Scholarly communication and open access: Perspectives from Korea

This article reports on scholarly communication and open access (OA) in Korea. Drawing on a range of databases, articles, and reference sources, it provides unique insights. In contrast to the UK/US model of scholarly communication, in Korea, most scholarly journals are published by discipline‐based scholarly societies and research institutes affiliated to universities. Payment for publication is the norm, and typically features article processing charges and scholarly society membership fees for both OA and toll access (TA) journals. Online access to journals in Korea is provided by commercial vendors who enter into contracts with the scholarly societies for exclusive use. Three online access models apply – TA, gold OA, and dual access – with the use of these models varying between disciplines. In parallel with this access provided by commercial vendors, there are a number of government‐funded open access repositories (OARs) to which university researchers are requested to deposit their research outputs, as well as OARs run by universities and other research institutes.


INTRODUCTION
Open access (OA) to the findings of scholarly research is a growing international movement, grounded on the principle that publically funded research should be freely accessible online immediately after publication. In addition, amongst the aspirations for OA is that it would lessen the strength of the commercial publishers, particularly their capacity to generate further revenues through ever-increasing journal prices (Satyanarayana, 2013). OA was viewed as a route to increasing access to research findings whilst also increasing the visibility of research. In particular, in less-developed countries, and other countries such as Korea and India that are investing in developing research cultures and expertise, it was expected to enhance access to scholarly resources, enhance research infrastructure, and facilitate an erosion of the divide between local and global journals (Mukherjee, 2014). It would seem that there is still a long road to travel before this can be delivered for developed and developing countries alike. Specifically, in the context of developing countries, analysis of involvement on a country basis through examination of the Directory of Open Access Repositories (DOAR) and the Directory of Open Access Journals (DOAJ) suggests that there are many countries where involvement in OA initiatives remains limited (e.g. Sahu & Arya, 2013). Furthermore, with the exception of India and Nigeria (which use English language for some of their scholarly communication, thereby navigating any potential language barriers to more global scholarly communication), there has been very limited research into OA initiatives, policies, and challenges in countries other than the UK and the USA. However, there is evidence to suggest that research funding and infrastructure, together with scholarly communications and publishing traditions, may have a significant influence on the engagement with and impact of OA initiatives. In addition, it is acknowledged that OA is a disruptive innovation (Lewis, 2012), which is likely to herald significant shifts in the ecology of scientific communication (Jubb, 2013).
Instead, there are both OA and TA journals, and both require payment of an APC. Journals in both of these categories are ultimately available for free access via OARs and under toll/ subscription access via commercial database providers. We term this approach 'dual access (DA)'.
The aim of this study is, then, to contribute to the knowledge on OA models adopted in different contexts and thereby complement the significant literature on 'western' OA models and to advance understanding of the challenges associated with global OA. More specifically, the objectives of this study are to: • Profile scholarly communication in South Korea.
• Investigate OA journal publishing (OAP) in South Korea.
• Identify and describe the major OARs in South Korea.
• Discuss the OA model adopted in Korea, and compare it with the 'western model' of OA.
The next section of this article briefly outlines the method adopted in this research. Then, the approach to scholarly communication in South Korea is reviewed in two core sections: (1) scholarly journals and OA publishing in Korea and (2) OARs and green OA. The Discussion section compares OA in Korea with the 'western' model of OA, focusing on types of OA, intellectual property rights, APCs and other payments, stakeholder groups, and disciplinary differences and similarities. The article concludes with recommendations for practice and policy.

METHODS
In drawing a profile of OA in Korea, this article draws on a range of secondary sources and databases. This, for instance, involved identifying and inspecting the data available on the websites of the seven major OARs in Korea and drawing on statistics produced by the Korea Citation Index (KCI) Korean Journal Database and other sources, such as the websites of open access korea (OAK) Central, KoreaMed Synapse, ScienceCentral, and dCollection. In addition, the article is informed by extracts from previous work by one of the authors, previously only published in Korean (e.g. Table 2).

Scholarly journals and OA publishing in Korea
Scholarly publishing in Korea According to statistics provided in association with the KCI, in 2016, there were around 8,600 academic organizations in Korea, of which 3,441 are scholarly societies and 5,047 are research institutions affiliated to universities (https://www.kci.go.kr/ kciportal/po/statistics/poStatisticsMain.kci). Between them, these organizations are responsible for producing numerous research journals, as shown in Table 1. Over half of these journals are listed in the KCI and benefit from the quality endorsement that this offers. In humanities and social sciences (HSS), 44% of the total are included in KCI, whilst in science, technology, and medicine (STM), 39% of the total are included. Many of the journals listed in KCI are published by scholarly societies, specifically 74% in HSS and 96% in STM.
In order to understand OA publishing in Korea, it is necessary to understand the scholarly publishing process (Joung, 2011;Kim, 2014;Lee, 2005;Woo, 2010) (Fig. 1). Researchers submit articles to scholarly societies, who are responsible for peer review, publication decisions, proofreading, and, sometimes, editorial design and printing. Articles are then deposited and made available through either one of the main OARs (e.g. KCI repository), societies' websites, academic libraries through a licensing arrangement that involves commercial database providers, or through sales to individuals.
This model of scholarly publishing leads to three types of online access to Korean journal articles: TA, gold OA, and DA. TA is fee-based, involving commercial vendors providing online access to articles to individuals for a fee and to libraries through a licensing agreement. The model is based on full journals, and access is through the commercial database providers' services. and through the websites of some scholarly societies. Scholarly societies receive support from the research and information service institutions to enable them to provide free online access to their journal articles. Under the third approach, DA, journals can be accessed for free by the public through the scholarly societies' websites and/or OARs whilst also being available through commercial vendors' databases, where individuals typically pay a fee or access the journals through their university library, which has taken out a license with a vendor.
According to Joung (2011), 34% of KCI journals were TA, 34% were DA, 22% were OA, and for 10%, the status could not be  determined due to lack of information provided. The main reason why publishers are prepared to offer their journal as OA is that it enhances the citation rate and thereby the ranking and reputation of the journal. As Table 2 shows, the distributions between OA, DA, and TA differ significantly between fields, with a significantly larger proportion of STM as opposed to HSS journals being OA; engagement with OA in medicine and agriculture is particularly high. These differences may be influenced by the existence of OARs in some disciplines, such as the Society Village for Science and Technology, managed by KISTI, and KoreaMed Synapse, managed by Korean Association of Medical Journal Editors (KAMJE), covering STM and medicine, respectively. OARs are discussed further in the next section. Certainly, the percentage of DA journals in agriculture is much higher than in other fields, probably due to the absence of a significant OA repository in this field.

Gratis, libre, and intellectual property rights
The license for OA journals is usually for access only and does not grant permission to reuse, such as in reproduction or transmission to the public. This OA is gratis OA, relating to the removal of price barriers only according to Harnard's terminology (Suber, 2008). Very few journals declare an OA policy or adopt an OA or Creative Commons License (CCL). This is in keeping with the tradition in Korean scholarly publication, in which the intellectual property rights for an article typically remain with the author. For example, according to Hong (2008), only 35 of the 906 KCI journals in 2007 regulated the authors' rights regarding their article. Furthermore, for 66.2% of the journals, no copyright information was provided, suggesting a laissez-faire approach to intellectual property rights. There is some evidence that, with the advent of OA initiatives, this situation is changing. The number of libre OA journals (involving the removal of price and at least some permission barriers; Suber, 2008) (Son, 2014). Most of these are HSS journals. Also, according to a recent report, only 13% of KCI journals had adopted CCL (National Library of Korea, 2015). Table 1 shows that most of the highly regarded journals, that is, those listed in KCI, are published by scholarly societies. In addition, most OA journals in Korea were also published by societies.

Publishers of OA journals and income sources
According to Joung's research (2011), 76% of OA journals were published by societies, 13% by research institutes affiliated to universities, and 11% by research institutions funded by government or non-commercial research organizations. Recent research (Son, 2014) on the KCI journals published by university research institutes showed that 85.8% were DA journals, 9.4% were OA, and 4.72% were TA.
Given that many journal publishers are scholarly societies or research institutes in universities who are keen to both generate income from their journal publishing activities and to build their viability and reputation in the research field, income from OA journals is a very important issue. There are various income sources, including APCs, advertisement revenue, membership fees, and grants from the government or from research funding agencies. One of the most contentious aspects of any OA model is APCs. APCs are a charge for publication and are paid by the author, their research funder, or research institution. As shown in Technology Societies]. In addition, some journals have some advertising revenue (45.3%), and others require a higher APC for funded articles and/or a board member fee (the fee associated with a special type of society membership), in addition to the membership fee, when academics join the society.
APCs vary significantly, from a minimum of £0.88 to a maximum of £44 (Joung, 2010). This diversity is also evident with  (22) Other 27 (7) 57 (12) 11 (18) 6 (15) 8 (9) 19 (11) 12 (8)  4 (6) 144 (10) DOAJ, for which the APC range was between $8 and $3,900 (Solomon & Björk, 2012). In further analysis, which compares APCs for OA and TA journals, Joung (2010) Table 4 shows the most important OARs in Korea. All except KoreaMed Synapse are government-funded initiatives. Each has a management agency that is either a research funding agency (such as NRF); a research and information institute, which provides research and information services and, in some cases, research funding; or the National Library of Korea. In addition to these OARs, many universities and research institutes have OARs. Twenty-six universities use the OAK software and provide access to their OARs through the OAK Portal. Some OARs

Overview
In keeping with its aspirations and investment towards the development of its research capacity, there has been considerable interest in the development of OA initiatives in South Korea over the past decade. These initiatives include those concerned with OA journal publication and with OARs. These developments pose some of the same challenges for Korea as they pose for western economies, in terms, for instance, of tensions regarding the role of commercial publishers and database providers, issues associated with the shifting economics of scholarly publishing, and concerns regarding intellectual property rights. However, these issues have been and will continue to be worked out against the backdrop of the rather different tradition of scholarly publishing in Korea. This section discusses some of the unique features of the OA model in Korea, comparing it with the more widespread model in the UK and the USA. In general, key stakeholder groups are consistent across contexts, but their precise roles and the way in which they interact may vary. In addition, there is some commonality regarding the models of OA adopted in different countries, but the precise nature of these and their relative importance varies. Intellectual property rights and their protection, alongside payments and funding for OA initiatives, are two other themes that recur. Finally, disciplinary differences are broadly consistent with those in western economies.

Stakeholder groups
Scholarly societies have a significant tradition of involvement in scholarly publishing, seeing academic journals as important for their reputation and the development of their discipline/profession, as well as offering an income stream and a benefit to their members. However, Solomon (2013) found that for Scopus journals, only 26.4% of OA journals were published by a society publisher and 23.6% by a university publisher. This is much lower than is the case in Korea. This is quite probably because in western economies, many scholarly societies have, in recent decades, outsourced the publishing processes associated with their journals to commercial publishers. These commercial publishers, under a license, take responsibility for managing the interface with authors, including managing a manuscript submission system, liaising with or appointing journal editors, copyediting, and other production processes. In Korea, these processes are still mainly performed by scholarly societies. As such, they potentially have a more direct influence on OA processes and policies. On the other hand, as a disparate group of organizations, they may find it more difficult to make their voices heard and to achieve the impact on OA approaches that major publishers can exert in western economies. Both the government and the database providers are keen to work with societies and acknowledge their pivotal role in OA initiatives.
Governments, more specifically their research funding agencies, have increasingly extended their role in research policy to dissemination of research outputs. However, in the early stages of OA in western economies, OA was very much seen as an ideological stance, supported by researchers themselves, and was driven by the belief that the outputs of publically funded research should be open to all (Crawford, 2015). It is only relatively recently that research funding agencies in, for example, the US and the UK, have started to fund the publication of research outputs by paying for APCs and to mandate OA publication. This has led to commercial publishers offering a gold OA route for the publication of articles in their previously subscription-only access journals. While in both Korea and many 'western economies, governments and their agencies have supported OARs and OAP, the Korean government's involvement in OA is unusually high. The  ; and there is some negativity towards the use of repositories (Jamali et al., 2014). No parallel research on researchers' attitudes has been conducted in Korea, but the availability of funds to support OA publication as part of research funding for a specific project may mean that funding is less of an issue (although there may be disciplinary differences).
Furthermore, the embeddedness of the scholarly societies in the publication process may allay any potential concerns regarding quality. The other key stakeholder group is university libraries, which typically are more likely to be engaged with OARs than with OAP. In both Korea and western economies, such institutional repositories are used more for theses deposit than for staff research. In Korea, individual university repositories contribute to the dCollection archive.

Intellectual property rights
In Korea, most journals are still published by scholarly societies. Traditionally, they have adopted a laissez-faire attitude towards intellectual property rights and their transfer, typically leaving the copyright with the author. However, more recently, there has been a move towards societies requesting copyright transfer from authors because this is necessary for their licensing arrangements with the commercial database providers. CCLs are not widely adopted, although their use is well-established in medicine.

APCs and other payments
APCs are not new to scholarly publishing in Korea. Scholarly societies have always charged APCs (arguably similar to page charges) for publication in their journals. However, APCs for OA journals are higher than those for TA journals, although there are significant differences by discipline. The issue of APCs and their level is pivotal, not only in determining which researchers can afford to publish but also in determining the revenue of publishers, and in turn their long-term viability. In western economies, there is evidence of significant differences in access to grant funding for APCs, with a divide between STM and HSS (Dallmeier-Tiessen et al., 2011;Solomon & Björk, 2012). For Korea, there is evidence that APCs are lower in HSS than in STM, which is most likely a result of the level of funding available (Joung, 2012). It is difficult to compare levels of APCs between Korea and other countries, partly because in addition to APCs, scholarly societies in Korea also charge other supplementary fees, such as a membership fee. Nevertheless, in both this study and that of Solomon and Björk (2012), there is a wide APC range, and fees are typically considerably lower than those charged by western OA journals, which are often around $2,000.

Disciplinary differences and similarities
There are many more HSS journals than there are STM journals published in Korea (Table 1), and more than twice as many HSS journals approved and listed by KCI than STM journals. For HSS societies publish more journals than universities, whereas the opposite is the case for STM. Furthermore, in 2011, a much smaller percentage of HSS journals were available as gold OA than was the case with STM journals, and there is a much higher level of TA for HSS journals than there is for STM journals (see Table 2). This may be linked to funding availability or to the availability of OARs in science and technology (KISTI) and medicine (KoreaMed). Such differences are consistent with those found in studies in western economies.

CONCLUSION Summary
By examining OA initiatives in Korea, including both OAP and OARs, this article offers a holistic perspective on progress with OA in a country that is committed to the development of a model of scholarly communication, which includes a significant OA element. This study offers some insights into the dynamic between scholarly communication, which may, potentially, inform progress both in western economies and in other less-developed nations.
Specifically, Korea shares many of the core characteristics of OA in other countries, for instance, the key stakeholder groups, types of OA, approaches to managing intellectual property rights, APCs and other funding issues, and disciplinary differences and disparities, but in each of these areas, there are nuances that are specific to Korea.

Recommendations for practice/policy
As discussed above, in Korea, as elsewhere, there are a number of stakeholder groups involved with OAP and OARs. Compared with the 'western' OA models, in Korea, government bodies and scholarly societies have been relatively proactive in promoting and funding OA models. This means that the sustainability of the current model, including both OAP and OARs, is pivotally dependent on the level of continued government funding. Indeed, extension to embrace a greater number of journals will require additional government funding. As such, it is important that government bodies, working with other stakeholders, continue to review their long-term aspirations in relation to OA. Furthermore, since OA constitutes a paradigm change, all parties need to work together to build a sustainable model of OA scholarly publishing that includes both OAP and OARs.
More specifically, policy or management decisions will be required in the following areas: • The DA model is inherently unstable, such that policy decisions need to be made regarding its long-term sustainability. • There is a need for further developments in the formalization of intellectual property rights.
• Although, currently, in Korea, the level of APCs does not present a problem for researchers, relatively low APCs do pose a threat to the sustainability and viability of society publishers. This situation needs to be monitored in order to inform policy decisions.
• The differences in the OAP profiles of disciplines are evidence of the role of university investment in the dissemination of scholarly outputs. This role would benefit from further examination, with particular reference to the role that universities and their libraries can play in OARs.

Limitations and further research
Examining the global landscape of OA is complex. This article makes a modest contribution towards this endeavour. Further studies in other countries in the world, with at least some emphasis on countries where the language of scholarly communication is not English, are required as a basis for achieving improved access to research outputs for scholars and professionals in lessdeveloped countries.