Open Access for Monographs is Here. But Are we Ready for It?
By John Sherer, Spangler Family Director of the University of North Carolina Press. He is the chair of the Association of University Presses Open Access Committee and is the Primary Investigator in the Andrew W. Mellon Foundation funded Sustainable History Monograph Pilot. This post originally appeared on The Scholarly Kitchen.
At the University of North Carolina Press, we recently completed a four-year initiative to support the publication of open access (OA) monographs by university presses in the discipline of history. Through the Sustainable History Monograph Pilot (SHMP)—generously funded by the Mellon Foundation—we published fifty-nine titles with nineteen different presses. SHMP was unique among OA initiatives in that it did more than simply provide offset funding. It required an intervention in publication workflows using our publishing services platform, Longleaf Services, where university presses and their authors had to submit to a standardized process with templated outputs that aligned with the requirements of digital platforms. The pilot recorded successes and challenges indicative of ongoing and future needs, some of them general to all OA publishing and some specific to OA monograph publishing.
There were several areas of encouraging news. The titles we published are being accessed and used exponentially more than digital editions that are behind paywalls. And standardization significantly reduced the costs required to create a high-quality university press monograph. But there were also discouraging results. Authors and presses remain deeply cautious about OA as a legitimate, sustainable mode of publishing. They are even more skeptical about the trade-offs associated with standardization. And we learned that there are serious gaps in the infrastructure needed to disseminate open scholarship efficiently and measure its impact.
The Obstacles Facing an OA Publisher
Our first area of setbacks and challenges were technical and knowledge based, ranging from the lack of availability of workflow tools to support the unique affordances of open digital publications, to the unique challenges of dissemination of OA metadata and digital files, to a dearth of best practices in how to measure usage of OA monographs.
We experienced significant challenges in managing metadata for our presses. The first issue was that OA metadata has unique needs, requiring the creation of new fields and rubrics that paywalled metadata doesn’t consider (for example, some content management systems cannot support a zero-priced product or store a DOI or ORCID). Additionally, commercial aggregators (like JSTOR, EBSCO, ProQuest) distribute book metadata for free because they will generate income on a sales commission. But if a press removes the sales commission possibility, what is the incentive for an aggregator to distribute the enriched metadata? This put us in the surprising scenario where metadata for OA editions could actually circulate less widely than metadata for paywalled e-titles. To overcome this, we had to create a supplemental metadata work-flow connecting presses and OA platforms. After we began our project, an independent initiative (funded by Arcadia) developed a tool called Thoth to try and address this set of challenges around metadata.
Beyond the distribution of metadata, we also had to build a workflow to distribute eBook files to OA platforms. In a paywalled environment, a press can use a digital asset management system (like Ingram’s CoreSource or Chicago Distribution Center’s BiblioVault) as a one-stop tool for eBook dissemination. But these commercial vendors don’t have connectivity with OA-only platforms like the Internet Archive, OAPEN, or an author’s institutional repository. This became another example of how distribution of an OA eBook is actually more laborious than a paywalled title.
Once we had published books, of course we wanted to measure their usage in open environments. We were once again surprised to learn that there are no best practices for how to define and measure a moment of reader usage. We relied on our six OA platforms to supply us with what data they could, and then we had to manually normalize it. Since SHMP was initiated, another Mellon-funded initiative has begun (the OA eBook Data Trust) to develop standards and communities of practice around this issue.
Lastly, we noted that OA workflows are challenging for most university presses, filled with numerous one-off requirements. Even though our program helped presses significantly with metadata and final file distribution, each project felt like it was exceptional. If OA remains on the fringes as a tool for presses, it will have the added burden of being more time-consuming to execute.
Our second area of challenges were the cultural and institutional barriers of author/press concerns, and the lack of financial and administrative resources at academic institutions.
Many authors we tried to work with expressed skepticism about the project. This was reported to us second-hand through the originating press, so our feedback is anecdotal. But in a survey we commissioned, our presses ranked “author concerns” as their own biggest issue in participating in the pilot. When we drilled down on this question, templated interior and cover designs were cited the most, but close behind was an impression that an OA monograph would be viewed less favorably than a traditional print monograph would in the tenure and promotion review process. Almost one third of projects submitted to us by presses were ultimately withdrawn from consideration for these reasons. Presses also cited financial challenges as a fundamental reason why they could not publish more OA titles.
We learned that academic institutions appeared ill-prepared to support OA monographs. In a number of instances, authors were advised by colleagues or supervisors to avoid a pilot like this on the grounds that any program that could be perceived as unconventional might harm the career prospects of a historian. We also asked participating authors to seek matching subventions at their home institution to support the program. Authors frequently didn’t know who to ask, and in our final results, of the fifty-nine titles we published, only thirteen institutions had funds to support OA monographs of their faculty with the average subvention amount from them being $2,525.
The Rewards of Standardized OA Publishing
It shouldn’t be a surprise to anyone that open digital editions get downloaded, read, and cited more than paywalled editions. Although it may be surprising just how much more they get used. Among university presses, sales are usually tracked in the low hundreds, and a highly successful monograph is one that sells more than 1000 copies. As of this writing, our fifty-nine SHMP titles have had more than 195,000 unique user engagements. Almost every title we helped presses publish exceeded 1000 usages within its first six months. After a year, it is usually more than double that and, in most cases, well over 3000 usages. These numbers are understated since we are unable to track “re-shares” of downloaded eBooks or downloads that came from authors’ own web sites or institutional repositories. <any monographs take years to find readers and our books are still in their infancy. We received in-depth reporting from JSTOR, one of the platforms where we distributed our OA eBooks. They compared the usage of SHMP titles (all of which are in the history discipline) against the paywalled history books in their full collection and SHMP titles showed an 8-fold increase over paywalled titles. And while SHMP titles made up only 4% of the JSTOR title set, it accounted for 22% of the use.
One of the keys to our model was the introduction of standardization early on in the workflow. In addition to speeding the publishing timeline, we hypothesized that it could help reduce costs. Our testing proved that we were able to reduce costs in copyediting/typesetting/file prep on average by almost 60%.[1] The savings were primarily in typesetting (because we used a templated design), cover design (again, a template), and in project management (which was less expensive because of our scale and use of an auto-composition tool).
As many presses have worked diligently to embrace diversity, equity and inclusion, they have confronted the inherently elitist and exclusionary distribution model for university press monographs. Our pricing models severely limit readership, and thus our requirements to sell enough copies of books to recover costs means that our publishing programs have privileged authors and subject areas from the wealthiest institutions and countries. The lack of institutional financial resources to support the dissemination of monographs reinforces that inequity. This is even more dire since in addition to the dramatic increase in the volume of OA usage, we are seeing usage in nations that have been historically marginalized in our marketing and dissemination efforts.
Where do we go from here?
While the benefits of OA are becoming clearer (expanded usage, better measurements of impact, and alignment with DEI goals), most presses still see financial sustainability as the biggest obstacle to publishing more OA. Furthermore, presses and authors can be extremely reluctant to embrace a standardized workflow and output, despite the possibility of saving thousands of dollars per title. And while we’re beginning to close some gaps, there is still more infrastructure that needs to be built to make OA publishing more seamless and efficient.
Some university presses have displayed admirable leadership in opening their lists (Michigan and MIT) and there’s an exciting cohort of newer, born OA presses that have the benefit of not having to convert from a legacy model. And there are new models for spreading the sources of funding like the Open Book Collective.
For our next step at UNC Press, we have been helping to develop a new initiative that is essentially a compromise between the legacy model of university press publishing and a fully -funded OA model. Path to Open is a concept modeled on the NEH Fellowship Open Book Program which provides for a three-year embargo period during which presses can participate in conventional cost-recovery activities, including selling print and consumer (e.g., Kindle) eBooks. During this time, JSTOR will be offering the digital versions of these titles to academic libraries and institutions in an exclusive subscription collection. JSTOR will pay presses an estimated $5,000 for each title put into the program. The value proposition to presses is that they can continue to use their legacy models to produce and market books during the time window when most cost recovery usually happens while ultimately making them more accessible. For libraries, they can buy a modestly priced collection of university press monographs and provide them for their patrons for three years. And at the same, they are contributing to a sharing of risk that allows everyone in the ecosystem to thrive: authors, presses, libraries, institutions, and readers everywhere.
The landscape remains wide open. OA feels like the most desirable future, but there’s still no clear roadmap for how to get there. If the first generation of OA pilots proved that OA can increase impact exponentially, the next generation of pilots need to find sustainable funding models to achieve this.
[1] Our costs for this work averaged $3,000. The recent TOME report prepared by Nancy Maron et al estimated an average of $7,040 for the same work.