Knowledge Unlatched: A Global Library Consortium Model for Funding Open Access Scholarly Books

Specialist scholarly books, including monographs, allow researchers to present their work, pose questions and to test and extend areas of theory through long-form writing. In spite of the fact that research communities all over the world value monographs and depend heavily on them as a requirement of tenure and promotion in many disciplines, sales of this kind of book are in free fall, with some estimates suggesting declines of as much as 90% over twenty years (Willinsky 2006). Cashstrapped monograph publishers have found themselves caught in a negative cycle of increasing prices and falling sales, with few resources left to support experimentation, business model innovation or engagement with digital technology and Open Access (OA). This chapter considers an important attempt to tackle failing markets for scholarly monographs, and to enable the wider adoption of OA licenses for book-length works: the 2012 – 2014 Knowledge Unlatched pilot. Knowledge Unlatched is a bold attempt to reconfigure the market for specialist scholarly books: moving it beyond the sale of ‘content’ towards a model that supports the services valued by scholarly and wider communities in the context of digital possibility. Its success has powerful implications for the way we understand copyright’s role in the creative industries, and the potential for established institutions and infrastructure to support the open and networked dynamics of a digital age. Introduction This chapter reports on an important attempt to enable more effective markets and wider access to specialist scholarly books: Knowledge Unlatched. The Knowledge Unlatched (KU) proof of concept pilot ran from 2012 – 2014, involving nearly 300 libraries from 24 countries, as well as research funders, publishers, digital intermediaries, authors and readers. It engaged stakeholders in the process of developing and testing a global library consortium model for funding the publication of specialist scholarly books on an Open Access (OA) basis. The pilot was a bold attempt to reconfigure a highly international area of publishing: moving it beyond the sale of ‘content’ towards a model that lowers barriers to access and supports services valued by scholarly and wider communities in the context of digital technologies. The chapter begins by introducing the monograph, and considering some of the historical factors that have led to the current crisis in this highly specialised area of publishing. It then goes on to outline the specific challenges associated with publishing monographs on OA licenses, before reporting on the Knowledge Unlatched pilot and considering its implications. Monographs in a changing landscape Monographs are book-length scholarly works on a single subject or theme, usually by a single author. Humanities and Social Sciences (HSS) research communities all over the world value specialist scholarly books, including monographs, as a form of writing that allows complex ideas to be developed and shared at length (OAPEN-UK 2014; Adema & Rutten 2010). In contrast to textbooks, which provide a broad overview of a field and are intended for a student audience, monographs present primary research and original scholarship, and their intended audience is generally other researchers. Having a book-length work published by a recognised scholarly


Introduction
This chapter reports on an important attempt to enable more effective markets and wider access to specialist scholarly books: Knowledge Unlatched. The Knowledge Unlatched (KU) proof of concept pilot ran from 2012 -2014, involving nearly 300 libraries from 24 countries, as well as research funders, publishers, digital intermediaries, authors and readers. It engaged stakeholders in the process of developing and testing a global library consortium model for funding the publication of specialist scholarly books on an Open Access (OA) basis. The pilot was a bold attempt to reconfigure a highly international area of publishing: moving it beyond the sale of 'content' towards a model that lowers barriers to access and supports services valued by scholarly and wider communities in the context of digital technologies. The chapter begins by introducing the monograph, and considering some of the historical factors that have led to the current crisis in this highly specialised area of publishing. It then goes on to outline the specific challenges associated with publishing monographs on OA licenses, before reporting on the Knowledge Unlatched pilot and considering its implications.

Monographs in a changing landscape
Monographs are book-length scholarly works on a single subject or theme, usually by a single author. Humanities and Social Sciences (HSS) research communities all over the world value specialist scholarly books, including monographs, as a form of writing that allows complex ideas to be developed and shared at length (OAPEN-UK 2014; Adema & Rutten 2010). In contrast to textbooks, which provide a broad overview of a field and are intended for a student audience, monographs present primary research and original scholarship, and their intended audience is generally other researchers. Having a book-length work published by a recognised scholarly press is understood as an important intellectual achievement (Adema & Ferwerda 2014) and the publication of a monograph serves as evidence of professional competence and is a requirement for tenure and promotion in many disciplines (Crow 2012;Williams et al. 2009).
In spite of their importance as a textual form through which new knowledge in the Humanities and Social Sciences is developed, certified and communicated among research communities across the globe, very few readers beyond the walls of the university are able to access these kinds of books. Books have been slower than journals to make a shift to digital formats (Adema 2010) and monographs are often available only in hardback. They are generally published in short print runs and their biggest market is university libraries in the United States and Western Europe, who pay between $50 and $250 per copy (Gasson 2004;Steele 2008). There is widespread consensus that sales of monographs are in decline: twenty years ago it was normal for to sell several thousand copies and today, most will sell just a few hundred (Wasserman 1998;Bunz 2014;Williams et al. 2009). At a moment when new technologies should be lowering barriers to access, monographs remain both closed and expensive.
The mission-focussed nature of many monograph publishers, particularly University presses, has made adapting to the pressures of a changing market and transformative technologies particularly challenging (Wasserman 1998). Humanities and Social Sciences academics depend on monograph publishers for professional certification purposes. Publishing at least one, and sometimes several monographs is a standard requirement for tenure and promotion in many disciplines. As a result, expansion of academic communities has been accompanied by an increase in the number of monograph titles being published, as publishers seek to provide the professional publishing services demanded by scholarly communities. However, there has not been a corresponding increase in book acquisition budgets, even at wealthier institutions (Cummings 1992). On the contrary, an expansion in the number of monograph titles being published has coincided with a decline in the book purchasing budgets of librarians, as the proportion of library budgets being spent on journals has grown. In the United States university library monograph acquisitions only increased by 1% over the 24-year period from 1980 to 2004, whereas journal acquisitions increased by 180% (Anon 2007). The result is that monograph publishers find themselves publishing fewer copies of each book that they take on, and attempting to recoup their costs through higher cover prices and a smaller and smaller number of unit sales.

The 21 st Century and Digital
Digital technologies and the internet are creating further challenges. The Internet is making it possible for academic communities, and whole populations, to access use and make knowledge in new ways. It is also making it conceivable that researchers located in the global south may gain a much higher level of access to existing stores of knowledge, and to processes of research and knowledge certification (Noronha 2006). Awareness of the value of ensuring the widest possible access to published research outputs is growing and what scholars expect from their publishers is changing, as are the expectations of Universities, research funders and reading publics (Willinsky 2009).
Funding agencies all over the world are becoming conscious of the potential for Open Access to maximise the impact of investments in research by ensuring that new knowledge is made available to all who might learn from, apply and build on it (Finch et al. 2013). The Registry of Open Access Mandatory Archiving Policies lists 90 OA mandates applied by research funders, and a further 228 institutional mandates (ROARMAP 2014). These mandates require researchers to make works available by self-archiving final, peer-reviewed drafts in a freely accessible institutional or disciplinary repository ("Green OA") or by publishing them in an OA journal ("Gold OA") or both. Funding agencies that have adopted OA mandates for grant recipients include the US's National Institutes of Health, Research Councils UK, the National Fund for Scientific Research, the Wellcome Trust, and the European Research Council.
To date, most OA mandates have focussed on peer-reviewed journal articles. Few research institutions or funding agencies currently require researchers to make booklength works available on an OA basis. 1 This is because policy makers and research communities remain anxious about the impact that requiring OA for monographs might have on an already troubled area of scholarly publishing (Finch et al. 2013;Mandler 2014). There is widespread consensus that the revenue models developed by journal publishers to support 'Gold OA' are unlikely to work for books (Gasson 2004;Adema 2010). There is also concern that business models capable of supporting a large-scale shift to OA for books have not yet been identified, and recognition of the value of continued experimentation in this area (Finch et al. 2013).
Researchers are justifiably worried about how OA requirements for book-length works might impact on their capacity to find, and afford, publication opportunities. Sustainable routes to OA for large numbers of scholarly books are needed in order to ensure that the core research outputs of the Humanities are visible, accessible and useful in a digital world.
A key challenge in ensuring that monographs are not left behind in the shift to OA relates to the fixed costs associated with publishing book length works. It costs more to publish a 70-100,000 word scholarly book than it does to publish a 5 -10,000-word journal article (Willinsky 2009). In 2013 Palgrave Open announced a £11,000 (US$17,200) fee for the publication of OA monographs on a CC-BY licence, while Manchester University Press is charging £5,900 -£7,800 (US$9,230 -$12,200) for the publication of books on a CC-BY-NC licence (Anon 2014a). OA charges for books are simply too high to be affordable for individual authors. Furthermore, research budgets in the Humanities and Social Sciences are much smaller than in STEM disciplines, and many authors aren't attached to a research budget at all.

Can Libraries Enable Open Access for Books?
The role of libraries as the only purchasers of monographs makes them especially vulnerable to changes in library budgets and purchasing patterns. It was also the starting point for the Knowledge Unlatched global consortium model for enabling OA for specialist scholarly books. Publisher and social entrepreneur Frances Pinter observed that libraries have a long history of working together in consortia to secure benefits for the academic communities that they serve. Her hypothesis was that libraries could work together in an international consortium to share the costs publishing books on an open access license, and that doing so would have benefits for all of the stakeholders in this system, particularly authors and readers, but also for publishers. With this in mind, Pinter established Knowledge Unlatched (KU) as a notfor-profit 'Community Interest Company' in 2012. Its mission is to create a sustainable route to OA for book-length scholarly publications by developing the coordinating mechanisms that libraries and publishers need to ensure that OA occurs efficiently. In doing so, KU aims to help libraries to maximise the positive impact of the money that they are already spending on specialist scholarly books and to help secure the future of the monograph.
The next section of this chapter outlines the Knowledge Unlatched proof of concept pilot, which took place between 2012 -2014. The Pilot involved libraries, publishers, authors, readers and research funders in the process of developing and testing a global library consortium model for supporting Open Access books. 297 libraries from 24 countries shared the cost of 'unlatching' 28 newly published Humanities and Social Sciences research books published by 13 well-known scholarly presses. The pilot demonstrated that it is possible for publishers and libraries to work together in new ways on a global scale. It also highlights the need for new approaches to coordination, governance and 'value' in scholarly communication landscapes in the midst of transformation, as well as the capacity of established institutions, like libraries and publishers, to add value in a digital context by making relatively simple changes in their operating models.

The Knowledge Unlatched Proof of Concept Pilot
Significant work was required to turn Pinter's broadly framed concept of a global library consortium into a defined pilot offering that could be presented to publishers, authors and research funders, and marketed to libraries as a viable route to securing access to content. Between mid-2012 and early 2014 the KU team worked to identify the needs of stakeholder communities through workshops and surveys; and to raise awareness of the project among stakeholder communities. Information gathered during this time was used to inform the development of the Pilot model (Montgomery, 2014). This work was funded through a combination of grant funding, from the Open Society Foundation and the British Library Trust, as well as by three founding Australian libraries: Queensland University of Technology; the University of Western Australia; and the University of Melbourne.
In October 2013 libraries were invited to pledge a maximum of $1680 towards securing OA for 28 new Humanities and Social Sciences research titles from 13 recognised scholarly presses. If at least 200 libraries worldwide agreed to contribute towards the cost of the Collection publishers would be paid a Title Fee in return for making the books available on Creative Commons (CC) licence 2 without an embargo, once the books had been published. If less than 200 libraries joined, the books would still be published, but they would remain closed. The publishers that took part in the Pilot provided a cross section of the different kinds of scholarly presses involved in the monograph space: university presses; commercial presses; large and small publishers from North America, the United Kingdom and Continental Europe were all involved. All of the KU pilot publishers operated according to traditional, closed publishing models. 3 The Pilot Collection included titles from the following 13 publishers: · Amsterdam University Press · Bloomsbury Academic · Brill · Cambridge University Press · De Gruyter · Duke University Press · Edinburgh University Press · Liverpool University Press · Manchester University Press · Purdue University Press · Rutgers University Press · Temple University Press · University of Michigan Press

Title Selection
During the second and third quarters of 2013, publishers were sent guidelines about the types of book that should be submitted for possible inclusion in the Pilot. These guidelines emphasised the need for rigorous peer review. They also encouraged publishers to submit titles that were likely to be relevant to an international audience.
Publishers then submitted a list of books that they would be willing to offer to libraries as part of a KU Pilot Collection.
The 13 participating publishers offered more than 100 front-list titles for possible inclusion in the pilot collection. KU worked with collections librarians at the New York Public Library in order to curate these books into a single 28-book collection that covered History, Literature, Political Science, and Media & Communications. One book in Anthropology was also accepted. The majority of the books were monographs and five were edited collections. Publishers and librarians anticipated that the core audience for the Collection was likely to be undergraduate and graduate level research students, as well as independent researchers.
Once acceptance of a title into the KU programme had been confirmed, publishers discussed the KU initiative with authors and, in consultation with them, selected the Creative Commons licence that should be applied to each work.

Licences
Copyright for scholarly monographs generally resides with the author, who assigns specific rights to a publisher. As such, permission from authors to make books available under a Creative Commons licence was required. Publishers worked with their authors to explain the Creative Commons licensing options and to select the specific Creative Commons licence that should be applied to each work. 4 'Non-commercial' Creative Commons licences were chosen by authors and publishers for twenty-five out of the twenty-eight books in the pilot collection. Noncommercial Creative Commons licenses reserve commercial rights in a copyright work. This means that a separate license must be negotiated with the copyright owner for all commercially related uses of the work (Lessig 2004). By maintaining an exclusive right to the commercial exploitation of a book, publishers remained confident about their ability to generate additional income for themselves and their authors by publishing and distributing the book through commercial channels in other formats. This helped to keep the Title Fees for the books in the Pilot Collection lower than might have been the case if they had been offered on licences that included rights to commercial re-use.
Many of the authors who agreed to the inclusion of their books in the Pilot Collection indicated to their publishers that they were not comfortable with granting a blanket licence allowing others to alter or adapt their work. In order to protect the integrity of their work, some authors chose to reserve the right to preserve the integrity of their work by selecting a Creative Commons licence that includes a 'No-Derivatives' condition. This licence restriction means that the authors expect down-stream users to seek permission before creating derivative works.

Cost
Publishers were asked to submit a Title Fee for each of the books included in the Pilot Collection, calculated on a Cost Recovery basis. That is, the Title Fee reflected the costs of bringing a monograph to first digital file: proposal review; internal manuscript review; editorial guidance to authors; external reader payments; copyediting; typesetting; proofing; design; permissions fees; marketing; and overheads. Costs associated with printing, binding and digital distribution were not included in the Title Fee. Book prices normally include a risk factor that addresses uncertainty regarding sales, which is not necessary with the KU approach.

Recognising Other Format Purchases
One of the most challenging aspects of the pilot design process was identifying a mechanism that would ensure that library contributions to the costs of bringing books to first digital file were reflected in the cost of other formats. Libraries were concerned that the KU model might amount to a double payment from libraries to publishers: requiring libraries to pay for the cost of unlatching a book in addition to the full cost of a print or closed e-book version. 8 Finding a way to ensure that this did not occur was vital to building confidence in the KU model among libraries. It was especially important for libraries that subscribe to pre-purchase plans or large (closed) e-book collections. It is not always possible or practical for libraries to opt out of purchasing a single title, or a small number of titles, because they are being supported elsewhere: for example, through the KU Pilot Collection.
Publishers indicated early in the pilot design process that they were willing to provide discounts on print or other format purchases to the libraries that helped to unlatch the Pilot Collection. However, finding a transparent, time-efficient mechanism for delivering these discounts in a complex, global distribution landscape in which discounts negotiated by library consortia, book-sellers, e-book vendors, digital platforms and other intermediaries play important roles was not easy.
For the Pilot, publishers agreed to waive the Unlatching Fee for books that a library indicated it was ordering through another channel. The Unlatching Fee was also waived for libraries that indicated that they intended to order a book within 30 days of the end of the pledging period through another supplier. This approach minimised disruption to established purchasing processes for libraries. However, it still required libraries to manually check to see whether additional formats of a title in the Pilot Collection were being ordered through other channels.
Publishers participating in the Pilot also accepted the risk that libraries might try to 'game' the model: claiming so many additional format purchases that the unlatching fee payment would fail to cover the fixed costs of publishing a book. In spite of this risk, publishers tolerated this approach to recognising additional format purchases and engaged with the KU proof-of-concept Pilot as an opportunity to learn about how they could work with libraries in new ways.
Promoting the Offer mathematics, physics, astronomy, computer science, quantitative biology, statistics, and quantitative finance. 6 The Portico digital preservation service is part of ITHAKA, a not--for--profit organization helping the academic community use digital technologies to preserve the scholarly record and to advance research and teaching in sustainable ways. Diversity in the official languages of countries involved in the Pilot highlights the international reach of the project. English is an official language in just 29% of the countries represented in the KU Pilot. English is not an official language in 71% of the countries involved.
84% of Charter Member institutions are in a country where English is an official language, and 16% are not.
National libraries in the following countries took part in the Pilot: Great Britain; Latvia; The Netherlands; Scotland and Wales. Five State Libraries also participated. These were: The State Library of Baden; The State Library of Bavaria; The State Library of Berlin and The State Library of Western Australia.

Content Delivery
Developing technical workflows and identifying content delivery and metadata bottlenecks, as well as strategies for overcoming these in order to scale the model, were important aspects of the Pilot. The results indicate that generally better metadata is required of publishers.

Hosting
KU elected not to develop its own platform for hosting and delivery. Instead, it chose to make the most of existing infrastructure by partnering with established hosting and content delivery services. The OA version of each unlatched book is being hosted in the OAPEN Digital Library and by HathiTrust. Some publishers have also elected to host the OA version of each book on their platforms and the possibility of hosting the OA version of books within the British Library's Digital Library is being explored.
KU's decision to work with partners to ensure that content is preserved, available and discoverable provided important advantages: allowing KU to build on existing investments in hosting and discovery services, and to benefit from the visibility and networks of its partners.

Preservation and Discoverability
KU has agreements with CLOCKSS and Portico to ensure that the OA version of each book in the Pilot Collection is digitally preserved. The OAPEN Digital Library is also acting as a preservation platform.
OPAEN provided MARCXML records for the KU Pilot Collection. These were enhanced using data available through other systems and refined to ensure consistency and quality by teams at Duke University Library, Kenyon College Library, Denison Library and the Boston College Libraries. 10 MARC21 records were also prepared. Both versions are available for download, for free, on the KU Collections website.
These records were then sent to: • HathiTrust as part of the package used to load KU content onto their platform • OCLC to load into WorldCat; • ProQuest for use in Summon; • CLOCKSS and Portico for preservation.
KU ensured that metadata provided to these services was complete and of a high quality. The opaque world of how web-based discovery tools integrate metadata into their services is an issue currently being addressed by a number of bodies, including the US's National Information Standards Organisation (NISO). KU has contributed to NISO's work in this area, 11 and will continue to share information arising from the Pilot with groups working towards increased visibility and discoverability of OA content.

Governance
The KU Pilot served as a mechanism for establishing a core group of member libraries that will help to govern the organisation as it moves forward. Each of the libraries that took part in the Pilot became a Charter Member of Knowledge Unlatched, with rights to nominate and elect the Library Steering Committee and Collections Committee, which will play a key role in ensuring that Knowledge Unlatched meets the needs of its library members in the future.

Usage
On 11 March 2014 the first of the KU Pilot Collection books became available via the OAPEN Digital Library. Books became available via HathiTrust several weeks later.
The books in the Pilot Collection were not uploaded onto OAPEN or the HathiTrust in a single batch. Rather, each book was uploaded as it was provided to KU by the publisher. Some of the books in the Pilot Collection were not published until the second half of 2014. As a result, the final Pilot Collection title became available via OAPEN in September 2014.
In spite of this, OAPEN recorded 6,301 downloads of KU books in the first 12 weeks of the Pilot Collection's availability online. Readers in at least 121 countries downloaded books from the Collection. After 24 weeks, the numbers had increased to 12,763 downloads from at least 138 countries.
OAPEN Digital Library usage data is COUNTER compliant. This means that raw download figures have been filtered according to a standard methodology, in order to ensure that an accurate number of downloads is reported, and that activity generated by online bots is excluded. OAPEN work with IRUS-UK, a Jisc-funded repository and infrastructure service in order to produce COUNTER compliant usage data. The period covered by the data is from 11 March to 31 August 2014. The corollary of allowing much wider sharing of KU books is that gathering comprehensive usage data becomes more challenging. It seems likely that downloads visible via the OAPEN and HathiTrust sites are just a fraction of the total number relating to any of the titles in the Pilot Collection.

Conclusions
Networked digital technologies are transforming the processes and institutions of knowledge creation and dissemination globally: remaking relationships between agents (which can include individuals, organisations or enterprises); enabling new forms of collaboration; and demanding new conceptual tools, infrastructure and business models to support activities that are valued by communities. The Knowledge Unlatched proof-of-concept Pilot sits within a broader context of efforts by scholarly and publishing communities to create such tools and infrastructure, and to enable the new business models demanded by disruptive change. It builds on a powerful innovation in copyright: Creative Commons licences (Lessig 2004), and engages with important developments elsewhere in the scholarly and communications landscapes. These include Cultural Science (Hartley & Potts 2014), the emergence of new possibilities for mapping and understanding use and impact (Priem et al. 2010), business model innovation (Hargreaves 2011) and the growth of community norms capable of supporting 'peaceful revolutions' in copyright (Suzor 2015).
The KU Pilot established that academic libraries and scholarly publishers are willing to work together in new ways to enable the OA publication of specialist scholarly books, successfully demonstrating that: 1. Publishers are willing to make high quality, front-list books available on an OA licence in return for the payment of a single, fixed Title Fee by a global community of libraries; 2. Libraries from around the world are willing to work together to share this fee; and 3. That doing so can provide a financially viable alternative to traditional content acquisition models for both publishers and libraries.
By recognising the key role that academic libraries already play in paying for the publication of monographs, as well as the potential for established scholarly presses to change the way in which the costs of high quality publishing are recouped with minimal disruption to their workflows and value propositions, KU was able to design a model that balanced the competing interests of different groups within the monograph system.
The consortium model trialled during the Pilot operated as a simple assurance contract between libraries and publishers: allowing libraries to collectively signal their willingness to pay for the OA availability of specialist scholarly books. In doing so, KU provided publishers with a low-risk opportunity to recoup the costs of publishing book-length works in a manner that met the needs of research funders, authors, libraries and readers more effectively than closed alternatives. The capacity of the KU model to enhance the value contributed to scholarly communication systems by libraries and publishers, and to leverage the value of earlier investments in infrastructure to support OA books, such as the OAPEN Digital Library, were key factors in the Pilot's success. The leadership and good will displayed by KU partners located in the United States, United Kingdom, Europe and Australia, as well as the authors and publishers that engaged with the model, reflect the truly global nature of the communities that care for and about the future of the monograph.
During the pledging period, which lasted from October 2013 until February 2014, 297 libraries from 24 countries signed up for the Pilot Collection, exceeding the original target for library participation by almost 50%. Each unlatched book was uploaded onto the OAPEN and HathiTrust platforms as its publisher provided it to KU. The first unlatched book became available via the OAPEN Digital Library on 11 March 2014, followed shortly after by HathiTrust. Because the Pilot dealt with new and forthcoming titles, rather than backlists, some of the books in the Collection were not published until the second half of 2014. As a result, the final Pilot Collection title became available via OAPEN and HathiTrust in September 2014. Between March 11 and August 28 a total of 12,763 downloads from the OAPEN Digital Library were recorded for the Pilot Collection. This equated to an average of 40 downloads per book, per week, by users from at least 138 countries. 13 In the first six months of the Collection's availability online, each book was downloaded a mean average of 473 times.
These figures are impressive, particularly as they relate to a period when titles were still being loaded onto the OAPEN platform. Viewed in the context of the small printruns of a few hundred titles that have become standard for most monographs, 14 they represent an extraordinary achievement: highlighting the capacity of OA monograph models to connect specialist scholarly books with readers located all over the world, regardless of their access to a university library, or their capacity to pay for access. The unequivocally positive outcomes of the Pilot highlight the power of international coordination and trusted, neutral, intermediaries to enable innovation in scholarly communication.