Adventures in Portlandia: Learning the Difference Between Publishing and Repositories

Late last year, I wrote a post in the Scholarly Kitchen on the subject of library publishing. I was pleasantly surprised by the level of impact and positive feedback that the post generated. As just one example, it was quoted as inspiration for a panel on library publishing at this year’s PSP conference. While I can’t really take credit for it, it is good to see library publishing becoming more influential and being taken increasingly seriously. Charles Watkinson (@charleswatkinso), who runs both the library publishing program and University Press at University of Michigan will be delivering the opening keynote at this year’s SSP conference in Arlington, VA.

I had the pleasure of moderating a panel at the Library Publishing Forum in Portland, Oregon last week. My panelists were Andrew Wesolek (@AndrewWesolek), Head of Digital Scholarship at Clemson University, David Scherer (@davidascherer), Digital Repository Specialist at Purdue, and the aforementioned Watkinson. The title of the session was A game of spot the difference: Librarians, Repository Managers, and Publishers, and my original goal was to explore the potential for convergence between repositories and library publisher programs. To my mind, funder interest in open scholarship, which has lately manifested itself as mandates for data sharing, as well as the increasing diversification of the scholarly record in both the sciences and digital humanities, will eventually lead to a blurring of the lines between what we traditionally see as a published article and other types of academic output. While traditional publishers are beginning to tackle this trend with the creation of data journals like Nature’s Scientific Data and PLoS’ support for data publication, libraries are well placed to support scholars by helping them disseminate output, as well curating and reporting it through research management software.

While planning the panel, however, it became clear that there exists a further level of complexity. As Watkinson put it to me, I was underestimating the strategic need to treat various types of output in different ways. During discussions, I and my three more qualified panelists discussed three main functions that library publishing platforms and repositories fulfill:

1. satisfying OA mandates and helping scholars disseminate work through green repositories,
2. providing data repositories to support open data and digital scholarship,
3. publishing, particularly within the humanities, but also grey literature like technical reports and student output.

The trend in many libraries seems to be towards separating those roles. The question then becomes how to support this level of diversity with a limited budget. Do you use a single platform and try to make it do as much as possible, or have three separate, more specialized platforms and perhaps struggle with interoperability and cross-linking issues? Our three panelists came from institutions with varying sizes of publishing/repository programs and so for a starting point, we described three different profiles or case studies; small, medium, and large.

• Small: At Clemson, the tigerprints repository hosts primary published works, preprints, data and other digital outputs. Built on Digital Commons, tigerprints does a fantastic job of catering to multiple use cases.
• Medium: Purdue’s approach is to support a diverse range of scholarly output with services ranging from publication strategy advice to metrics. Their publishing platform (ePubs) is separate from the Purdue University Research Repository (PURR).
• Large: Michigan are in the middle of implementing Hydra-Fedora, funded by a nearly $900k Mellon Foundation grant. The idea is to create a best of both worlds solution to provide a central repository and infrastructure for all digital output, with multiple heads. Each head may look and function like a journal, preprint server, or data repository as needed. The downside being the comparative expense and complexity involved in such a project.

Although program size (read budget) was our starting point, as the conversation on the day unfolded, it became clear that many other factors influence the type of publishing and repository support that a particular institution might offer. The attitude of the vice provost for research was cited as a factor by more than one librarian. At some institutions, the VPR is highly supportive of data repository or advanced digital efforts and is willing to support funding but at other institutions, that’s less true. The culture and type of outputs generated at the institution was also discussed as a factor. An institution that has a large amount of student research output or technical reports, for instance, might have more of a need for in-house publishing solutions as opposed to a data repository. Lastly, somewhat of an elephant in the room was the role of the institutional IT department. Since IT are involved in the implementation of any technology solution, they exert a strong influence on the types of services provided. For example, many IT departments seem uncomfortable with cloud based solutions and some are focused on local infrastructure rather than external usability and discoverability, at least for the time being.

There are many unanswered questions when it comes to libraries’ role in dissemination of scholarly works. It may be that different institutions will always adopt different strategies, reflecting the types of scholarly output that they generate, local needs and priorities, and the occasional political constraint. In recent years, as publishing efforts have developed out of institutional repositories, the need to become more outward facing and support a wider range publishing services has led to the three functions that I mentioned earlier becoming increasingly separate. In a sense, we might think of our small, medium and large library publishers as analogous to how library publishing may develop over time, as it becomes a larger and increasingly important sector. I find it telling that as we go from small to medium, there becomes enough funding to support multiple specialist platforms, but for large library publishers, the option of a single centralized infrastructure with multiple front ends or heads becomes the best of both worlds solution that ties the repository and publishing efforts back together again.