The Role of Institutional Repositories in Higher Education: Purpose and Level of Openness

Abstract

The aim of this chapter is to consider repositories as distributed learning systems, exploring how they can influence higher education by allowing learners and teachers to deposit their educational resources for open sharing and use in teaching and learning processes. To this end, three essential issues will be addressed. The first is the current state of institutional repositories in higher education at international level. An overview will be offered—covering content, metadata, licences, educational and reuse aspects, etc.—to identify the different levels of development of these repositories, followed by a more specific contextualisation of the European case. The second issue, causally related to the first, is the question of what should be done to (further) increase the level of openness and interconnection of repositories in order to integrate them into education. In other words, how can we advance from open content and Open Educational Resources (OER) to Open Educational Practices (OEP), and from OAI-PHM (Open Archives Initiative Protocol for Metadata Harvesting) protocol interoperability to Linked Data and Open Informational Ecosystems? Finally, several ongoing initiatives will also be discussed as good practice models to be considered.

1 Introduction

There are two main types of repositories containing Open Educational Resources (OER) in higher education. The first is the OER-exclusive central repository model (national or thematic), and the second is the hybrid model of institutional repositories (IRs) combining teaching and research outputs. There are also web portals called ‘referatories’ that host links to OER or directories for locating repositories (Neumann & Muuß-Merholz, 2016). Some repositories combine both functions (deposit and referral), such as the Dutch OER Platform Wikiwijs.¹

OER-exclusive repositories are made up solely of OER collections and are usually national or thematic in nature, often maintained by an associated institution or run by public agencies or private companies. These evolved out of what was formerly referred to as learning object repositories (LORs). Institutional repositories are hybrid platforms, created mostly for storing research outputs although they may also include learning materials. However, there are also some examples of institutional repositories created specifically for educational purposes (e.g., the cooperative repository of Catalan universities MDX -Materials Docents en Xarxa-²).

As will be discussed below, both models of repositories have various benefits and drawbacks and, thus, neither is necessarily better than the other. It is the specific context, purpose, availability of resources, and intended outcomes of the OER project concerned that influence the decision of which model to use. Other factors affecting this decision include stakeholder needs, existing institutional policies and practices, technical infrastructure, sustainability, staff skills and understanding, quality workflows, and copyright licensing (Risquez et al., 2020).

In many cases, the existing repository determines the institutional OER strategy and vice versa. This is why many institutions are using their institutional research repositories for the conservation and dissemination of OER. In this way, IRs can support education by encouraging students and teachers to deposit their educational resources and share and use them openly in teaching and learning processes, as well as by providing interconnections with other educational platforms.

With the above in mind, three essential issues will be addressed in this chapter. The first is the current state of institutional repositories in higher education at international level. The second issue, causally related to the first, is the question of what should be done to (further) increase the level of openness and interconnection of repositories in order to integrate them into the educational system. In other words, how can we advance from open content to Open Educational Practices (OEP) and from OAI-PHM protocol interoperability to Linked Data and Open Informational Ecosystems? Finally, some ongoing initiatives will be discussed, as good practice models to be considered.

2 The Current State of Institutional Repositories

Overview of OER Repositories

In recent decades, a few types of OER repositories have been widely adopted to advance the aims of the OER initiative as stated in the 2012 Paris Declaration, and to help implement OER as encouraged in the UNESCO (2019) OER Recommendation.

However, there is no single authoritative source listing all OER repositories, and their total number varies depending on the international repository directory consulted. Therefore, it is not clear how many repositories of this kind exist wordwide, and diversity is clearly one of their main characteristics.

According to the OpenDOAR,³ a directory of open access repositories developed by the University of Nottingham, there are 786 Learning Object repositories⁴ (Fig. 1), 89% of which (699) are IRs, while the rest are classified as Disciplinary, Governmental, or Aggregators. This directory shows an exponential increase in the number of institutional repositories over time, which is related to the development of open access policies around the world.

Fig. 1

Open DOAR content types (From Statistics: https://v2.sherpa.ac.uk/view/repository_visualisations/1.html)

Another important international source, the Registry of Open Access Repositories (ROAR),⁵ lists 77 learning and teaching object⁶ repositories, most of which are IRs. Finally, the OER World Map,⁷ a project funded by the Hewlett Foundation, registers 328 entries of repositories, 125 of which are managed by higher education institutions.

Beyond the numbers, some qualitative studies provide an overview of the state of OER repositories at international level. For example, Tzikopoulos et al. (2007) conducted a pioneering, comprehensive analysis of OER main features, while McGreal (2008) offers a review of the most popular ones. Other global studies include holistic approaches to quality assurance (Atenas & Havemann, 2013; Clements et al., 2015), the promotion of OER through institutional repositories (Castaño et al., 2016), and the state of higher education OER repositories in Latin America (Morales & Montoya, 2014) and worldwide (Santos-Hermosa et al., 2017). Based on this literature, specific characteristics (typology, origin, metadata, quality, licenses, educational and reuse aspects, etc.) can be identified to create a picture of the current state of OER repositories.

Most repositories are institutional (created mainly by universities and government agencies), and the majority are based in the US and Europe (UK, Spain, Germany, and France). The countries with the most repositories appear to be the countries where also the most publications about OER can be found (Zancanaro et al., 2015).

In terms of quality, internal quality assurance (IQA) criteria for repositories have not yet been widely implemented, although some recommendations on the feasibility of incorporating IQA into current and future systems have been offered by experts. The most common top-down criteria seem to be internal assessments (mainly the OER submission process), followed by author prestige and external reviews (peer review systems). Bottom-up quality controls based on community expertise (user ratings and comments) are also used (Clements et al., 2015; Atenas & Havemann, 2013; Santos-Hermosa et al., 2017).

OER repositories are at different levels of development regarding reuse and educational aspects (Santos-Hermosa et al., 2017). Repositories usually include more incentives for reuse, mainly through open licences (most of them using Creative Commons) and social networks, than features that facilitate OER retrieval and use based on educational needs (such as learning objectives). On the other hand, other specific parameters for reuse (such as quality, versioning,⁸ granularity, and open formats) are less evident, and only a few categories of educational metadata are offered (generally content type, subject, and educational level).

This situation suggests that repositories focus more specifically on OA licences (allowing different levels of openness) and on facilitating the creation of communities of users (to comment on and share OER) than on providing specific educational metadata. The most common metadata standard used in IRs is Dublin Core (DC), which is not educational but is conditioned by the type of repository software used most (DSpace). However, some cases of metadata crosswalks have also been observed to incorporate LOM (Learning Object Metadata) educational descriptors into DSpace's default-qualified Dublin Core (DC) schema.

European Case

In Europe, apart from the European Commission’s communication on ‘Opening up education’, several national and institutional policies on Open Education (OE) (or with an OE component) promote the development of OER. These also influence the creation and development of better OER infrastructures and, consequently, better repositories.

Some examples (Santos-Hermosa, 2019) at the national level are the cases of Finland (Declaration for Open Science 2020–2025^{9 10}), France (second National Plan for Open Science 2021–2024¹¹), and Slovenia (Opening Up Slovenia¹²), while at the institutional level, there are Tilburg University’s Action Plan Open Science 2018–2020¹³ and Universitat Oberta de Catalunya’s Open Knowledge policy 2021.¹⁴ In addition, as mentioned above, there are some European countries with a longer tradition of OER and IR repositories, such as Germany, the UK, France, and Spain.

Although Germany could be characterised as a relative latecomer to the OER scene, there is now a strong OER community in the country. OER are developed in Germany using both bottom-up (the OERinfo and EduArc government funding programmes) and top-down (a national working group for OER repositories) approaches (Orr et al. 2017). In addition, numerous OER portals have been launched in recent years, such as Twilo, ZOERR, or the OER World Map (although this last one is not specific for Germany).

The success of OER repositories in the UK might be due to the OER programme funding support from the Joint Information Systems Committee (JISC),¹⁵ which has supported the development of various OER projects. These initiatives have required records to be deposited in the Jorum open repository and many have also used open institutional repositories to host their original materials (Beetham et al., 2012). However, currently, JISC and Jorum are not active anymore and the OER Research Hub,¹⁶ located in the Institute of Educational Technology (IET) at the Open University (OU), is leading the research on OER and their impact on teaching and learning practices.

Spain also has a specific OER initiative implemented by REBIUN (the Spanish network of university libraries) and launched in 2017, which has been analysing IRs and open policies in terms of educational resources (REBIUN, 2019). A recent study on higher education OER collections in IRs (Santos-Hermosa et al., 2020) has established that there are 45 IRs with collections of this kind in Spanish universities (58.4% of the total) and that there has been an increase in the presence of educational resources in IRs.

Finally, a substantial number of international networks has also been established to promote OER and repositories, such as the SPARC (the Scholarly Publishing and Academic Resources Coalition) Europe¹⁷ and a new working group on educational resources within LIBER (Ligue des Bibliothèques Européennes de Recherche – Association of European Research Libraries).¹⁸

OER Content in Institutional Repositories

An increasing number of universities have chosen their institutional repositories (IRs) for primary access and storage of the OER produced by their faculty, students, and staff. Some of the reasons for using IRs are the existing infrastructure they provide, the possibility of long-term preservation and versioning of OER they offer, their visibility and availability to the academic community, and the benefits in terms of internal promotion and awareness (Ferguson, 2017).

In many cases, the existing IR determines the institutional OER strategy and facilitates compliance with Open Access mandates and policies. This is why some institutions choose their existing institutional repositories of research for OER curation and dissemination. Timing and workflow in ingesting materials are additional considerations that must be factored in, as well as their value as low-cost or no-cost alternatives to open textbooks (Mitchel & Chu, 2014). This kind of repository also has a long lifespan for archiving different updates and editions of an OER (necessary because their modular nature allows instructors to customise materials) and preserving all of them (Hess, 2016).

Moreover, IRs allow institutions to disseminate all their academic production to an external audience but also to the internal community, to make members of that community aware of an OER available that has been created by one faculty member but may also benefit others. Indeed, for an IR to be successful and serve its purpose, it is important that members of the academic community are aware of its existence, understand its value, and are interested in contributing content of their own (Yang & Li, 2015).

Reasons given by OER and IR stakeholders (faculty, librarians, and IR administrators) in support of the suitability of institutional repositories for educational resources include ease of access, sharing and collaboration, and profile enhancement. Faculty also consider that IRs provide protection not offered by other online platforms (ensuring that resources have the proper licensing) and curate resources already shared online (as a form of validation by the institution). Technicians and repository managers also point out that IRs do not have storage capacity problems, allow for multi-object OER deposits, and could provide support for research-led teaching, in the sense that OER might be a type of ‘associated material’ to evidence the impact of research on teaching practice (Risquez et al., 2020).

Conversely, arguments against the suitability of IRs for sharing OER mention the availability of other more flexible platforms, lack of visibility, critical mass, the need to keep research and teaching outputs separate, and other concerns such as the lack of a culture of sharing and the need for quality control (Pirkkalainen et al., 2014).

In any case, the success of institutional repositories depends on contributions made by the faculty and students. Similarly, OER are only useful if they are integrated into teaching practices and learning experiences. Therefore, the two workflows need to be brought together: awareness of and commitment to creating and sharing OER (through the repositories); and the use of OER (in classrooms). To achieve this the institution must offer specific training and support (technical, legal, etc.) to engage faculty and students. On the other hand, policies, communication strategies (Tovar Gutiérrez et al., 2014), and interconnected learning systems (Kerres & Heinen, 2015) are crucial for promoting the use of OER in teaching processes through repositories.

3 Integration of Repositories into Education: Openness and Interoperability

According to Ferreras et al. (2013), institutional repositories store, disseminate, and preserve an institution’s digital documents. However, depositing educational resources on a server does not guarantee that they will be effectively retrieved and integrated into classroom learning.

To facilitate the implementation of these educational resources, they need to be open (so they can be used and reused by teachers and learners), described with adequate metadata (so they can be retrieved in accordance with specific educational needs), and exchangeable (by connecting with other technological systems, such as repositories, learning management systems, search engines, and other educational platforms, facilitating interoperability and cross-searchable networks).

For all these reasons, it is important to increase the level of openness, the suitability of metadata, and the interoperability of IRs. Doing so would also promote cross-institutional collaboration and increase the visibility of OER creators.

Finally, once content requirements have been properly ensured, it will be possible to bring practices into the equation, i.e., introduce Open Education Practices (OEP) that support the (re)use and production of OER into teaching and learning.

3.1 Openness: From Open Content to Open Educational Practices (OEP)

Openness in education has attracted considerable attention and discussion in recent years. However, it is a complex concept, subject to different interpretations (openness as free access, openness as open licensed resources, openness as open educational practices, etc.) and with various related elements (OER, MOOCs, OEP, co-creation, networked participation, etc.). This section attempts to identify the best possible way of depositing OER into IRs (content) for implementation and use in teaching (practices).

There are many reports that stress the importance of open licences for OER use. The Paris Declaration (UNESCO, 2012), for instance, promotes open licences ‘in order to facilitate the use, reuse, revision, combination, and redistribution of educational materials worldwide’. Furthermore, various international guidelines (Butcher, 2011; Conole, 2011) and research by the Commonwealth of Learning (McGreal et al., 2013) exist, which demand the use of Creative Commons licences and open standards as a precondition for the educational use of OER. However, other critical approaches to open education broaden this perspective and encourage a view that moves beyond the binaries of open and closed (Cronin, 2020).

The widespread adoption of Creative Commons licensing and the sense of public ownership of resources and content developed with public budgets are helping to foster openness. In the case of IRs, studies show that the openness of deposited OER has been increasing over the years regarding the number of resources with a CC license (Tzikopoulos et al., 2007; Atenas & Havemann, 2013; Amiel & Soares, 2016; Santos-Hermosa et al., 2017; Santos-Hermosa et al., 2020).

Although institutional repositories can provide OER for use in the classroom and pedagogical development, there are still two important considerations regarding open licensing that need to be taken into account for the successful implementation of OER. The first has to do with identifying the most appropriate licences for OER, while the second is related to opening up education through OEP.

One of the main concerns of teachers when sharing resources in repositories is the loss of ownership control (Risquez et al., 2020). Related to this, when looking for OER by other authors, teachers also need to know the legal conditions under which a particular resource can be used in their classroom. Therefore, in order to use and reuse OER, teachers need training and advice so that they will be able to choose suitable licences for their own material and recognise the licences on other colleagues’ resources.

In this sense, librarians are doing an excellent job of supporting faculty and advising them about copyright, open licences, and permissions, for example, through OER toolkits and other specific services (West et al., 2018). However, especially in relation to IRs, administrators have been dealing with research outputs for a long time and although there is a workflow in the research domain (definitions for peer-review, copyright, quality control, etc.), they do not usually have clear guidelines for OER. It is also true that there is no general agreement or criteria on when and how OER are suitable for teaching and learning, e.g., when a resource has reached a quality that is acceptable for sharing or how to recommend the most appropriate open licence for an OER (REBIUN, 2019; Risquez, 2020; Santos-Hermosa et al., 2020).

However, some steps towards international collaboration have been taken, and OER librarians organise themselves in networks, such as the European Network of Open Education Librarians (ENOEL)¹⁹ in SPARC Europe and a recent Group on Educational Resources in LIBER.²⁰

Regarding the question of opening up education, it is important to highlight that the increasing number of OER repositories and the availability of (many) open license educational resources will not necessarily suffice to achieve. The proliferation of OER alone will not change educational practices and will not automatically improve the quality of education (Knox, 2013). Open education is more complex and should include practices that support the reuse and production of OER through institutional policies, promote innovative pedagogical models, and respect and empower learners as co-producers on their lifelong learning paths (Ehlers, 2011). Wiley and Hilton (2018) even propose a new term as a reaction to the development of Open Pedagogy ideas, ‘OER-enabled pedagogy’, defined as the set of teaching and learning practices that are only possible or feasible with the permissions that are characteristic of OER. In this sense, it is necessary to support innovative approaches to teaching and learning and expand participation in education, not only by facilitating access to content but also by creating open communities of practice (Cronin, 2017).

3.2 Interoperability: From OAI-PMH to Linked Open Data and Open Informational Ecosystems

In addition to making openly licensed resources available and facilitating their use in teaching, repositories should include proper educational metadata and be connected to other repositories, educational platforms, and LMS. All this is needed to build federated systems of networked platforms and to establish open informational ecosystems.

Metadata

Repositories require an efficient search mechanism based on metadata in addition to intuitive navigation, quality controls, etc. According to Atenas and Havemann (2013), metadata is an indicator of quality assurance and a prerequisite for a repository’s success.

Thus, an excellent metadata-based strategy has positive effects on using a repository for both educators and students (in terms of achievement and dropout rates). For example, the description of OER using specific educational metadata generally receives a positive response from users, since they recognise that just using the title, keywords, and type of resources is not enough to find suitable OER (Wojcik & Rataj, 2020). On the other hand, the inability to use repositories effectively to find content that meets the needs of potential users (educators, students, etc.) can inhibit the broader adoption of these platforms and prevent them from having greater impact on open education (Dichev & Dicheva, 2012).

In this sense, the use of metadata standards (such as IEEE LOM²¹) facilitates the description of didactic characteristics of the OER (educational level, learning objectives, competencies, etc.) that are essential for retrieving and using them in accordance with the teaching and learning purposes of the users (Santos-Hermosa et al., 2017). It is also recommended that educational metadata be enriched with ontology-based semantic indexing for better results (Ruiz-Iniesta et al., 2014).

Artificial Intelligence

In the last few years, artificial intelligence (AI) and machine learning (ML) have become ubiquitous terms. They are applied to solve existing problems as an increasingly popular practice and should be considered for problems emerging in the future (Gonzalez, Zimmermann & Nagappan, 2020).

Machine learning algorithms require access to large amounts of data and metadata from multiple, cross-linked files, which should be easy to consult and retrieve the correct information from (Prior et al., 2020). Such approaches can complement repositories’ discoverability. For example, the role of AI in automatic metadata generation and machine learning extraction methods is expanding the possibilities for OER search (Tsay et al., 2020).

OAI-PMH Interoperability

Interoperability plays a crucial role in the sharing and global discovery of content in the education environment. Driven by the open access movement, the interoperability of metadata has been the most active area in developing digital repositories (Aschenbrenner et al., 2008).

The OAI-PMH (Open Archives Initiative Protocol for Metadata Harvesting), developed by the Open Archives Initiative, is based on open standards that guarantee automatic interoperability between senders and receivers of digital resources. Most institutional repository software providers (such as DSpace, EPrints, Digital Commons, etc.) support Dublin Core metadata export and use OAI-PMH to allow interoperability with other repositories to harvest bibliographic data (OAI, 2015). The repositories are constantly fed with new description metadata and send their own to one another. This perpetual activity of search queries allows continuous updating of these platforms.

Although OAI-PMH is a well-established and widely-used standard that facilitates the exchange of metadata, it still has some problems when integrating data that is extracted from various repositories, and its use for OER is somewhat limited.

OAI-PMH has difficulties to standardise standardising processes for the description and publication of metadata in the open environment of the web, resulting in a heterogeneous context in which each institution individually manages various metadata schemas, data formats, or vocabularies. Therefore, this metadata standard achieves only the lowest level of interoperability in repositories by facilitating the extraction and exchange of metadata. However, it does not ensure the reuse and integration of information since it does not have the possibility of unification and interoperability with other datasets (Piedra et al., 2014). Therefore, OER data should be encoded in a machine-readable format to be easily accessible to any user (human being or machine agent) and, thus, improve the exchange, reuse, sharing, and enrichment of data.

In recent years, it has been generally noted (OpenAIRE, 2021) that some customisations and significant advances have been made in IRs concerning enriching metadata (through mechanisms to adapt the DC to specific information needs) and extracting and recovering useful information for users, as well as the processing of meaning (semantics of information). However, in the European context, most institutional repositories are currently focusing on improving scholarly communication and research resources.

This is evidenced by the Confederation of Open Access Repositories (COAR) initiative named Next Generation Repositories (NGR), which identifies protocols and technologies that will enable new and improved functionalities for 21st-century repository systems. The initiative aims to support interconnected, resource-centric repositories that manage access to diverse outputs, making resources the focus of services (Rodrigues et al., 2017).

Although the most recent NGR reports (COAR, 2019; Rodrigues et al., 2017; Walk et al., 2020) advocate moving towards new levels of web-centric interoperability and web technologies and also recommend keeping repositories open to new kinds of content (such as OER), they are still focused mainly on the development of technologies that will enhance research outputs.

Recent advancements include review services for repositories, which enable the linking of preprints and other repository resources to external services (Walk et al., 2020). Another advancement is the technology Signposting,²² an approach to informing machine agents about the nature of the resources linked to the resource they currently interact with. Finally, ResourceSync²³ is a mechanism for large-scale synchronising of web resources, allowing aggregation services to ‘harvest’ metadata and content from repositories (OpenAIRE, 2021).

All these improvements focus on research outputs and the connection of several different platforms in scholarly communications (by creating gateways between IRs, data repositories, preprint servers, publishing platforms, CRIS systems, etc.). In addition, some of these refinements have been integrated into widely used institutional repository platforms (Ferreras, 2021).

Nevertheless, it seems clear that they are not yet oriented towards OER or educational ecosystems and that a more specific approach needs to be developed. There are two main challenges in this respect: integrating OER into IRs using semantic web technologies and Linked Open data and connecting different open information systems (repositories, LMS, and other educational platforms) in an educational ecosystem.

Integration: Semantic Web and Linked Data

The integration focus facilitates the combination of existing information resources in various contexts. Semantic web technologies and, specifically, linked data (referring to connecting structured data on the web) change the way information is stored, published, connected, and exploited (Berners-Lee, 2006).

When applied to IRs, this creates the potential for global networks of cross-searchable information. Networked repositories lower access barriers and offer the widest possible dissemination of content (Ukwoma et al., 2019). This is why there is a need to improve the metadata interoperability between different collections of open material through the application of a semantic web approach and linked data technologies.

An example of a repository pursuing this kind of development is MIIDAS, in the areas of computing and electronics, which includes the metadata of the OAI-PMH protocol and adds educational metadata. This platform uses the Resource Description Framework (RDF), ontologies, and SPARQL query language to improve resource retrieval (Durán & Ramírez, 2021). This kind of development in IRs is disseminating OER and allowing the construction of semantically enriched data sets. Another case of using linked data technologies for open access repositories to optimise the level of interoperability is the network of repositories belonging to the Ecuadorian Consortium for the Development of the Advanced Internet (CEDIA)²⁴ (Piedra et al., 2015).

Open Informational Ecosystems

Technical discussions have evolved from the fundamental issues of creating repositories connected through interoperability protocols to the integration of repositories into a broader academic infrastructure (COAR, 2015; Rodrigues et al., 2017).

To avoid a situation where repositories behave like local silos, the repository infrastructure needs to evolve rapidly and interact with other local infrastructures and external systems that facilitate education (such as collaborative teaching environments, e-learning systems, editorials, etc.). There is a need to extend interoperability activities beyond repository-to-repository efforts to include interoperability across the diversity of systems that exist in the educational context.

When IRs go beyond merely providing content and offering systematic exchange with different external platforms, they become ‘open informational ecosystems’. According to Kerres and Heinen (2015), these systems are defined as a technological infrastructure of related services, in which several independent providers are linked and their collaboration relies on standards for interfacing content and metadata.

Open information systems consist of a reference platform that provides access to all associated content platforms. An example is the publicly financed German educational server Deutscher Bildungsserver,²⁵ which acts as a reference platform that links to private and public content providers with (open and closed) learning materials for primary and secondary education. In addition, this server is linked to the reference-run platforms of the 16 German states and supports connections between state education servers to develop technical, informational, and editorial infrastructures.

Other examples of OER integration between heterogeneous learning platforms—specifically regarding the connection of repositories with learning management systems (LMS) and MOOCs—are discussed below.

Some preliminary steps towards building bridges between learning systems were taken by the Massachusetts Institute of Technology (MIT) when it created an open source e-learning platform called .LRN (dotLRN), which integrated different services in the virtual learning environment (including OER repositories). The MIT also defined a global e-learning strategy, named Intellectual Commons, that includes .LRN and its other initiatives (Sotelo et al., 2019).

The DSpace software for IRs has also been integrated into the Moodle LMS using various plugins (Gómez et al., 2014), a search engine (Guzman-Arias et al., 2019), and online laboratories (Ruano et al., 2020). One plugin performs the integration using the permanent URL (or handle) provided for the DSpace software to every item (file), which allows the incorporation of OER from the repository into the virtual classroom in Moodle. The Open Education Austria Advanced project²⁶ has developed another OER plugin that makes it possible to index OER in the LMS (Moodle) with the corresponding OER metadata, to transfer them into the university’s repository, and to make them available via other services, such as the Austrian OER portal (Ruano et al., 2019). A different strategy is to enable access to OER using the institution’s own LMS and offering it to a wider public via the university’s own OER repository, as Graz University of Technology (TU Graz) has done (Ladurner et al., 2020).

Other institutions establish interactions between formal and non-formal systems (e-learning and MOOC platforms) through frameworks based on linked data technologies and integration layers that allow MOOC designers to find and access OER in repositories (Mrhar et al., 2020).

Through such initiatives to integrate different platforms into open information ecosystems, the gap between educational resources and educational systems can be filled, resulting in an immense contribution to open education.

4 Current OER Repository Initiatives

This last section considers some interesting ongoing initiatives involving OER repositories. Some of these initiatives have been chosen because they create global, international, or national informational ecosystems (such as 5Xgon, Open Discovery Space, or ENCORE+), while others provide a connected national infrastructure (OERSi and Open Education Austria). However, all of them have in common that they seek ways to influence the future of open educational resources by applying the latest technologies to the educational ecosystem.

OERSI

The Open Educational Resources Search Index (OERSI)²⁷ is a central entry point for OER searches in higher education in Germany across distributed heterogeneous sources. It is a new repository created by the Technische Informationsbibliothek (TIB) in Hannover under the responsibility of the Ministry for Science and Culture of Lower Saxony (Van Wijngaarden et al., 2021). It has been developed with reusable open-source software, whose code is available on GitLab.²⁸ The search sources are various OER portals and repositories and the OER that can be accessed are mainly videos, although there are also courses, images, etc. The target groups are end users (teachers, learners) and service providers (repository operators, LMS operators).

Its technical features include standardised JSON-LD metadata²⁹ based on the Learning Resource Metadata Initiative (LRMI), schema.org, and SKOS; a Modular ETL (Extract, Transform, Load)³⁰ for import processes, and an option for embedding OER in LMS or other contexts. Thus, the OERSI can be considered an open infrastructure good practice for connecting distributed scenarios.

Open Education Austria

Open Education Austria Advanced (OEAA)³¹ is a project by Austrian universities (running from 2020 to February 28, 2024) to jointly develop a national infrastructure for OER. It is a specialist portal with a search function across the decentralised OER repositories provided by the participating universities, e-learning centres, and central computer science services. Its modular system architectures and open-source software developments facilitate universities’ (technical) participation beyond the project. The project is also working on establishing an open source repository and implementing interfaces with university-owned services. Furthermore, the definition of an application profile based on LOM facilitates metadata transformations, thus enabling OER outside local repositories and the specialist portal.

OEAA is also conceived as a good practice in cross-institutional open-source development that aims to contribute to the unrestricted use of OER by gradually establishing open practices analogous to research (Open Access, Open Data). It also provides training and a national OER certification body, operated by the Forum for New Media in Teaching Austria (FNMA).

5XGon

The X5Gon³² is a cross-site global OER network that aims to connect European OER repositories. Its database contains OER from registered OER repositories in its network of 22 partners (including MIT OCW, University of Bologna, and others). The materials currently shown are three types of OER (text, video, and audio) that have been enriched through wikification.³³

The indexed OER platforms are connected in this network via the Connect service and APIs. This is made possible by adding a JavaScript code snippet to each resource to be indexed (and enriched with transcription and translation when needed), which will itself call a library on the own server.³⁴ Concerning OER aggregation services, it has been decided to develop an option that also connects with the OAI-PMH protocol.

Other RTD projects related to X5Gon technologies include lecture recording solutions (the VirtUOS platform³⁵ run by Osnabruck University), transLectures (TraMOOC and Videolecture repository³⁶), and RTD projects³⁷ related to MOOCs (Polimedia at Universitat Politècnica de València³⁸).

X5Gon is an interesting case of merging services and developing technologies that constitutes the stepping stone towards a wide range of derivative applications in educational technologies. It is also working on adapting new solutions for near real-time multilingual delivery of OER content and MOOCs, since coverage of under-resourced languages is needed.

ENCORE Projects

The European Network for Catalysing Open Resources in Education (ENCORE+)³⁹ is part of the 2020 ERASMUS+ programme and responds to the European priority of opening up education. One of its five main objectives is to foster the collaboration and connection of repositories in a European OER repository ecosystem. A working package (WP3) named ‘Technology for the Future European OER Repository Ecosystem’ is currently working specifically on this aim. The first deliverable will be a proof of concept of key features of the OER infrastructure, identifying and exploring key technological features of OER repositories to support quality assured production, sustainability, and innovation.

The International Council for Open and Distance Education (ICDE) is the coordinator of the project, which also includes several institutional partners, such as the Open University in the United Kingdom, Universidad Internacional de la Rioja in Spain, and Dublin City University in Ireland. The private sector is also represented among the project partners by Joubel AS in Norway, and Knowledge 4 all Foundation and Instructure Global Ltd. (Canvas) in the United Kingdom.

ENCORE+ seems to be very timely in view of the newly adopted UNESCO (2019) OER Recommendation, and it offers a new vision for collaboration and connection between OER repositories in a European OER Ecosystem, encouraging entrepreneurship and empowerment through OER.

5 Conclusion

The Covid-19 pandemic has reinforced the general need for OER in all educational sectors. Undoubtedly, repositories play a vital role in this regard since they improve and facilitate access to OER. The overview presented here shows the increasing number of OER collections, their presence in institutional repositories (IRs), and their different levels of development. We can also conclude that for greater adoption of OER, more effectiveness is needed in the search and location in the IRs and more interoperability with the other educational platforms of an institution.

As we have seen, a variety of actions is needed to turn repositories into distributed learning systems. On the one hand, repositories should take advantage of emerging technologies in terms of cross-searchable information, with semantic web and linked open data increasing discovery and findability, and artificial intelligence and machine learning supporting this. Therefore, a good combination of both human and computer access methods should be considered in future. On the other hand, increasing the openness of OER is also a good way since the open licence allocation allows the actual reuse and remix of OER.

Last but not least, strong institutional coverage should be provided, with strategies top-down and bottom-up, to establish a culture of sharing and OER adoption in teaching and learning. Examples are amplifying and promoting the adaption of OER through institutional policies, strategies, and incentives and providing technical and infrastructural standards for the construction and design of OER repositories. Other complementary strategies are increasing awareness and training for practitioners and teachers in open skills and open education methods since many of them are not yet familiar with the term OER. Accompanying and tutoring faculty and students on their path to OER, with some practice in sharing them in repositories and using them in established pedagogical approaches, would be possible strategies to implement.

OER is on the road to mainstream acceptance, and a mix of strategies to stimulate their adoption through repositories should be applied, embracing all relevant stakeholders within institutions.