by Peggy Seiden, College Librarian
In 1991, when Hurricane Bob roared up the eastern seaboard, the inhabitants of Cape Cod lost power for nearly a week. Twenty years later, when Irene promised to do the same, the prospect of losing our ability to cool and light our homes, cook, or watch television was daunting. But a loss of power also means, in most cases, a loss of the internet. We might all breathe a sigh of relief to be freed from the tyranny of email and social media for a few days - there are other ways to communicate. But increasingly, we have come to depend upon the internet to connect us to both current and historic scholarship. We still have access to the print for most of this digital scholarship, but will this always be so?
Early in June, NPR’s "On the Media" broadcast an interview with Brewster Kahle, founder of the Internet Archive and its related initiative to digitize millions of books. On the show, Kahle raised the question: if we have digital books, is there any need to have physical books? In his case, he answered in the affirmative. Kahle goes on to say, "We're discovering what librarians have known for centuries in this new digital world, so I'm feeling like a little naive." Kahle has built a high density, long-term preservation facility that could be expanded to hold 10 million volumes and serve as part of a distributed preservation system. His rationale is based on both the poor quality of previous digitization efforts, such as Google Books, and knowing that there may likely be something that we will want to do next with the physical books (though we may not know what that is).
It's not certain whether or not we have definitively solved the problem of preserving digital materials, but there has been a coordinated national effort to develop strategies that at least begin to address preservation of the digital scholarly record. However, is digital preservation enough to guarantee that researchers will continue to have access to these materials? There is a growing consensus that for some period of time, we will also want to continue to preserve print, even when there is a digital facsimile.
The original concept behind JSTOR was to digitize commonly held journals so that libraries could remove their journal back runs from their shelves in an effort to save or repurpose stack space. But despite generally high quality digitization processes, concerns remained, and many libraries, including Swarthmore, continued to retain their print copies of nearly all the JSTOR titles. In 2005, the TriColleges decided that they would divide up the responsibility for maintaining print holdings of JSTOR journals. Even though many of our students and faculty showed a marked preference for accessing these journals online, there was "an emotional attachment" to the print, a sense that having disciplinary core journals in print was essential to the library's or college's identity. There was some validity to the argument because of the intrinsic value of the printed artifact, the possibility that there was missing content, or the need to consult the print if there was significant amount of image content. But use patterns did not bear out the utility of retaining print.
Nevertheless, the TriColleges upheld their commitments to each other until this year when renovations at Bryn Mawr forced us to reconsider our retention decisions. McCabe Library has little room for growth and in fact has exceeded its projected maximum shelving capacity. There is the possibility that we will need to build a classroom in Cornell Library. We don't have the luxury of retaining volumes upon volumes of little or never used print content. De-accessioning JSTOR journals housed in McCabe alone would free up some 3500 linear feet of shelving or nearly a quarter of the stack space on the second level. In a recent study, Paul Courant, University Librarian at Michigan (Swarthmore '68), and Matthew Nielsen estimated that open stack storage of a volume costs $4.26/year.
But can we do this and be confident that should we ever need access to the print copy, we could get it? Given the expense of housing print collections, and the need to repurpose library spaces, it does not make sense that every library needs to retain their print holdings. In 2009, Ithaka, the parent organization of the JSTOR project, published a report entitled "What to Withdraw: Print Collections Management in the Wake of Digitization." The underlying study determined that the needs for continuing print preservation are "significant but not unlimited." It made recommendations about the number of copies needed and minimum time period to vouchsafe access to the original for print authentication (in case of failure of digital delivery mechanisms or for re-scanning).
Fortunately, certain library organizations have been giving a great deal of thought to how we guarantee that a sufficient number of print journals, government documents, and monographs will be preserved for future scholars and readers. These efforts, collectively referred to as shared print management, offer up the possibility of regionally based shared collections of print in trusted repositories. In most cases, the shared collections are housed in existing library storage facilities, but in others, consortia member libraries take responsibility for long-term retention of titles in a distributed model. The motivation behind these efforts is two-fold: to build trusted repositories (so that libraries have some level of comfort with de-accessioning their collections with the assurance that their patrons can still access these materials if needed) and to serve as a guaranteed source for these materials.
The initial focus of these efforts was on the JSTOR corpus. A number of consortia/organizations, such as the Five Colleges in Massachusetts and the Center for Research Libraries, have taken on the responsibility of comprehensively collecting the print issues and volumes and serving as trusted repositories for these materials. Across the country, other organizations are taking on responsibility for preserving other collections of print journals. The Committee on Institutional Cooperation (CIC) libraries are preserving works of a number of major STM (scientific, technical, medical) publishers including Springer, Elsevier and Wiley. Locally, the Pennsylvania Academic Library Consortium, Inc. (PALCI) has a fledgling program to distribute the responsibility for archiving American Chemical Society, American Institute of Physics, and the American Physical Society journals across member libraries. Both Ohiolink and the Associated Southeastern Research Libraries (ASERL) have projects to develop shared comprehensive federal documents collections.
While libraries have been placing low-use monographs in storage for years, to do so in a more strategic way is far more difficult and complex. However, there have been some promising conversations, and if the infrastructure under development for shared print journal and documents archives is sufficiently robust, one can imagine that libraries will begin to tackle the harder issues around books. Until we do, we run the risk of libraries independently making decisions to withdraw little used materials and seriously diminishing access to these works. Key to the success of developing a national shared print strategy is the ability to disclose information about permanent holdings (at a volume level) as well as the development of operating agreements about access and ownership. The Center for Research Libraries, with grant funding from the Institute for Museum and Library Services (IMLS), is leading efforts to build the infrastructure necessary to a national shared print strategy, as well as facilitating conversations among consortia regarding priorities for preservation and development of standards and best practices.
It will be a long time before we can comfortably de-accession all the journals we access online. JSTOR only accounts for a few hundred titles among thousands of online journals to which we subscribe. But for many of these other titles, we do not have extensive backfiles either because they have not yet been digitized or because they are too costly to purchase. Our policy is to continue to retain print if we do not own the backfiles or if there is no trusted long-term digital preservation strategy. We are participating in the PALCI initiative mentioned above and are responsible for archiving several ACS and AIP titles.
In the last five years, attitudes towards print retention and use have begun to shift on campus. This past year, the TriColleges became affiliate members of the Five Colleges storage repository in order to ensure access to print JSTOR volumes should the need arise. Now we are in the process of querying faculty about JSTOR volumes we might de-accession. In most cases where images are not a significant (greater than 5%) portion of the articles, faculty have approved our removal of the volumes.
Given Swarthmore's historically rich holdings, it is likely that we still have a role to play in preservation, albeit not on the scale of major research libraries. However, we will also benefit greatly from our ability to repurpose stack space for people to use or for additional collections. There is no question that print is important to scholarship for the foreseeable future, but the nature of our local collections is certainly about to change.