Please tell us what you think of this issue!  Feedback

Bulletin, April/May 2009

Institutional Repositories: The Great Debate
Topic 2
Libraries Should Lead the Institutional Repository Initiative and Development at Their Institutions

Affirmative Argument 
Jim Ottaviani, coordinator, Deep Blue, University of Michigan Library
Email: hellpop<at> 

Negative Arguement 
Carolyn Hank, TRLN doctoral fellow, School of Information and Library Science, University of North Carolina at Chapel Hill
Email: hcarolyn<at>

Libraries make people crazy. Nobody says it that way or in those exact words, but some of the things librarians do, I swear…

We ask people to return things, and if they’re overdue sometimes we charge them for it (unless they’re faculty, for whom even the idea of overdue fees and returning materials may be objectionable). We don’t buy enough good stuff, and we buy too much stuff that nobody cares about (meaning for many users: not enough of what I read, too much of what they read). Perhaps it’s a complaint unique to Michigan, but also generally we won’t violate copyright law for anyone’s convenience. (“I know you and Google scanned it, so let me have the digital copy!”)

Crazy. But for all our flaws, people also know libraries are expert at two things: free and forever.

Not that the material we provide access to is free, and not that we don’t realize that our best efforts at preservation sometimes fall short. But librarians are magicians when it comes to hiding costs from users – articles from licensed journals materialize on desktops at the twitch of a finger, leading their readers to think they cost nothing. And if it’s old or fragile or rare? Researchers know better than to turn to their own shelves or rely on publishers for access to this kind of material. They come to libraries assuming that we have it and that we won’t make them pay to get it.

What does this all mean for libraries’ role in institutional repository development? Since the concept of free and forever is baked into the definition of an IR, librarians should find leading IR initiatives as natural as breathing.

Better still, IRs offer many additional features that libraries are adept at providing, such as access control and format migration. Most authors don’t think hard, and perhaps not at all, about those things, and they shouldn’t have to do so. That’s the library’s job, not theirs. To be fair, others have made it their job as well: Academic publishers are, after all, excellent at control and have realized the importance of migration. (Though in the early days of digitization, where did they go to fill in the often-huge gaps in runs of their backlist journals? Three guesses, and the first two don’t count.) So libraries are not uniquely capable in these areas, but they are arguably the best choice to lead institutional repository development and long-term preservation of access. Commercial academic publishers don’t provide preservation services to individuals, on-demand or without charging a lot of money. Libraries do those things, with handholding and a pat on the back added in as a bonus when needed.

Libraries are also good at acquisition, though here’s where things get dicey. Libraries are good at paying for things on behalf of our scholars, but not so good at getting things for free, especially when doing so means asking faculty to step outside the bounds of tradition. On a related note, we’re also not good at marketing and promotion. (Exhibit A: The phrase institutional repository. Those are two very…institutional…words and a bland shorthand for a robust and valuable service.) These are functional areas in which librarians must improve. Our livelihood and our mission both depend on it, because in an era where 24/7 worldwide access to the published record is possible, our great libraries will be considered great not because of what they offer that duplicates what’s in other collections – though we won’t be able to get away with not having such things – but because of what we offer that’s available nowhere else. To that end, for Deep Blue, we promote the idea of presenting the director’s cut of the research. Sure, we want the published paper; everybody wants that. And many will have it. But we can now offer the raw data, the equipment list, the color images, the video and/or audio of interviews, the negative results or results that don’t reject the null hypothesis, the code book…the list of things that other scholars and researchers would find valuable (but aren’t likely to appear in a final, edited paper or book because of space, money or logistics) is huge. And IRs can bundle all of those facets of the work together in a citable package with a persistent URL. That’s new and different and valuable and, in my experience, an easy sell. And also a little bit of a hard sell. Because of the traditional behaviors and roles mentioned above, this sort of service still feels new. It shouldn’t, but it does.

So back to easy stuff by way of reconsidering those crazy things libraries do in the context of an IR…since IRs fix them all! We don’t need anything returned to an IR once it’s downloaded – it’s still there, in perfect shape. Our faculty determines an IR’s content themselves – if they want something good in the library, all they have to do is put it there and we’ll take care of everything else. And since it’s their stuff, it’s there to be downloaded if they say it should be. Few (and ideally, none) of the things in an IR cannot be accessed immediately upon deposit, so there are no worries about us hiding the digital versions of useful work, unless that’s what the depositor asks us to do. And we understand copyright, so we can help faculty cut better deals with publishers or at least go into the publishing arena with open eyes.

But these considerations are all secondary. It’s really about free and forever, and libraries proved we’re the best at those things decades, maybe even centuries, ago. In fact, we’ve been so far out in front of these issues and ideas for so long that we’ve become invisible. We can change that. Libraries should lead, because they already do, and do it well.

Recent surveys show that libraries indeed are leading in institutional repository initiatives and developments at their institutions. The Census of Institutional Repositories in the United States: MIRACLE (Making Institutional Repositories a Collaborative Learning Environment) Project Research Findings [1] reports that libraries take on between 40 to 60% of the responsibility for institutional repository planning and pilot testing. Other institutional units such as archives, central computing, CIO offices and central administration participate at these stages though with a considerably reduced amount of responsibility. At the deployment and implementation stage, however, the disparity of responsibility between these institutional units shifts considerably, with libraries taking the lead and assuming the majority share for their respective institutional repositories. A survey of members of the Association of Research Libraries (ARL) on institutional repository development and deployment also found that libraries lead [2]. Units within the libraries, for the most part, are primarily responsible for institutional repository administration. 

Returning to the topic of this position piece, it is not so much a question of whether or not libraries should lead. Libraries should very much be active leaders in institutional repositories, although institutional repositories require an extensive investment of resources for development and administration. The associated tasks necessitate a planned and deliberate coordination of a diverse and extensive collection of stakeholders, both within the library and most definitely outside of the library, including scores of producer and consumer sub-groups. Libraries have a rich history of serving the needs of these subgroups through traditional library functions and services, but transferring this traditional library expertise to the requirements of a successful institutional repository program is not so transparent. It raises fundamental questions of both capacity (for instance, sufficient staffing levels) and capability such as sufficient staff skill and knowledge. 

Instead, I believe this is a question about how libraries should lead. Libraries should not lead institutional repository initiatives if leadership, and subsequent institutional repository management, is exclusive. That is, libraries should share leadership and operational responsibilities with other campus units. From several years’ experience participating in institutional repository planning activities at a large research university, I understand the many challenges associated with institutional repository development, particularly in coordinating with numerous campus constituencies in a landscape of too many priorities and too few resources. Libraries are well positioned to lead; this responsibility should be, however, a joint management endeavor at the operational level, with concrete contributions from other institutional units as opposed to simply expressions of affirmative sentiment. 

Consider trends in institutional repository deployment. In a 2004 interview Cliff Lynch described institutional repositories as “terribly fashionable” and said “everyone wants to claim to be building one” [3]. Three years later, in the forward to the MIRACLE census, Abby Smith wrote, “One of the paradoxical findings of the survey [is] that there is detectable urgency on the part of libraries to implement institutional repositories, even as early adopters report difficulties in achieving the purposes for which they built them” [1, Foreword]. While there are many mitigating factors that impact adoption and success in institutional repository development, these quotes reflect a challenge for leaders. How do libraries manage institutional repositories, negotiating among the ever-evolving needs, objectives and issues, in light of the current climate of shrinking budgets, a lousy economy and shifting priorities at institutional and departmental levels? If they wish to take on this task just as early adopters are revisiting their institutional repository programs in light of post-deployment difficulties, libraries should reconsider their approaches to leadership and management by revisiting the seminal definition of an institutional repository provided by Lynch in 2003 [4].

Lynch wrote, “While operational responsibility for these services may reasonably be situated in different organizational units at different universities, an effective institutional repository of necessity represents a collaboration among librarians, information technologists, archives and records managers, faculty, and university administrators and policymakers.” Seven years after this statement was published, operational responsibility sits squarely within libraries. This responsibility may have been charged (that is, delegated by the provost) or self-fulfilled (that is, adopted by libraries). Do we know if a co-managed, shared operational model would be more beneficial than exclusive operational management by libraries? No, because we do not have a pool of examples from which to evaluate. Granted, such a shared management and operations model would not be easy to implement since collaboration, like leadership, is a challenging task. The challenge is reflected in the second part of the excerpt above and raises the issue of the (at times) ambiguous distinctions between collaboration and cooperation. 

To cooperate in institutional repository development is to support it. “Cheerleading,” or advocacy, should be placed on the side of cooperation. To collaborate in institutional repository development, however, is to contribute to it. Resource allocation is very much on the side of collaboration (for example. programming, hardware and software support; staffing and funding). Again, considering the findings from the MIRACLE census and the ARL survey, there is evidence of both collaborative and cooperative partnerships between the cast of players described by Lynch. However, these partnerships are most active during planning and pilot testing phases and tend to drop off at the operational stage. This trend was confirmed in interviews this author conducted for a study on institutional repository planning [5]. Institutional repository managers were asked to comment on their libraries’ collaboration with other campus units. Interviewees described their activities as a library-specific enterprise, with the exception of campus-wide representation on planning committees. As institutional repositories continue to develop and emerge, new collaborative models for operational management may very well contribute to improved and sustainable services. 

Continuing to borrow from Lynch, an institutional repository is “most essentially an organizational commitment to the stewardship of these digital materials, including long-term preservation where appropriate, as well as organization and access or distribution.” Organizational commitment may be manifest in many ways, including dollars. Just as surveys show a trend in responsibility between planning (shared) and operational management (exclusive), there is also a trend in funding between initial infusions of project and planning support to subsequent operational, recurring funds. Per the MIRACLE census, funding is primarily provided – or will be provided for planned and piloted instances – by libraries. 

A fuller treatment grounding Lynch’s definition in actual, real-world deployment activities is outside the scope of this paper. But even this abbreviated consideration leads to the question: Can libraries, exclusively, accomplish the tasks explicit in Lynch’s definition in their institutional repository development and deployment activities? From reports on planning, libraries have been deliberate in building a cohort of stakeholders from across the institution to inform project development activities, reflected in the diversity of institutional repository planning committees. However, in the transition from projects to emerging, and presumably sustainable, programs, the libraries’ role has transformed from major player to sole player. This change complicates the already intense task of building and managing institutional repositories. 

For an appreciation of the extent of this endeavor, consider the complex and comprehensive criteria enumerated in the Trustworthy Repositories Audit & Certification (TRAC): Criteria and Checklist [6]. This reference is offered not to advocate audit and certification for all repositories (though that would be another useful exercise for the affirmatives and negative stances taken throughout this special section), but simply to demonstrate the series of critical steps, requirements and decisions necessary for the development and operation of a “trustworthy” institutional repository. Potentially, libraries, and in turn institutional repositories, would be better served though a sharing of these responsibilities with other expert agencies on campus. It would certainly decentralize the requisite funding and thus present a potentially more sustainable model.

Further, the need to reconsider shared leadership is not just exclusive to the institutional setting. Libraries should also look to collaborations and resource sharing with other libraries and organizations external to their respective institutions. There are several examples of deployed consortial institutional repositories, but there are other ways, bound by consortial boundaries or not, that libraries can build these ties or improve upon existing collaborative relationships. 

To return to the question: Can libraries, exclusively, accomplish the tasks explicit in Lynch’s definition in their institutional repository development and deployment activities? My simple response: No. My irresistible, hokey response: Just like there is no I in Team, there is no I in Lead (and yes, I know I should have just resisted).

Resources Mentioned in the Debate
[1] Markey, K., Rieh, S. Y., St. Jean, B., Kim, J., & Yakel, E. (2007). Census of institutional repositories in the United States: MIRACLE Project research findings. (CLIR Publication No.140). Washington, D.C.: Council on Library and Information Resources. Retrieved February 23, 2009, from
[2] Association of Research Libraries. Institutional Repositories Task Force, & University of Houston Libraries. (2006, June). Spec kit 292: Institutional repositories. Washington, DC: The Association. Retrieved February 24, 2009, from
[3] Lynch, C.A. (2004, August 15). Editor’s Interview with Clifford A. Lynch. RLG DigiNews, 8(4). Retrieved February 24, 2009, from
[4] Lynch, C.A. (2003, February). Institutional repositories: Essential infrastructure for scholarship in the digital age. ARL: A Bimonthly Report, 226. Retrieved February 24, 2009, from
[5] Hank, C., Tibbo, H., & Barnes H. (2007, May). Building from trust: Using the RLG/NARA Audit Checklist for institutional repository planning and deployment. In Archiving 2007. Springfield, VA: Society for Imaging Science and Technology. Retrieved February 24, 2009, from
[6] Center for Research Libraries. (2008, July 11). Trustworthy Repositories audit & certification (TRAC): Criteria and checklist. Chicago, IL: Center for Research Libraries. Retrieved February 24, 2009, from