I’ve had the great privilege to help develop and lead Ithaka S+R for more than 20 years, including research and advisory projects supporting the work of libraries, publishers, and the research enterprise. It’s a moment of professional transition for me, as I assume responsibility for JSTOR Digital Stewardship Services and entrust Ithaka S+R to my colleagues. Through JSTOR Stewardship, we will help libraries, archives, and museums steward the distinctive collections that are essential for grounding academic work and public discourse about our history and society. My perspective has been shaped in many ways by my work over these past two decades, and here I want to reflect on some of what I’ve learned.
Shared Infrastructure
Much of my work has focused on innovation and sustainability for the shared infrastructure that makes the work of libraries and publishers possible, with a significant emphasis on collections, preservation, and digitization. A key challenge is that shared infrastructure must activate the benefits of cross-institutional scale in ways that preserve and advance academic values without sacrificing innovation and agility.
In a project supported by STM Solutions and several major publishing houses, The Second Digital Transformation of Scholarly Publishing, Oya Rieger, Tracy Bergstrom, and I examined the shared infrastructure needs and sustainability issues for scholarly publishing. In the paper, we described key infrastructure categories — Identifiers; Enterprise Publishing Systems; Discovery, Collaboration, and Trust; and Preservation. For each of these, we looked at the dynamism that is brought by startups, the stagnation that can arise as these become mature monoliths, and the challenges associated with membership organization governance. As Oya has pointed out to me so many times, these social and organizational dynamics make shared infrastructure hard — the technical part is comparatively easy.
With Gwen Evans, then at OhioLINK, I had a chance to examine deeply the shared infrastructure needs of libraries and their user communities in It’s Not What Libraries Hold; It’s Who Libraries Serve. Ultimately, we identified many ways in which integrated library systems could better align with the modern needs of academic libraries. At the same time, the diversity of library types and the desire to act together through a variety of collaborative vehicles at once introduce real complexity for infrastructure providers. Gwen taught me through personal example the importance of a strong and coherent voice to drive forward collaborative work — every chorus needs a conductor.
Over the years, I have had so many thoughts about how to develop sustainable business models for shared infrastructure! I wrote several post-mortems about initiatives that had to sunset or consider major pivots here in the Scholarly Kitchen, including for The Digital Preservation Network, The Digital Public Library of America, and DuraSpace. Based on these examples and broader dynamics facing shared infrastructure providers and membership organizations for libraries, I proposed some lessons learned in Restructuring Library Collaboration, particularly related to the interplay between membership governance models and the marketplace for software, systems, and services. Organizations should exist to serve a purpose, and not every initiative needs a new organization.
Collections as a Public Good
Some of the projects that most deeply captured my imagination involved efforts to think about the general collections of libraries as a public good, or as Lorcan Dempsey has termed it, a collective collection. The Google digitization initiative was an important touchstone for me in this work. Soon after the initiative was announced, Brian Lavoie and I had the opportunity to examine the scope and characteristics of the global collective collection. This analysis, which had never been done before, led to a tremendous amount of collections analysis by OCLC Research to support shared print initiatives. Brian and I learned together that an enormous share of library collections are rare, with tremendous significance for how we think about discovery, access, and collection management for distinctive collections.
More than 15 years later, Deanna Marcum and I wrote a history of library digitization, Along Came Google, in which, we examined the profound challenges that libraries have faced in coordinating around digitization initiatives, which ultimately enabled Google to serve as an effective catalyst in organizing this activity, even beyond the material resources it contributed. In exchange, Google generated substantial proprietary value, which it has leveraged in building advanced technologies. Research libraries eventually asserted their ability to provide for the public good through HathiTrust. Deanna taught me again and again, and we learned together in this project, about the importance of purpose-driven leadership for libraries in pursuing the vision of expanding access.
Additionally, I have had the chance to think about format migrations and obsolescence in partnership with the shared print community. This included a project Ithaka helped fund in which Candace Yano and colleagues analyzed the optimal number of print copies that should be retained and preserved once a given journal is digitized. I learned from Candace how we could bring a more scientific approach to collective collections work, which Ross Housewright and I tried to bring into applicability in What to Withdraw. At the same time, the scientific approach is only as good as the commitment that libraries are able to make to preservation in a managed down environment, and later I urged library leaders to begin Taking Stock, as I began to fret that as we move to manage collections collectively, to expand access and drive efficiency, it is vital we not lose sight of the preservation imperative.
Distinctive Collections
The work that I am embarking on now with JSTOR Stewardship involves building and sustaining shared infrastructure to help libraries, archives, and museums manage their distinctive collections. This is an enormously important and, as Rick Anderson has argued, an increasingly significant category of institutional holdings. These are the materials that provide the grounding not only for scholarship but for our understanding of our cultural and intellectual history. Through primary sources, we understand our own humanity.
I first had the opportunity to use special collections materials as an undergraduate, and I worked with my colleague Jennifer Rutner to study their use in an NEH-funded project on Supporting the Changing Research Practices of Historians. We found that researchers face key challenges in using archival collections, indicating substantial opportunities and complexities for discovery and access wrought by new technologies. My favorite example is how the digital camera has transformed the work of individual archival researchers — and the resistance faced to achieving its full transformative potential. Later, my colleague Matthew Long and I undertook a similar project on Supporting the Changing Research Practices of Art Historians. We found that the discovery environment, including for primary sources, was fragmented and frustrating.
In a project originally suggested by Robin Dale, then at the Institute of Museum and Library Services, Oya Y. Rieger, Liam Sweeney, and I assessed The Effectiveness and Durability of Digital Preservation and Curation Systems. Through this project, we concluded that notwithstanding the strong commitment of numerous professionals across the sector, digital collections infrastructure and practices faced some vulnerabilities. Notably, many libraries selected their systems not based primarily on values but also to a large degree on market competition, with a number of struggling not-for-profit offerings becoming displaced by growing commercial providers.
Looking Ahead
At a high level, I see an opportunity to expand the impact of distinctive collections to the same degree that we experienced in the first wave of general collections digitization, where JSTOR was an early leader. In our JSTOR Stewardship work, we will apply some of what has been learned, including many of the principles discussed above, in providing a shared infrastructure for distinctive collections in a way that addresses the public good. I welcome your reactions to these reflections and your suggestions for future directions.
