This post was written by Graham Stone, Senior Research Manager, Jisc Collections.
Key studies, such as the OAPEN-NL project, have shown that open access has a positive impact on the usage and discovery of monographs. However, a recurring theme for open access monograph publishers is that of discoverability, dissemination and metadata.
In 2017, the Jisc report Changing publishing ecologies: A landscape study of new university presses and academic-led publishing suggested that best practices in metadata were drawn up as the quality of metadata at New University Presses and Academic-led publishing initiatives was at various levels of maturity. This view was confirmed at a European level by the Knowledge Unlatched report to the OPERAS project on the visibility of metadata, which stated that: “The metadata held and managed by OPERAS partners is inconsistent and variable in quality. Collecting and aggregating data from multiple OPERAS partners was a challenge due to inconsistency in bibliographic metadata processes and formats”.
New University Presses (NUP), Academic-Led Publishers (ALP), and open access presses in general, have difficulty in accessing the library supply chain – the channels that library acquisition departments use to buy print and e-book content. As part of the ALP interviews for the landscape study Rupert Gatti of Open Book Publishers noted that it would be helpful to have a service that “looks at how to bring academic content into the catalogues and the digital learning environments of the universities and to allow universities to also relate back to the publisher, so that there is a flow of information going back both ways.
In order to take the recommendations from the Jisc report forward, a community workshop was held in July 2018 to surface issues with key stakeholders for the following problem statement:
“OA publishers have difficulty accessing the channels that library acquisition departments use to buy print and e-book content”
The workshop brought together experts from NUPs, ALPs, book suppliers and distributors, metadata suppliers, libraries and other experts in OA publishing to discuss the above statement. The core aim of the workshop was to allow the experts to share their experiences and knowledge in order to get a better understanding of the supply chain and to gain insight into the problem statement.
Stakeholders from NUPs, ALPs, book suppliers and academic libraries were asked to give a brief overview and an outline of the problem statement from their perspective as part of the publishing workflow/library supply chain. Other experts at the table then shared their knowledge to understand and develop these issues in relation to the problem statement and to consider what learning needs to take place to address the problem statement.
The four discussions are captured below. In addition, these discussions highlighted four key areas, which the group agreed needed further development. These have been presented as a set of recommendations and suggestions for best practice.
New University Presses
Library-led New University Presses can be defined as a “set of activities led by college and university libraries to support the creation, dissemination, and curation of scholarly, creative, and/or educational works” (Library Publishing Coalition).
It was highlighted that the problem statement was not just an open access problem, but an issue that all small presses face. The complexity of the acquisition process proved an issue for these presses. For example, the Jisc landscape report found that most UK NUP/ALPs had 1-1.5 FTE staff and that specialist knowledge of library systems and the supply chain was lacking. Furthermore, presses found that understanding the metadata required for the supply chain, such as ONIX, can be intimidating for small presses. There are also lots of new channels, constant change and it is hard to keep up. It was felt that there was a need for a degree of standardisation to cope with metadata channels. There were also issues around chapter level metadata in that there is not always a means to capture it.
It was suggested that suppliers do not always see the benefits of ‘free’ and that NUPs have paid and unpaid channels, for example most presses publishing monographs provide a free open access version, usually a PDF and a print on demand for sale. However, presses have difficulty registering open access in non-OA systems as there is no option for £0. A key point was made that NUPs do not care how their books are read, whether it is free or paid, just whether it is read it at all. The priority is readership and not sales. Presses also need to demonstrate the value of open access and it was felt that this was a key issue for the library supply chain stakeholders to understand.
For this reason, NUPs regarded journal articles as more discoverable than books and this resulted in a perception that the arts, humanities and social sciences were being disadvantaged.
What is needed are practical workflows and durable solutions with an avoidance of duplication in metadata creation. This can only come from shared knowledge within the supply chain.
Looking at library discovery in the supply chain, it was noted that one of the biggest issues was how to work with library systems as there is nowhere to add ‘free’ to MARC records. NUPs wished to know how they could work with libraries to help make OA content discoverable. This led to a discussion about ISBNs and presses were asked if they attributed separate ISBNs to each format. It appears that this is done by some and not others. Therefore, this is a potential area for standardisation of processes. However, limitations with ONIX mean that metadata may not be captured for multiple ISBNs.
Finally, it was noted that there needs to be discussion with the British Library around possible inclusion in the Cataloguing-in-Publication (CIP) Programme, which “provides records of new and forthcoming books in advance of publication in the United Kingdom and Ireland”. Due to the importance of Library of Congress categorisations and metadata, which publishers do not tend to engage with, it was suggested that this should also be taken to the Library of Congress committee for discussion.
Academic-Led Publishing can be described as “a publishing initiative set-up and run by academics… …Academic-led presses are most often not-for profit, independent, highly ideological entities, set up to provide an alternative publication route to the commercial presses or to support the open access publishing of books for example” (with thanks to Janneke Adema for this definition).
Academic or scholarly-led publishing has developed from a view from scholars that the current options to authors are not delivering what the scholars want. For example, different content, design, data, interaction, etc. Open access creates these new interactions with readers and research. As a rule, ALPs do not enter into competition with each other, rather they see the need for collaboration in order to progress a common goal.
However, ALPs also find dissemination difficult. Libraries were asked what they do to allow researchers to be made aware of research content made by researchers and what does the library community want and what can presses do to help achieve this.
The ALP view echoed that of the NUPs in that there was need for a process to generate metadata easily, such as MARC, ONIX, KBART and all of the other formats required at different stages in the supply chain.
It was highlighted that there could be a possible solution led by Jisc’s National Bibliographic Knowledgebase initiative and that data could be converted to MARC 21 format, where libraries could download records with some sort of open access identifier. At this point, an important differentiation was made between discovery and acquisition and that the two areas had different sets of systems and workflows. Discovery via library systems is an easier area to resolve through platforms such as the Directory of Open Access Books (DOAB) and OAPEN. However, metadata held in library systems tends to include items that have been ‘purchased’, which reinforces the need to resolve metadata issues in the library supply chain.
Regarding the embedding of open access material in digital learning environments for teaching, it was felt that open access content was typically presented in a very informal way, whereas paid for content was often formally organised because of purchases and the flow of metadata through the library supply chain. It was perceived that discovery in these systems via the library supply chain could assist in assumptions that open access content, particularly monographs, were not peer reviewed and were of low quality than paid for content.
It was noted that there were tools in place to bring in open access content, such as open educational resources into learning environments without necessarily using the library supply chain. For example, by using platforms such as Kortext and Talis Aspire.
Library books suppliers
Taking the difference between supply and discovery further, book suppliers noted that for the supply chain, MARC records were not required. Metadata needs to be supplied in ONIX and that this is even better if it is part of a BDS/Nielsen Bookdata feed. Suppliers can take data direct from publishers, but that would mean that every publisher would need to supply all book suppliers, a task that would be too big for any small-scale publisher.
It was noted that the more fields that could be supplied in the metadata the better, such as content, summaries, abstracts. It was also noted that BDS do not charge the publisher for this service, the charge is passed onto the book supplier. Therefore, the supply of metadata has costs even if the publications are free.
ONIX does cater for open access. However, an appropriate licence, such as a Creative Commons licence is more important as the definition of open access may differ from publisher to publisher. Therefore, the data the suppliers are sent is as good as the publisher provides. If there is no reference to open access and an appropriate Creative Commons licence, then there will be misinformation between the different parties in the supply chain.
It was suggested that there needs to be further discussion with open access publishers in order to supply the whole library supply chain. This would entail understanding what NUP/ALPs see as supply, and how libraries want the data to be provided. Business models need to be understood by all parties in order for the supply chain to be fully mapped. There may be a clear advantage to ALP/NUPs in that they could react faster than commercial publishers to engage.
The library view confirmed the complexity of the library purchasing landscape. A particular point for UK universities was raised, that of the national books purchasing framework agreement. This agreement commits those universities that have opted in to the agreement to make a certain percentage of spend through the suppliers listed in the agreement. This, in part, is the reason many library acquisition teams are not aware of open access monographs. If they are not listed in the database of their chosen suppliers, they are unlikely to look elsewhere.
Examples were shown where suppliers and publishers are not listing the fact that a print version of a monograph is also available free via open access. This provides more evidence of a failure in the library supply chain for open access monographs. Open access versions are invisible to library purchasing systems.
A question was asked whether, for future framework agreements, an open access element is included.
It was noted that libraries are very student focussed in everything they do. Therefore, this features highly in content acquisition where libraries need to bring as much open access content in to students and researchers as they possibly can. However, this does not always fit with library’s collection management and development policy. Although the policy is often focussed on providing value, it often does not contain a section on the acquisition of open access content. This supports the ALP comments about open being less formal than purchased content.
It was suggested that a series of recommendations were needed for libraries and that there was a role for SCONUL in this area, together with making open access central to information literacy.
The group agreed that themes had been surfaced in the discussion. These included two practical areas, the library supply chain and metadata and two aspirational areas, cultural change in the acquisition process and new forms of content. These are further developed below.
Library supply chain
There is a crossover between discussions on the library supply chain and metadata. However, it was decided that it was important to be able to map the library supply chain for open access monographs so that all parties reach equal understanding of each other’s processes and workflow. This may include some reprioritisation in order to accommodate open access monograph supply. To avoid further confusion, it is important for open access monograph publishers to understand who the audience is for the different types of metadata. For example, library suppliers, library acquisitions teams and researchers and other end users.
As part of mapping the open access monograph library supply chain, it is important to understand how the various costs will be covered and by whom.
It may be useful to map the discovery workflow in addition to the library supply chain to understand any commonalities. In this instance, the discovery workflow is described as the process of making open access monographs available through other means, such as DOAB and research discovery systems.
It was agreed by the group that there is a need to agree a minimum metadata requirement, which could then be used in all metadata in the library supply chain, such as ONIX, MARC, KBART etc. This would go some way to allow all parties to understand what they each mean by the term metadata and what it is describing. The minimum level of metadata must include ISBN, chapter level identifiers and abstracts. There is potential to scale this model internationally.
Two possible solutions were put forward for further development. This could help to reduce the number of multiple MARC records in the supply chain.
- An NBK solution
It was suggested that a two-step approach could be offered. Firstly, libraries and consortia with presses who had the capacity to create their own metadata in a consistent format could upload the data directly to NBK. This could then be pushed to the library supply chain in whatever format was required.
For smaller presses without the means to create their own metadata, presses could also be offered a web form with the required metadata fields. This would also allow data to be pushed to the supply chain, as well as providing the presses with their own MARC records using an open license.
- BDS solution
An alternative solution would use an online form via BDS to create records, which could then be converted to ONIX, XML etc. This model would be free for presses, instead it would include a processing charge for end users. This is the standard BDS business model and would require licensing models to be in place for the metadata
Any minimum standard would have to ensure that there was a way to enable library acquisitions teams to see that there is an open access version of the monograph in addition to any print copy via the supplier
Regarding the discovery workflow, if a press had been pre-approved by DOAB there is also potential, as long as the correct licence was included, to send the records to DOAB for inclusion in library discovery systems.
Cultural change in the acquisitions process
On a more aspirational note, it was suggested that, as part of the transition to open access for monographs, libraries need to reposition their thinking on collection management in internal communications and at a national level. Different teams within the library deal with open access in different ways. For example, discovery is more easily addressed via the delivery of metadata to DOAB. However, a major issue for acquisitions teams is how to recognise zero cost for open access while the print version is available.
A possible way forward in the UK is to create a position paper outlining the problem statement and to recommend a way forward at a strategic level. It was thought that this could be proposed by Jisc Collections as part of its Content Strategy Group. It was then suggested that further work could be progressed under the auspices of SCONUL/RLUK if this was seen as appropriate.
Key themes were agreed as:
- Buy-in by Library Directors around the idea of centrally funding OA monograph initiatives, e.g. ensuring that these became part of the central book/monographs budget based
- Ensuring that the acquisition of open access content featured in the library’s collection management and development plan or policy
- Measuring the impact of open access monographs. For example, in reading list software, or via OAPEN COUNTER stats if appropriate.
New forms of content
Discussions regarding the library supply chain and appropriate metadata necessarily centred around more traditional forms of monograph publishing, albeit an electronic version of the printed monograph, or ‘print under glass’ version. However, many presses and potential authors are interested in experimental content, but this often does not get noticed as doesn’t ‘fit’ with traditional forms of publishing. It was felt that open access may well act as a trigger for these new forms of content.
The group considered how to take the next step and address any issues arising from these new forms, such as multiple platforms and fluid versions. For example, how should presses make suppliers and libraries aware of new form of content?
Comments and Feedback
Jisc Collections are now exploring the options discussed in the workshop. As such, we would welcome further comments and feedback from the community. This will assist us in defining where our focus should lie in partnership with the major stakeholders. We would also like to hear from stakeholders who would like to express an interest in further work.
2 replies on “OA monographs discovery in the library supply chain: draft report and recommendations”
Since this blog post was released, ScholarLed (a collective of six Academic-Led Publishers) have posted a web page about the types of infrastructure projects they have been thinking about.
Project 1 under the open infrastructure tab is directly related to this blog post and can be found at https://scholarled.org/
Unglue.it, a program of the Free Ebook Foundation (a not-for-profit registered charity in the US), has been working (slowly but steadily) in this area for the past 5 years. Publisher participants work with us to upload metadata, either with a custom loader or a web form. Their catalogs can then be exported in MARC, ONIX, and OPDS vi our API. All the code is open source, using Python and Django as a platform.
We’ve observed the very issues noted in this post, particularly the hesitance of libraries to put OA monographs in their catalog. Working with library discovery vendors will be key to the success of this effort.