workshops/lsta04/handout9.pdf. Library of Congress’ American Memory Technical Information, Building Digital Collections: A Technical Overview—http://memory.loc.gov/ammem/about/techIn.html. National Archives and Records Administration’s Technical Guidelines for Digitizing Archival Materials for Electronic Access: Creation of Production Master Files—Raster Images—http://www.archives.gov/preservation/technical/guidelines.pdf. National Information Standards Organization, A Framework of Guidance for Building Good Digital Collection, 3rd ed. (2007). http://www.niso.org/publications/rp/framework3.pdf. Northeast Document Conservation Center’s Handbook for Digital Projects: A Management Tool for Preservation and Access—http://www.nedcc.org/resources/digitalhandbook/dman.pdf. OCLC’s Best Practices and Planning for Digitization Projects—http://www.webjunction.org/documents/webjunction/Best_Practices_ and_Planning_for_Digitization_Projects.html. Penn State University Libraries’ Digital Toolkit—www.libraries.psu.edu/psul/toolkits/digitaltools/standards.html. Stanford University Libraries Digital Production Services Quality Assurance—Cropping Guide–http://lib.stanford.edu/digital-production-services/quality-assurance-cropping-guide. Stanford University Libraries Digital Production Services Quality Assurance—Image Defects–http://lib.stanford.edu/digital-production-services/quality-assurance-image-defects. University of Michigan’s Digitization Guidelines for Photographs and Textual Documents—http://bentley.umich.edu/dchome/resources/digitization/20120719_ digitizationguidelines_final.pdf. Newspaper Digitization Project Resources International Federation of Library Associations and Institutions (IFLA) Newspapers Section’s Digitization Projects and Best Practices—http://www.ifla.org/node/6777. Library of Congress’ National Digital Newspaper Program’s Guidelines and Resources—http://www.loc.gov/ndnp/guidelines/. University of Kentucky’s meta|morphosis–http://www.uky.edu/Libraries/NDNP/metamorphosis/resources.html. Digital Preservation Association of Library Collections and Technical Services, Preservation and Reformatting Section, Working Group on Defining Digital Preservation’s Definitions of Digital Preservation—http://www.ala.org/alcts/resources/preserv/2009def. Cornell University’s Digital Preservation Management: Implementing Short-Term Strategies for Long-Term Problems—http://www.dpworkshop.org/dpm-eng/eng_index.html. Northeast Document Conservation Center’s (NEDCC) Surveying Digital Preservation Readiness: Toolkit for Cultural Organizations–http://www.nedcc.org/resources/digtools.php. University of Michigan’s Digital Preservation Management: Implementing Short-term Strategies for Long-Term Problems—http://www.dpworkshop.org/dpm-eng/eng_index.html. Metadata Dublin Core—http://www.dublincore.org. University of Denver’s Collaborative Digitization Program’s (CDP) Dublin Core Metadata Best Practices—http://www.mndigital.org/digitizing/standards/metadata.pdf. WHAT MORE TO SAY? SHARING LESSONS LEARNED In my 20 years of working with library curators and subject specialists to convert materials to digital form as a collection building strategy, I have learned some important lessons along the way. Some of these lessons were easy to grasp at the outset thereby averting potential problems later on; others resulted in redoing our work or even worse—ending up with unusable scanned collections. As an early adopter of digital technologies for collection development, we were motivated to try untested emerging technologies, but often uncertain as to where we were headed or what we might encounter. Initially it was a struggle. Each day was a challenge. However, over time and by gaining the necessary experience, the challenges and struggles became fewer; we righted our planning direction and focus and developed a vision for the foreseeable future. My hope is that by sharing what we have learned, you can avoid costly digital project pitfalls and experience success with your own digitization projects. Lesson 1: Do not bite off more than you can chew; bigger is not always better. A cool idea, enthusiasm, and ambition are a lethal combination and can easily distort reality. Our first digital project was to scan 350,000 archival pages. Needless to say, this project is still among our projects yet to be finished. Consider projects smaller in scope from 400 to 600 items in size, or in the case of newspapers, process batches of 10,000 pages at a time. Lesson 2: It takes a team. You need more than a single scanning technician to convert materials and build a digital collection. Subject curators, subject specialists and faculty experts, systems and information technology staff, preservation and reformatting staff, financial office personnel, and administrators each bring an important and necessary skill set to the digital planning and implementation process. Lesson 3: It takes longer than you think. How naive we were to think we could scan 350,000 pages in-house in two years. Today, our projects, albeit smaller in size, generally take 8–18 months from conception to public display. And regardless of the size of the project, the steps to digitizing a collection remain the same, although the time to inventory, stabilize, and scan the collections will vary. Lesson 4: Quality is everything. Quality source materials equals quality searching results. One important lesson we learned when working with microfilm is that poor film quality reduces the success in OCR processing and ultimately affects searching accuracy. With several of our historical newspaper digital projects we had to re-film portions of a newspaper due to the poor microfilm quality. Moreover, our need to re-microfilm resulted in additional project costs, increased time to locate and prepare materials for microfilming, a delay in the digitization phase of the project, and missed project deadlines. Lesson 5: Garbage in—garbage out. Just as important as having quality source materials is having quality description. High-quality description equals high-quality access and search navigation of collections. Metadata needs for different types of collections can be very different from what is supplied for traditional cataloging. With each digital project, it is necessary to work with the subject curator to determine access needs. The source of the metadata may be the cataloger, but depending on the project, it may also be supplied by a subject specialist, it may be purchased, or it could be converted from existing metadata such as MARC records, a finding aid, or a thesaurus. Lesson 6: Build useful and useable Web sites that are intuitive to navigate. Anticipate user needs first. Some of our first projects were all about production scanning with little attention paid to how users might access, navigate, or use the digital collection. With the eventual implementation of a team approach to digital projects, and the initial reference interview with the project curator to discuss scope, objectives, and expected outcomes of the project, we learned that user needs should be among the basic pieces of information that is gathered first. This also means building an intuitive user interface. Build search tip pages, provide derivative file formats for students to copy and paste into their research papers, add links to similar online content held elsewhere, and encourage use by incorporating interactive social media publishing options. Lesson 7: Web sites should be forever. Just when you think you’re finished with a digital product you find out that your institution recently acquired another 50 items to be scanned and added to the online collection. Eventually, too, you will be faced with the need to enhance your Web site, upgrade and migrate your collection to the newest software platform, and more than likely you will need to move, add, or delete objects from the image database. Lesson 8: Remember the original. Build time into your digital project plan to stabilize and care for your original materials before, during, and after the scanning process. Whether microfilm reels, glass lantern slides, or paper copies, repairing, rehousing, or binding may be warranted. Also, remember to document the preservation methods and techniques used to prepare the collection on the Web site for future reference. Lesson 9: Celebrate your accomplishments and let everyone know. Publicize. Early on in our digitization program we were so focused on production and targeted completion deadlines that we did not record the names of the faculty, staff, and students who worked on projects. Today, not only do we record all participants involved in the project, but we also prepare an acknowledgment page on each project Web site and celebrate the launch of our new digital products. To publicize our digitized collections, paper and email postcards are sent to donors and institutions with similar collections. We also use blog posts and other social media venues to spread the news. Keep in mind too that any promotional activities just might encourage future financial support for additional digital projects. Lesson 10: Once you build it, they will come—to your Web site and to your library. Never underestimate the power and potential of worldwide access. Users will find your products quickly, so be prepared for inquires, requests for physical access to originals, and especially requests to purchase images and/or print reproductions. CONCLUSION Despite the scores of challenges that you may encounter in your digitization project—from handling fragile materials in less than ideal conditions to the countless computer files that will need to be preserved over time, digitization as a collection development strategy in building online research collections is achievable. To recap, a summary of the steps for you to follow are included in the following. 1. Select materials 2. Clear rights 3. Locate and retrieve materials (annex, branch location, other campus locations) 4. Inventory, collate, and stabilize the materials (disbind, mend, repair, deacidify, encapsulate) 5. Determine digital content management platform for display and access 6. Determine file naming convention (OCLC# or barcode, folder/box or title/date/volume/issue number), for example, 3465879_R102_A_0001.tif 7. Collect metadata in a spreadsheet (bibliographic information, such as author, title, date of publication) 8. Scan (color, grayscale, or black/white) 9. Collect structural metadata (size, dpi, etc.) 10. Assess and treat original materials (rebind, box, etc.) 11. Return original materials to home location 12. Upload scanned images and metadata 13. Create user interface Web page giving context for users 14. Request MARC tag 856 link be added to the bibliographic record in your online catalog—link to the online product 15. Request digitized collection be added to your main portal Web page for your digitized collections 16. Generate publicity or marketing plan for the digitized collection 17. Update all internal spreadsheets of the collection going live online; report collection status to project curator or subject specialist and other project stakeholders; calculate costs incurred, and so forth 18. Add e-copy sticker to the physical analog item or collection container to alert users that the content has been scanned and is available online Libraries have embraced digitization as a means to convert analog collections to digital. Each year hundreds of digitization projects are launched and presented online thereby exposing collections to users worldwide. I hope that the information presented in this chapter equips and encourages you to consider project digitization as a feasible and vital collection development strategy for your institution. NOTE 1 “Digital Collections Review Team Digital Toolkit,” The Penn State University Libraries. https://www.libraries.psu.edu/psul/toolkits/digicoltoolkit.html (accessed December 30, 2012). REFERENCES Association of Library Collections and Technical Services, Preservation and Reformatting Section, Working Group on Defining Digital Preservation. 2009. “Definitions of Digital Preservation.” Available at: http://www.ala.org/alcts/resources/preserv/2009def. Accessed December 30, 2012. Conway, Paul. 2010. “Preservation in the Age of Google: Digitization, Digital Preservation, and Dilemmas.” Library Quarterly 80 (1): 61–79. Lynch, Clifford A. 2009. “Special Collections at the Cusp of the Digital Age: A Credo.” Research Library Issues 267.Available at: http://publications.arl.org/9ishf/prvp3/4. Accessed December 7, 2012.
APA (American Psychological Assoc.)
Albitz, B., Zabel, D., & Avery, C. (2014). Rethinking Collection Development and Management. Santa Barbara: Libraries Unlimited.
MLA (Modern Language Assoc.)
Albitz, Becky, et al. Rethinking Collection Development and Management. Libraries Unlimited, 2014. EBSCOhost.
Terry Sanders’ film Slow Fires (Sanders, Maddow, and MacNeil 1987) defined the challenges facing library and archives preservation efforts in the 20th century. Throughout the stacks, acid hydrolysis was constantly breaking down the cellulose fibers of paper, rendering the collections unusable. The problem was inherent to the media itself, and the only way to keep the information was to reformat it to new media. Microfilm was the technology that carried the day. Libraries adopted microfilm because it allowed for more compact storage, but it was repurposed for preservation because of its physical durability and chemical stability. By the time Slow Fires made its debut, the characteristics were nearly synonymous: microfilm was compact and eternal. It turns out though, that longevity is a minor consideration in libraries compared to storage space and ease of use.
Ironically, Slow Fires, a film about microfilming, concludes with a scene of an early digital library project that used optical discs to store images and text files. The filmmaker seems to have intuited the obvious narrative arc of ever-denser storage media, each generation one step further removed from physical presence and implicitly, from physical harm. This is the promise of ubiquitous computing: the data is fixed in no one place in particular, but can be delivered to many places simultaneously and every place potentially. Something is going on here at a deeper level than simply putting information on smaller or more durable media, and a better framework is needed to understand this. The Function Requirements for Bibliographic Records (FRBR) and the Open Archival Information System (OAIS) serve well in this role. In this chapter, I use these frameworks to explore way that preservation is accomplished across the domains of digital libraries and cloud computing, and printed or paper-based collections.
These two models, already widely used in their practitioner communities, are useful for harmonizing activities across the library and archival enterprise. The remainder of this chapter is in four sections. The first section gives a brief review of FRBR and OAIS with particular reference to how they work conceptually outside of their original spheres. The second and third sections discuss the state of digital preservation activities with particular reference to cooperative efforts, and examine the management of print storage in cooperative and consortium arrangements. Finally, these elements are unified into a preservation strategy that incorporates both print and digital preservations, each one supporting the other to ensure that libraries can reliably accommodate many types of use.
FRBR1 and the OAID2 are two of the key theoretical developments in the evolution of library collection management and technical services. FRBR is an ontology that shows how a Work can be Expressed in a variety of ways that are Manifested in particular media and mediums that are managed as individual Items by the library.
The FRBR hierarchy—Work, Expression, Manifestation, Item—was developed as an aid to describing collections but is highly effective for evaluating preservation as well. Indeed, engaging in preservation without clear reference to the level of the hierarchy where preservation should be targeted can create substantial problems. Work-level preservation efforts can succeed while Manifestation-level efforts fail. This type of confusion is at the heart of the ongoing conflation of digitization and preservation. Digitization can be an effective strategy at the Work and Expression levels, while simultaneously failing at the Manifestation and Item levels.
Deriving an alternate version of a Work at any level of the FRBR hierarchy is an important test for preservation. At a certain point, a work may be so greatly modified on its way to becoming an Expression that it is better to call it a new work entirely. Take the case of something like Heddatron, a version of Hedda Gabler with a half robot cast, wherein Ibsen is repeatedly thwarted by a kitchen maid and August Strindberg while trying to write Hedda Gabler. In parallel, a Michigan woman is abducted by robots and forced to perform Ibsen’s masterpieces over and over again. If every printed copy of Ibsen’s Hedda Gabler was lost—eliminating an entire chain of Expression, Manifestations, and Items—but a recording of a theatrical performance survived, we could use that recording to derive new scripts to create new Expressions, Manifestations, and Items. If the theatrical recording was subsequently lost, but the new scripts retained, the Work would have still been preserved throughout. However, if every Expression, Manifestation, and Item representing Hedda Gabler were lost, but we still had Heddatron, we could achieve at best a fragmentary and confused version of Ibsen’s work. The Expressions, Manifestations, and Items of Hedda Gabler are sufficient for preservation of the Work called Hedda Gabler. The Expressions, Manifestations, and Items of Heddatron are not.
This brings us back to Slow Fires and the subsequent debate over microfilming that Nicholson Baker launched in Double Fold (Baker 2001) and the articles that preceded it in the New Yorker, such as “Deadline” (Baker 2000) and “The Author vs. the Library” (Baker 1996). Baker excoriated librarians for microfilming newspapers and discarding the originals in the name of preservation. Librarians were preserving the Works and their Expressions, but were not preserving the Manifestations and Items. This observation is not meant to resolve the discord, but rather to facilitate debate. Accepting the concept that preservation plays out at different levels of the FRBR hierarchy, and that preservation at one level can occur in parallel with loss at another level, is necessary in harmonizing print and digital preservation strategies. Baker was objecting to a failure to preserve Manifestations and Items; librarians were rejoicing that they had preserved Works and Expressions.
There are numerous methods available to care for and repair items. These should be called conservation techniques, to clarify that they are a distinct class of activities related to preservation. Conservation techniques need to be deployed within a larger strategic framework of preservation and curation. The OAIS describes three main function of an archive: Submission (S), Archival Storage (A), and Dissemination (D) (CCSDS 2012). Each function is coupled to an information package (IP) and a method of transforming the IP between each function. The Submission Information Package (SIP) becomes and Archival Information Package (AIP) as it is Ingested into Archival Storage, and the AIP becomes a Dissemination Information Package (DIP) to give consumers Access to Information.
In terms of preservation theory, it is important to recognize that the SIP, AIP, and DIP may be nearly identical to one another. This would not be the norm in digital archives, where there are likely to be different formats for Archival Storage and Dissemination, as well as substantially different metadata at each stage in the OAIS reference model. For a print library, however, a book and its catalog record can be traced through this whole process and except for adding a few labels on ingest and providing a local call number, the object and metadata in the various Information Packages will be nearly identical.
Although the impetus for developing the OAIS was the recognition of digital preservation challenges, OAIS itself is fairly agnostic about the type of materials that the “Archive” might contain. As we attempt to develop strategies for managing preservation across different media, it is important to emphasize the media-independence of the OAIS model. Therefore, I make reference to the OAIS Functional Entities of Archival Storage and Preservation Planning in the discussion of print and digital preservation systems that follow, to emphasize the similarity of concepts across both media and by contrast, to emphasize where digital and artifact-specific traits present preservation advantages or risks that can be offset by one another.
Digital preservation efforts have had a long collaborative and cooperative tradition. E-journal preservation (and increasingly, eBook preservation) is the area of digital preservation most widely encountered by libraries and most closely coupled to the decisions libraries make about managing their print holdings. E-journal preservation occurs predominately through the Lots of Copies Keeps Stuff Safe (LOCKSS) (http://www.lockss.org) and Portico (www.portico.org) projects. LOCKSS and Portico both have strong records for preservation, but substantially different approaches. Alongside LOCKSS and Portico, HathiTrust (www.hathitrust.org) and the Internet Archive (www.archive.org) are the most important eBook preservation efforts at present, containing works digitized by Internet Archive and Google’s large-scale digitization projects. Finally, it is important to note that the digital preservation efforts discussed here are chosen because they are closely coupled to libraries’ general research collections, rather than library and archives special collections. Special collections are increasingly important to library services in the 21st century and are one of the major drivers of digital library projects, but almost by definition, special collections are not managed in a cooperative fashion as a shared resource.
The LOCKSS software has a substantial and successful record of maintaining content over time and has developed into several related projects that build on the same core preservation engine. The Global LOCKSS Network (GLN) has subsumed the original e-journal preservation function of the LOCKSS project. Private LOCKSS networks are being established as cooperative arrangements of digital libraries to share the costs of preservation. There are 10 or more now in operation, including MetaArchive (http://www.metaarchive.org) and Controlled LOCKSS (CLOCKSS) (www.clockss.org). MetaArchive is an international membership organization run by the Educopia Institute, a 501(c)3 organization. MetaArchive “caters to cultural memory organizations that are collaborating to preserve very high value locally created digital materials” and is reported to be among the largest Private LOCKSS Network implementations. CLOCKSS provides a dark archive of scholarly journal content, kept in a network of 12 LOCKSS boxes across Europe, Asia, and North America. When content in CLOCKSS is no longer available from a publisher, it is copied from the CLOCKSS Archive and made available at no cost to the library community.3
Critical to understanding the role of LOCKSS, and distributed preservation strategies generally, is the idea of storing many copies of identical objects in discrete locations. In FRBR terms, distributed preservation assumes many Items and, by preserving them, ensures the continued existence of a Manifestation, preserving in turn an Expression and thus a Work. Damage can occur to an Item in one node of the network without affecting the existence of the rest of the WEM chain.
Portico is the other major e-journal preservation system in use in the United States. Portico bears many similarities to CLOCKSS in the way it acquires and releases content, but uses a substantially different approach to preservation. LOCKSS receives materials from publishers and replicates them as ingested. Originally this meant that LOCKSS harvested the Web-delivered versions of the articles, generally HTML or PDF files. In OAIS terms, LOCKSS harvests the publisher’s DIP and treats it as the LOCKSS SIP. With minimal changes, this as-published content becomes the LOCKSS AIP. There is a potential problem if the publisher’s DIP, the content that is actually delivered to library users, is substantially different from the publisher’s own AIP, which lives in its proprietary content management systems, especially if they were to contain metadata or higher-quality data than the publisher released through its AIP (McCargar et al. 2009). The degree to which this is a real problem has not been well quantified, though. CLOCKSS may also mitigate this potential problem by acquiring files more directly from the publisher than the original LOCKSS system.
Portico, by contrast, was established to take publisher’s source files and normalize them for preservation. This allows Portico to mitigate several potential problems related to format obsolescence as Portico fully controls its own preservation formats. There is still some question as to when in the chain of production the publishers create the package of content that is received by Portico, and it is fair to assume that the answer differs between publishers. However, by working with publishers directly, Portico has the potential to get at production source files instead of distribution files. The loss of data through the digital publishing chain has received some attention, especially in the area of news preservation (Center for Research Libraries 2011). Although this is an emerging issue in digital preservation, production files potentially contain data and metadata that are valuable for long-term preservation.
Portico stores its archive in several on- and off-line versions and on different hardware platforms. The master copy of the archive in Princeton, New Jersey, is maintained using the Documentum content management system and Oracle database platform, and the Portico content storage area for publisher content is backed up to tape. One online replica of the archive is located on a file system in Ann Arbor, Michigan, and a second in commercial cloud storage. Both replicas are stored in a file system rather than a database/CMS and are self-describing and boot-strappable, so that the archive can be rebuilt from either replica. An off-line replica of the Portico archive is stored off continent at the National Library of the Netherlands, for preservation in their secure vault. Incremental backups of the archive are shipped to the National Library of the Netherlands on a quarterly basis and a full backup of the archive is created and shipped to the National Library of the Netherlands annually.4
Despite their different technical approaches, both LOCKSS and Portico achieve good results in preserving their holdings. HathiTrust and the Internet Archive do not represent radically different models for storing and managing digital content, but they are distinctive and important collections of material. Like the e-journals in LOCKSS and Portico, the digitized books in HathiTrust and the Internet Archive are closely coupled to libraries’ printed and general research collections, in contrast to digital versions of library special collections.
HathiTrust and the Internet Archive preserve huge amounts of data. There are about 475 terabytes of data representing nearly 10.6 million volumes in the HathiTrust.5 The Internet Archive hosts over 3.8 million texts, more than 1.1 million videos, and over 1.4 million audio files, totaling some 1.7 petabytes (1,741 terabytes) as of December 2010.6 Both repositories use a fairly conventional datacenter model to store and replicate their content. In the case of HathiTrust it stores its data in a file system managed using California Digital Library’s Pairtree micro service (University of California 2008), on two Isilon storage systems, one in Michigan and the other in Indiana. These are replicated on a frequent, periodic basis, usually daily but with some alteration depending on the amount of data in play on any given day. The two systems are mirrored and load-balanced, as well, so that a user may be served data from either system at any time, an important disaster mitigation technique.
The Internet Archive stores its data on its Petabox systems (Internet Archive 2004). The Petabox uses commodity hardware to provide 650 terabytes of storage in a single rack with low power consumption level of 6 kilowatts per petabyte, plus little need for air conditioning. Indeed, Internet Archive uses excess heat from its Petaboxes to help heat its building in San Francisco. At present, it reports having four datacenters in operation and has also used Bibliotheca Alexandria as a mirror site for its Web archive (Bibliotheca Alexandrina 2002).
All of these systems have good records in preserving their digital content, with no notable failures or loss of content to date. Although they represent different philosophies and levels of mitigation and risk-tolerance, they collectively make a strong case for the position that large-scale digital collections are reliable and available in the present. And, because these digital collections have a strong relationship to library print collections, they can be an important element in library preservation planning when used in conjunction with appropriate print storage strategies, discussed in the next section.
Shared collection practices, from ILL to large-scale digitization, have always mixed economic necessity with service aspirations. The development of purpose built high-density storage facilities (Payne 2007) that bring together sharing functions such as digitization and ILL (Seaman 2003) with preservation through environmental controls, preservation imaging and on-site conservation laboratories has opened the way to a new era in library collaboration around the management of the printed record (Malpas 2011).
Cooperative collection management allows access to a greater variety of material by distributing the costs of some resources across a network of partners. In many cases, there is a convenience cost for some users, as they wait for materials to be delivered. There are also outright and overhead costs to the library as they pay fees for using and maintaining the delivery system, but costs and delays often trump no access at all. Some libraries are electing to purchase materials outright rather than use ILL, an interesting network effect related to online booksellers making materials more discoverable for purchase and shared cataloging lowering the cost of metadata creation for new acquisitions.
Although the problem of scarcity is inherent in print collections, it plays out in digital libraries as well, with a key difference. In print collections, shared resources overcame a problem of scarcity. There were either very few items to begin with, as in the case of museums and special collections, or there were limits on the funds available to acquire items. In digital collections, scarcity is deliberately imposed.
Scarcity still persists in the experience of using digital resources, but it occurs at a different point in the supply chain. Print collections may suffer from a shortage of actual items to read and have to balance access versus the preservation needs of a particular physical object, whereas digital collections may be made scarce from a lack of usage environments; they are a metaphorical print library that has neither chairs nor light bulbs. Sometimes these constrains are rooted in available infrastructure. In 2012, 100 percent of public libraries reported that they offer Internet access, but in the same study 41.7 percent of report that they lacked sufficient broadband access to meet patron needs, and 69 percent report speeds lower than 10 megabytes per second (Bertot et al. 2012). The digital divide is a systematic barrier to access, but it is shrinking, and should be strongly contrasted to intentional constraints such as pay walls and systems that limit the number of simultaneous users for a particular resource. In the early 20th century, an argument that limited processing power and bandwidth-required restrictions on usage was believable, but it is harder to sustain in the 21st century, when most publishers, academic libraries, and public libraries have connected to very high-speed networks, and computation resources are available as a commodity service from many vendors.
Usage restrictions in the digital library are increasingly driven by license terms that are tied to a particular business model, not to an inherent characteristic of digital information. These restrictions are solutions to fiscal problems, not intrinsic characteristics of digital information (Farb 2006). The potential for ubiquitous digital access has enabled libraries to pay more attention to the economics of shared print storage and to bring preservation concerns to the forefront of that picture.
One of the most germane examples of distributed print storage is the U.S. Government Printing Office (GPO), which was from its inception intended to couple preservation and access functions and keep the nation’s records safe through “such a multiplication of copies, as shall place them beyond reach of accident,” rather than through “vaults and locks which fence them from the public eye and use in consigning them to the waste of time.”7 The GPO Depository Library Program was a direct inspiration to LOCKSS and has served well to keep government publications widely available since 1813. Today, the same core concept is driving a reinvestigation of shared collections and alluding to digital terminology to cloud-source the print collection.
Other programs are not as long established, but equally venerable. The Five Colleges created a shared collections system and depository in 1951, as the Hampshire Inter-Library Center (HILC). Since 1949, CRL has acquired resources that are difficult to for an individual library to acquire—for reasons of cost, scarcity, or both—and provided them to the CRL member libraries. In addition to maintaining objects in storage facilities, CRL has been involved in microfilming and digitization, and tellingly, was one of the early service providers for the Trusted Repositories Audit and Certification process (TRAC) (Center for Research Libraries 2007). OhioLINK (www.ohiolink.edu) was established in 1987, the same year the University of California system also created a shared collections program operated out of the Southern and Northern Regional Library Facilities (SRLF and NRLF). NRLF (www.srlf.ucla.edu) was developed in 1982, and SRLF (http://www.lib.berkeley.edu/NRLF/about.html) in 1987. Together they hold over 12 million items and served as the foundation for the Western Regional Storage Trust Program (WEST) (http://www.cdlib.org/services/west/). As of July 2007, as least 68 high-density storage facilities existed in the United States and Canada, holding more than 70 million volumes in total (Payne 2007, 6).
Throughout the 20th century, these programs have primarily served an access function by overcoming the scarcity problems of print collections. Preservation has always been an important secondary element in their missions, but has come to the fore recently. Access to adequate digital versions of the works kept in shared print networks has enabled a shift in focus, from preservation as a beneficial effect toward preservation as an intentional good. This change is playing out for two reasons. The first, discussed earlier, is the ability of digital libraries to provide Manifestation- and Expression-level access to a vast collection of materials that is large enough to dwarf all but a few American research library print collections, and accessible in a format that is intrinsically able to be ubiquitously available, even if law, economics, and policy proscribe use. The second, less discussed here, is the renewed attention to library collections as artifacts and the implicit requirement that these receive appropriate conservation treatment.
Several major developments in core preservation practice have likewise moved the preservation function of collaborative storage to the forefront. One is the development of a number of new conservation labs in conjunction with collection storage facilities. The other is the development of a more sophisticated system for the evaluation of preservation environments. The final element is an emerging effort in the preservation community to adapt risk management frameworks that deal with scarcity and endangerment to library networks.
Conservation lab construction was stimulated by several factors in the 21st century; in particular by the Andrew W. Mellon Foundation’s support for new lab spaces coupled with the economic advantages of constructing new high-density storage facilities (Teper and Alstrom 2012). For our discussion, the critical outcome is the close coupling of collections in need of services with the service providers. At the UC SRLF, for example, the preservation-imaging unit has been colocated with the existing storage facility since its inception, and plans for that facility include relocating the conservation laboratory into the expanded storage space. Indiana University and the University of Illinois Urbana-Champaign both constructed new storage facilities with integrated conservation labs in the 21st century. The New York Public Library has developed what may be the rule-proving exception, a conservation facility located not at the collection storage location, but at the central logistics point, which is still a hub for collections on their path from storage to use. Preservation and conservation facilities and expertise are at a key point of flow for library collections in all of these models, providing built-in opportunities to utilize preservation services, especially for general research collections.8
The cornerstone of preservation environment research is Donald Sebera’s (1994) work on isopermanence. Sebera’s key observation is that any given rate of deterioration can be caused by many combinations of temperature and relative humidity, and likewise, multiple combinations of temperature and relative humidity are capable of producing preservation benefits. James Reilly and other researchers at the Image Permanence Institute (IPI), William Lull, Tim Padfield, and others developed Sebera’s work in the interest of determining if there were a better path to conservation than the traditional effort to maintain a flat-line, constant temperature, and relative humidity environment (Tedone, Pavelka, and Becker 2008). From this, IPI’s Time-Weighted Preservation Index (TWPI) was developed and used as the basis for a series of field trials and research projects that led to the creation of hardware and software for monitoring and evaluating preservation environments. At the same time, the introduction of direct digital controls (DDC) into facilities management enabled more fine-grained, programmatic control of systems.
Adoption of DDCs is another variation on the theme: a change adopted out of fiscal concerns, but providing the opportunity for better preservation as a secondary benefit. A key collaborator on the IPI field trials, and the environmental optimization portfolio that developed out of them, is Herzog/Wheeler Associates, consulting engineers concerned with energy efficiency. Coupling Herzog/Wheeler’s process for energy efficiency with IPI’s metrics for preservation environments has led to multiple benefits: collections receive a better storage environment, energy usage and concomitant expenditures are reduced, and oftentimes (due to better review and understanding of mechanical systems in a storage facility), relatively inexpensive preventative maintenance makes catastrophic mechanical system failure far less likely. While this final utopic outcome is highly dependent on institutional priorities and politics, the dual benefit of reduced energy use for better preservation environments is compelling on its own, and intrinsic to the optimization process.
As technology has altered the usage of print collections, a delightful irony has emerged. The development of the Internet was predicated on the robustness of packet-switched networks, and, preservation models that effectively treat printed volumes as packets are now being developed. A working paper by Candace Arai Yano’s IEOR group at UC Berkeley, commissioned by JSTOR, is the foundation for this model. Yano, Shen, and Chan (2008) adapted models from manufacturing to develop risk of loss curves for print journals. Their basic argument was that the question of how many printed items were needed to ensure the existence of a single perfect item at the end of a given period of time was in essence the same question as asking how many spare parts needed to be produced during a production run to ensure sufficient supply to cover repairs for the life of a manufactured device.
Based on Yano’s adaptation of this line of thinking to the library environment, Nadal and Peterson (2013) developed a framework for evaluating risk in shared print collections that incorporates a greater degree of preservation information and library-domain-specific information. The results are notable. In many cases, 12–24 copies in usable condition are sufficient to satisfy a century-long planning horizon. For many items fewer than 12 copies are extant, which calls for significant preservation effort. However, in many cases these are already held at institutions with some level of preservation capability. More significantly, for a vast collection of materials, well over 50 copies are extant, making it possible to think of preservation as a service that focuses on a defined subset of the volumes in libraries. Preservation capacity still may not be robust enough across the system to meet that need, but being able to define the need is a critical step toward developing a viable strategic plan for preservation.
In the emerging scenario of large-scale digital collections backed by reliable digital preservation systems and equally large-scale shared print collections supported by reliable artifact preservation systems, FRBR and OAIS can be valuable ways of talking about unified preservation strategies that are more economical in capitalizing on the benefits of different formats and use cases.
For strategic preservation planning in research collections, Expression-level preservation is the goal. Work-level preservation is too abstract for useful policy making, while Manifestation- and Item-level preservation is more clearly the province of special collections. As an example, a researcher may want to consult different editions of a work (Expressions), but can use these in either print or digital form (Manifestations), yet does not specifically require the copy in the Rare Book Division with a Nobel Prize acceptance speech written on the flyleaves (Item).9
Print and digital collections play different and complementary roles in achieving Expression-level preservation. Conservator Gary Frost has made a compelling argument that print collections play a role in verification of digital resources, coining the term leaf-master to refer to a printed work kept for this purpose (Frost 2008). The adoption of the concept of managed scarcity in WEST attests to the necessity of this role for print collections. JSTOR’s studies also provide some information about how often this function is required, noting that half of its error reports occur in the first two years of posting an item online, and 92 percent within five years (Schonfeld and Housewright 2009). So far it appears that needs are minimal, which attests to the overall quality of digital collections and should give some comfort to anyone who cannot quite trust the math in the scarcity risk-management frameworks.
In fact, this dimension of trust may be one of the key coming-of-age challenges to librarians in the early 21st century. There is substantial data and information available to suggest that digital collections are reliable and usable and that shared print preservation can be effectively risk-managed. These assertions warrant healthy skepticism, but they are often subject to a chauvinism toward tangible objects that is harder to defend. It is important to differentiate between a few concepts that are sometimes misunderstood in our contemporary information environment: downtime as opposed to loss, and obsolescence versus decay.
Downtime is possible in a digital environment for a variety of reasons having nothing to do with the integrity of the underlying data. Momentary interruptions in access are independent of preservation outcomes, which are hedged against by processes for detecting and mitigating risks of obsolescence, data corruption, and physical damage and decay. All of these strategies work in tandem to preserve an Item that is sufficient to accomplish Manifestation-level preservation, and in turn keep at least one Expression of a Work available. Simultaneous failures of all these systems are required for the complete loss of any Work that exists in both a print and digital Manifestation. For a research library community concerned with preservation of Works and their Expressions, maintaining both print and digital Manifestations are critical elements of preservation strategy, but there are clearly diminishing returns on maintaining dozens or hundreds of redundant Items across the system.
Print and digital preservation do not exist in a zero-sum relationship, but instead, they should be treated as elements of a balanced preservation strategy. Many libraries will weight their efforts strongly toward print preservation. This is a primary function in rare book and manuscript collections. The Othmer Library at Brooklyn Historical Society where I work, for example, is a National Interior Landmark where place-based and primary-source research and education are intended to come together. A recent visit from architecture students at the New York City College of Technology involved viewing George Post’s architectural drawing for the library and then making their own measurements and drawings within the library itself. There is no digital equivalent. Artifact preservation is intrinsic to our services.
For other libraries, print collections are all but irrelevant. For example, the engineering libraries at Stanford and Cornell have made news by moving away from print collections (Sydell 2010), yet both universities are central players in e-journal preservation efforts. Indeed, Stanford hosts the LOCKSS project and Cornell both hosts arXiv (http://arxiv.org/) and was the original developer and site for the Digital Preservation Workshop and Tutorial (http://dpworkshop.org) that has been one of the primary venues for training and education in digital preservation.
Most libraries fall between these extremes. (Stanford and Cornell both maintain substantial book collections beyond their engineering libraries, of course.)10 A review of the membership lists for LOCKSS and Portico will turn up dozens of libraries that also maintain print storage facilities. Print collections and digital collections serve different functions and perform complementary roles. Taken together, a hybrid print and digital preservation strategy supports a wider range of use-cases and presents a lower risk of loss than either type of collection can provide on its own.
I have argued that physical artifacts and digital resources have complementary roles in preservation, and that the FRBR and OAIS models are complementary frameworks for managing stewardship efforts across these two domains. There are some types of inquiry that are essentially artifactual and others that are essentially digital, but for supplying a document to a reader, for the core function of libraries and archives, either type of resource can serve well enough. Some may prefer reading a text on screen and others on the page, but reading occurs in either case.
The ALA’s definitions of digital preservation state that “[d]igital preservation combines policies, strategies and actions that ensure access to digital content over time.” Remove the word “digital,” and this could be a workable definition of preservation of any sort, describing a goal of “access over time” that is achieved through “policies, strategies, and actions.” In this chapter, I have described two sets of high-level actions and a strategy for evaluating them. Print preservation is enacted through environmental controls, conservation, and agreements to limit risk and control scarcity. Digital preservation is enacted through replication of copies and encoding digital objects in well-understood formats that are tailored to their user-community’s needs. These actions can support a strategic preservation effort when FRBR entities are used to decide when print and digital resources serve the same need and when the OAIS is used to evaluate the viability of each preservation approach.
1 IFLA Study Group on the Functional Requirements for Bibliographic Records. 1998. Functional requirements for bibliographic records: final report. München: K.G. Saur.
2 Open Archival Information System. http://www.iso.org/iso/home/store/catalogue_ics/catalogue_detail_ics.htm?csnumber=57284 (accessed July 18, 2013).
3 “Who Gets Access to Content Impacted by a Trigger Event?” https://www.clockss.org/clockss/FAQ (accessed July 18, 2013).
4 “Portico Replication and Backup Policy, v. 1.1.” http://www.portico.org/digital-preservation/wp-content/uploads/2011/03/Portico-Replication-and-Backup-Policy.pdf (accessed July 10, 2013).
5 “HathiTrust Statistics and Information.” http://www.hathitrust.org/statistics_info (accessed July 10, 2013).
6 “Internet Archive PetaBox.” http://archive.org/web/petabox.php (accessed July 10, 2013).
7 Thomas Jefferson to Ebenezer Hazard, Philadelphia, February 18, 1791. In Thomas Jefferson: Writings: Autobiography, Notes on the State of Virginia, Public and Private Papers, Addresses, Letters, edited by Merrill D. Peterson. New York: Library of America. (1984)
8 There is more work to be done on the costs of preservation, especially the balance between direct costs for services and the overall costs across a collection or a library’s portfolio of service. For instance, some of these libraries had special collections adjacent spaces that were replaced or reconfigured in response to the construction of new lab spaces. The direct cost per special collections item treated may be higher, in these cases, even though preservation services overall may be having greater impact at lower cost. For an introduction to these issues, see Calvi, Elise. 2006. The Preservation Manager’s Guide to Cost Analysis. Chicago: Preservation and Reformatting Section, Association for Library Collections & Technical Services.
9 I owe this particular example to Michael Inman, Rare Books Curator at the New York Public Library, who showed me an exceptional inscribed presentation copy of John Marquand’s Thirty Years, from Ernest Hemingway to Lee Samuels, with Hemingway’s holograph draft of his Nobel prize acceptance speech on two blank leaves at the end. http://nypl.bibliocommons.com/item/show/13127997052_thirty_years (accessed July 18, 2013).
10 Cornell is actually one of the most notable examples of the genealogy of print and digital preservation. Anne Kenney, now Cornell’s university librarian, was associate director for Cornell’s Department of Preservation and Conservation from 1987 to 2001, and then transitioned into a leadership role in digital libraries and preservation.
ALCTS Preservation and Reformatting Section, Working Group on Defining Digital Preservation. 2007. “Definitions of Digital Preservation.” ALA Annual Conference, Washington, D.C. (June 24). Available at: http://www.ala.org/alcts/resources/preserv/defdigpres0408. Accessed November 4, 2013.
Baker, Nicholson. 1996. “Letter from San Francisco: The Author vs. the Library.” The New Yorker (October 14): 50.
Baker, Nicholson. 2000. “Deadline.” The New Yorker (July 24): 42.
Baker, Nicholson. 2001. Double Fold: Libraries and the Assault on Paper. New York: Random House.
Bertot, J. C., A. McDermott, R. Lincoln, B. Real, and K. Peterson. 2012. 2011–2012 Public Library Funding and Technology Access Survey: Survey Findings & Report. College Park, MD: Information Policy & Access Center, University of Maryland College Park. Available at: http://plinternetsurvey.org/sites/default/files/publications/2012_plftas.pdf. Accessed July 11, 2013.
Bibliotheca Alexandrina: Internet Archive. 2002. Available at: http://www.bibalex.org/internetarchive/ia_en.aspx. Accessed July 10, 2013.
Center for Research Libraries. 2007. Trustworthy Repositories Audit & Certification: Criteria and Checklist. Available at: http://www.crl.edu/sites/default/files/attachments/pages/trac_0.pdf. Accessed July 10, 2013.
Center for Research Libraries. 2011. Preserving News in the Digital Environment: Mapping the Newspaper Industry in Transition. Available at: http://www.digitalpreservation.gov/documents/CRL_digiNews_report_110502.pdf. Accessed July 11, 2013.
Consultative Committee for Space Data Systems (CCSDS). 2012. Reference Model for an Open Archival Information System (OAIS). Washington, DC: CCSDS. Available at: http://public.ccsds.org/publications/archive/650x0m2.pdf. Accessed July 11, 2013.
Farb, Sharon E. 2006. “Libraries, Licensing and the Challenge of Stewardship.” First Monday 11 (7). Available at: http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/viewArticle/1364. Accessed July 11, 2013.
Frost, Gary. 2008. “BookLab: Verging on the Future of the Paper Book.” BookNote (20). Available at: http://cool.conservation-us.org/byauth/frost/frost1.html. Accessed July 11, 2013.
Internet Archive. 2004. Petabox 4. Available at: http://archive.org/web/petabox.php. Accessed July 10, 2013.
Malpas, Constance. 2011. Cloud-Sourcing Research Collections Managing Print in the Mass-Digitized Library Environment. Dublin, OH: OCLC Research. Available at: http://www.oclc.org/content/dam/research/publications/library/2011/2011-01.pdf?urlm=162949. Accessed July 18, 2013.
McCargar, V., J. Nadal, H. Snyder, A. Vanek, and F. Zarndt. 2009. “Newspapers, Data Formats, and Acronym Stew: Preservation and Distribution of Born-digital Newspapers Using METS/ALTO, NITF, and PDF-A.” Proceedings of the IFLA International Newspaper Conference. In Newspapers: Legal Deposit and Research in the Digital Era, ed. Hartmut Walravens, 115–24. Berlin/Munich: De Gruyter Saur
Nadal, Jacob, and Annie Peterson, “Scarce and Endangered Works: Using Network-Level Holdings Data in Preservation Decision Making and Stewardship of the Printed Record” (Preprint, accepted for publication in ALCTS Monographs). Available at: http://www.jacobnadal.com/162. Accessed July 10, 2013.
Payne, Lizanne. 2007. Library Storage Facilities and the Future of Print Collections in North America. Dublin, OH: OCLC Programs and Research.
Sanders, Terry, Ben Maddow, and Robert MacNeil. 1987. Slow Fires: On the Preservation of the Human Record. DVD or VHS. United States: American Film Foundation. Available at: http://www.americanfilmfoundation.com/order/slow_fires.shtml. Accessed July 18, 2013.
Schonfeld, Roger C., and Ross Housewright. 2009. What to Withdraw? Print Collections Management in the Wake of Digitization. New York, NY: Ithaka S+R. Available at: http://www.sr.ithaka.org/research-publications/what-withdraw-print-collections-management-wake-digitization. Accessed July 10, 2013.
Seaman, Scott. 2003. “High-Density Off-Site Storage: Document Delivery and Academic Library Research Collections.” Journal of Interlibrary Loan, Document Delivery & Information Supply 13 (3): 91–103.
Sebera, Donald K. 1994. Isoperms: An Environmental Management Tool. Washington, DC: Commission on Preservation and Access.
Sydell, Laura. 2010, July 8. “Stanford Ushers in the Age of Bookless Libraries.” National Public Radio. Available at: http://www.npr.org/templates/transcript/transcript.php?storyId=128361395. Accessed July 10, 2013.
Tedone, Melissa, Karen Pavelka, and Snowden Becker. 2008. “From Gray Areas to Green Areas: Developing Sustainable Practices in Preservation Environments: Symposium Papers.” Austin, TX: Kilgarlin Center for Preservation of the Cultural Record, School of Information, The University of Texas at Austin. Available at: http://www.ischool.utexas.edu/kilgarlin/gaga/proceedings.html. Accessed July 18, 2013.
Teper, Jennifer Hain, and Eric Alstrom. 2012. Planning and Constructing Book and Paper Conservation Laboratories: A Guidebook. Chicago: Association for Library Collections and Technical Services, a Division of the American Library Association.
University of California. 2008. CDL Pairtree, Version 0.1. Available at: https://confluence.ucop.edu/display/Curation/PairTree. Accessed July 10, 2013.
Yano, Candace Arai, Z. J. Max Shen, and Stephen Chan. October 2008. “Optimizing the Number of Copies for Print Preservation of Research Journals.” IEOR Department and the Haas School of Business.
