The academic library is often characterized as ‘the humanist's laboratory’ the journals and particularly the books it contains analogous to a biologist's cell samples. Libraries and librarians play a crucial role in supplying humanities researchers with the raw materials that they use to undertake scholarship. As such, any open access (OA) monograph model will need to take their preferences and priorities into account, and ensure that it can supply and promote OA books on a par with those published under a more traditional business model.

Academics themselves recognize the importance of the library as a source of information. Studies consistently emphasize the relative importance of the library for books compared to journals, and for humanities and social science (HSS) researchers compared to those working in other disciplines., In return, librarians recognize that their job is to meet the needs of the academics and students who use their services, ensuring they shape their offer to give users what they need – even when those users are not entirely sure quite what that is.

This article presents findings from interviews with library and repository staff at four UK universities, undertaken as part of the OAPEN-UK project; we then reflect upon how these findings chime with the University of Huddersfield's experience.

OAPEN-UK

E-books and the library

The question of open access monographs is closely related to more general moves towards e-books within academic library services. In fact, e-book expenditure has tended to be low as a proportion of overall book expenditure in most academic libraries: figures from 2009 indicate that e-books accounted for around 10% of overall book expenditure in SCONUL member libraries, and more recent interviews with librarians through the OAPEN-UK project suggest that things have not changed significantly since then. Although many libraries did have an electronic-first acquisition policy, they stressed that in some instances the content they required was still not available in electronic format (although this seems to relate more to textbooks than to monographs). In part this may also reflect academic resistance to e-books; although a recent survey suggested that most academics used e-books relatively frequently, it also identified a distinct lack of support for e-only collections of books. This support was especially low among arts, humanities and social science (AHSS) researchers.

“The academic library is often characterized as ‘the humanist's laboratory’…”

Librarians in our study affirmed that many academics prefer to have print copies of the books they use, even when the library already owns or licenses an electronic copy. In some cases, librarians are happy to buy these, while in others they will encourage the researcher to use the existing electronic version unless they have a very good explanation for why this is inadequate. When money was available at the end of the year, librarians were more likely to support multiple formats of a single text; where budgets were tighter, money was less likely to be forthcoming. This has important ramifications for the likely success of certain OA business models, outlined in more detail below.

“…Librarians appreciated the simplicity of having everything in a single location for users …”

All the librarians that we interviewed use one or more e-book aggregator to provide access to monographs and other types of books. Librarians appreciated the simplicity of having everything in a single location for users, but felt that the policies and particularly the digital rights management (DRM) imposed by aggregators could lead to a less satisfactory experience for researchers and students. Indeed, this was reflected in some of the interviews we undertook with academics, who spoke of confusion about whether checking a reference in a book would cost the library money (in some models it would) and frustration at the three-day expiration of books that they download to read later.

There are thus two concerns for librarians thinking about how OA books might fit into their existing e-book environment. The first is whether the way those books are developed, presented and enter the supply chain will be consistent with existing practices for subscription e-books. At the time of our interviews the Directory of Open Access Books (DOAB) and the OAPEN library, two primary indices of peer-reviewed quality OA monograph content, had only just been integrated into major discovery systems such as Summon and EDS, and most interviewees were not aware that their web-scale discovery systems probably would lead users to OA monograph content (although their library catalogue may not, depending upon how they had set it up). But if a user went directly to the e-book platform used by the library and searched there, they might not necessarily find the open access content.

The second problem is, to some extent, in tension with the first. Some interviewees mentioned a concern that OA books, if supplied by a third party such as an aggregator, would become subject to all the DRM processes routinely imposed by such parties and thus lose many of the advantages of open access. One interviewee identified that this is already evident when you compare the version of a paid-for book made available on the publisher's own website with the version supplied by aggregators. The publishers that we interviewed as part of a separate work package were even more concerned about this.

Collection, acquisition and preservation

Most librarians involved academics in some way in the selection of books for both acquisition and disposal. The method of doing this varied between institutions and in some cases even within institutions, with different departments operating more or less formal methods to communicate their needs to librarians. But in general, librarians felt that the arts and humanities were particularly consultative when it came to monograph selection. Librarians – and the academics that we spoke to in their institutions – were clear that the library often reflected the specific research interests and specialities of academic staff. Many institutions had also begun to experiment with patron-driven acquisition (PDA), which is primarily geared towards students rather than researchers.

Acquisition of print and electronic content is primarily done through large-scale third-party suppliers who often supply metadata alongside the content itself. Some interviewees talked about sourcing out-of-print books via Amazon or AbeBooks, but the overwhelmingly common way to acquire new content was via large-scale aggregators. Any OA books will need to be part of this supply chain, in the same way that they need to be discoverable and accessible in supply chains.

Librarians we interviewed noted a long-term trend with electronic resources which moves away from holding or hosting content locally to long-term availability via third parties. Although many expressed a wish for a greater degree of control over the content that they license, or more flexibility in selecting this content, few wished to take on the job of maintaining and preserving their e-book collections in-house – this was generally seen as inefficient and a retroactive step. In an open access model, this creates all kinds of challenges. One interviewee noted the lack of a contractual and financial relationship between the publisher and the library which would guarantee long-term access to open content. This was a particular concern as far as new open access presses are concerned, but librarians gave examples of even well-established publishers pulling content without clear recourse for those who had previously subscribed to it.

New roles and responsibilities

Several librarians we interviewed mentioned that their university was considering setting up some kind of institutional press, although the nature of this operation, and the relationship with the library, were very different in different institutions. Nonetheless, most were unsure that monographs were currently within the scope of the very nascent plans at their institutions.

“… there may be some scope for including monographs in repository collections.”

We also interviewed repository managers in our case studies to understand how a green OA model might work for monographs. As yet, this is not a model that has been extensively trialled or studied, but our interviews suggest there may be some scope for including monographs in repository collections. In principle, interviewees felt that monographs were sufficiently similar to journals to make inclusion relatively straightforward from a technical point of view, although tweaks might be needed to permit, for example, associating chapters with each other.

But there may be bigger problems from a licensing point of view. Most of our interviewees ask depositors to check the licences of the work that they are sharing through the repository, but will subsequently re-check before making content live. This may reflect a fairly realistic estimation of the attention that researchers pay to their contracts. Several interviewees mentioned the challenge of establishing specific terms and conditions for green deposit of books or book chapters, as many publishers have not developed or publicly expressed these. One interviewee suggested that this problem may be exacerbated by the large number of small monograph publishers, requiring institutions to maintain contacts with a lot of individuals to establish and confirm policies which small publishers are particularly unlikely to have.

Repository managers were, like the librarians discussed in the section above, largely unwilling to be responsible for guaranteeing access to monograph content in the long term. Interviewees questioned the efficiency of every institution retaining its own copy of open access content and felt that, unless properly indexed by search engines, it might be difficult to ensure that content held in a repository could be discovered by users.

“Librarians were ambiguous about their ability to underwrite first-copy costs.”

There may also be a role for libraries as funders of OA monographs. Many of the business models currently being trialled for OA monographs depend upon contributions from the library, either by paying towards first-copy costs, or by purchasing alternative formats of a monograph beyond the basic HTML or PDF. Librarians were ambiguous about their ability to underwrite first-copy costs. Most felt that this could not be done from existing library budgets, although some were willing to contribute small amounts of money to experiments which aggregate across libraries to fund a small number of books. Similarly, on the alternative-format question, some librarians suggested that they would expect this to operate in a similar way to their existing arrangements, i.e. if money was available or an academic made a compelling case for alternative formats, they would be happy to buy them, but that they would not necessarily expect this to become a routine way of acquiring content.

University of Huddersfield

E-books and the library

The Huddersfield experience has shown that e-books seem to have been slower to take off than e-journals. However, e-book expenditure has been rising over the last couple of years, as has use. The majority of recent expenditure has been through the acquisition of individual titles purchased from library suppliers and through PDA. This has built upon a number of aggregated packages and small publisher collections that Huddersfield has had for some years, as the Library has not yet bought any larger publisher collections.

The Library is in the process of implementing a new collection management and development policy, which includes the introduction of a loose formula for reading list material. Although this is more of a guide, based on student numbers in a particular module combined with the in-depth knowledge of each subject team, the policy is to always purchase an e-book where available. It is important to note that, although many popular textbooks are not available at an affordable price, a reading list does not always feature textbooks. For example, our history reading lists cover a wide range of titles and many of these will be academic monographs – therefore the availability of monographs can be very relevant to undergraduate studies as well as to research.

Although e-books may have been slow to catch on, usage is now on the increase and, in addition, Huddersfield has seen a steady decline in print book usage per student. The rise in access to e-books has almost been in spite of DRM, which continues to be a major source of complaint from users. Users also comment on confusion over the wide variety of platforms. There is a big difference between e-journal usage and e-book usage in that users are used to going straight into a journal, downloading the PDF for printing or reading later, and moving on. DRM and the large variety of platforms for e-books confuse the users in some situations, and the inability to save the PDF for more than a few days often causes irritation. Finally, there are also many issues round mobile and access for users with a disability that are not encountered to the same degree for e-journals.

“DRM and the large variety of platforms for e-books confuse the users …”

The University of Huddersfield, like a growing number of libraries, has moved away from the library catalogue. Since August 2013, the Library's web discovery tool, Summon, is the gateway to both the print holdings of the Library and its online resources. The addition of OAPEN to the knowledge base means that OA monographs can be searched alongside the rest of the Library's collection; the addition of DOAB will increase the amount of accessible content. However, the Library does not seek out other open access material on publisher websites due to staffing constraints, therefore if an OA book is not available via OAPEN/DOAB (or a limited number of other platforms) it is not discoverable to the researcher via Library resources.

Collection, acquisition and preservation

It is because of some of the concerns expressed by the community around the ‘quality’ of open access material that Huddersfield does not simply turn on every open access collection it finds within the knowledge base. Services such as OAPEN and DOAB go a long way towards assuring that quality. However, it is interesting to note that the same stringent quality checks do not always apply to purchases of e-book content. This is not a contentious statement aimed at academic publishers, or academics or librarians, but if some of the titles held within the large aggregated e-books packages are investigated, there are certainly some ‘fillers’. The wide use of these large packages also raises the question that many have about preservation for open access: there is no preservation clause in these popular packages – titles move in and out of the collection on a month-by-month basis. In fact, if a title in a package proves popular or is added to a reading list, the title is often purchased a second time via a book supplier or publisher to maintain access.

Of course, preservation is a concern for OA monographs partly as it is also more expensive than for journals. New open access presses, particularly scholarly library publishing, often operate on a very small budget. A quick look at Portico shows that e-journal preservation is far more affordable than e-book preservation. It is for this reason that The University of Huddersfield Press has chosen to preserve its OA journals via Portico, but not its OA e-books. The lack of a strategy in the UK for the preservation of UK OA monographs is an area of concern. However, it is hoped that the National Monographs Strategy will help to address this in some way.

“… preservation is a concern for OA monographs …”

New roles and responsibilities

Scalability of library publishing services

Hahn compares library publishing programmes to the traditional publishing model, describing them as a truncated list of services. This is a potential advantage as libraries seek to identify the ‘minimum suite of services’ that would be required to operate a successful journals and monographs publishing model. However, Hahn goes on to question the sustainability of library publishers: the 2007 ARL survey found that very few library publishers were able to ‘…support even 10 journal titles or more than a handful of monographic works.’ This is exactly the case for new scholarly library publishers such as the University of Huddersfield Press. Otherwise these new ventures could prove a victim of their own success: a view echoed by comments made by Georgia Tech, where the Library has hesitated in more aggressive marketing of the Press due to fears that this could generate more demand than the Press could satisfy. The Press has to identify the resources it needs in order to expand successfully. However, Bryn Geffert, librarian at Amherst College and director of the new University Press there rightly says, “We want to do a few things well, not overextend.”

Staffing

Borchert described the establishment of OA journal publishing at University of South Florida in 2007 as being carried out with no increase in staff. Crow et al found similar figures in their survey. The number of staff allocated to publishing activities ranged between 0.9-2.4 FTE, with staff dedicated to Library publishing programmes described as ‘relatively rare’. Kosavic identified this as a service-level boundary, a ‘crucial step’ in identifying a manageable workflow in order to ensure that too much is not taken on. Georgia Tech operated on an informal basis for its first five years, with staff funded from the Library's standing budget; Huddersfield is in a very similar position. This is really the crux of the matter and was also identified by Hahn: either the Library needs to reallocate a greater staffing resource from the Library or new resources need to be identified. Having the same staff running the repository, the journals team, gold open access and the University Press is clearly not sustainable if the work of the Press increases.

Repositories

Although Huddersfield has had some success with chapters being made available on green open access, this is a time-consuming process. Often monograph publishers simply don't reply to requests for information about their self-archiving policies, and without their permission the repository is unable to make deposited work publicly available. It is hoped that discussion about establishing a UK pilot to implement the OAPEN deposit service will be successful as it will provide better discovery for the monographs and facilitate better access for repositories to harvest the full text.