Verification of Education Credentials on European Blockchain Services Infrastructure (EBSI): Action Research in a Cross-Border Use Case between Belgium and Italy

Abstract

The European Blockchain Services Infrastructure (EBSI) is a major policy initiative of the EU and European Blockchain Partnership, to leverage blockchain to create cross-border services for public administrations, businesses, citizens, and their ecosystems, to verify information and increase trust in services. The EBSI aims to provide a secure and interoperable system infrastructure, utilizing new digital technologies such as digital wallets, verifiable credentials, and decentralized identifiers. These technologies are expected to change the way government services are organized between citizens and service providers. This article presents empirical findings from a cross-border pilot that was the first case to utilize this new blockchain infrastructure, and which tested the institutional, technical, and user-specific requirements for wider adoption. The pilot tested and assessed the verification of education credentials through the EBSI blockchain in a cross-border setting between a Belgian and an Italian university. The research was based on action research with the ICT units of the participating universities, wallet solution providers, the Belgian government, and EBSI officials. The findings highlight the following as key challenges for the wider adoption of the EBSI and verification credential use case: (1) onboarding of the EBSI ecosystem governance, (2) issuance of EBSI-compliant digital wallets and data schemes for transcript validation, (3) interoperability issues concerning digital identity systems.

1. Introduction

Blockchain and blockchain-supported concepts, such as self-sovereign identity systems, verifiable credentials, and decentralized identifiers, are gradually transforming information management systems in various public sector domains [1,2,3]. Education is a prominent public sector domain where blockchain solutions are actively being developed and tested [4,5,6]. The literature provides a plethora of exploratory studies and systematic literature reviews that detail the prospects and challenges of blockchain applications in the public sector in general [1,3,7,8,9] and in the education domain in particular [4,5,6]. However, there have been very limited empirical findings related to the actual implementation challenges of blockchain-based solutions. Furthermore, we lack practical insights into the use of verifiable credentials, self-sovereign identity systems, and digital wallets, to facilitate cross-border academic mobility.

In this article, we provide empirical findings from one of the biggest blockchain initiatives in the world targeting the public sector: the European Blockchain Services Infrastructure (EBSI). The EBSI is a joint initiative of the European Union (EU) and the European Blockchain Partnership (EBP), an intergovernmental initiative of the 27 EU member states, Norway, and Liechtenstein. The EBSI aims to create a Europe-wide infrastructure, to support cross-border services for public administrations, businesses, citizens, and their ecosystems and to verify the information and make services trustworthy. Since 2020, the EBSI has been deploying a network of distributed nodes across Europe and developing and testing use cases in education, social security, and other public service areas that utilize self-sovereign identity systems, verifiable credentials, and decentralized identifiers.

Through a pilot on the verification of education credentials between two universities in Italy and Belgium, this article investigates the implementation challenges of EBSI blockchain and self-sovereign identity systems for the verification of student identification and the transcripts of student records. Furthermore, this article provides the first use case in the literature to assess the implementation challenges of the European Blockchain Services Infrastructure (EBSI) and the European self-sovereign identity framework (ESSIF). Through action research, the article provides empirical evidence regarding the obstacles and potential solutions to the transformation of digital credential frameworks, toward decentralized systems compatible with self-sovereign identity systems.

The findings suggest that the efficiency of education credential systems can be significantly improved with EBSI blockchain and SSI solutions, but onboarding issues, missing data schemes for transcript records, political hesitations, the alignment with Single Digital Gateway and Europass, and interoperability issues concerning digital wallet solutions and digital identity systems appear to be the key challenges in moving to the production phase.

The structure of this article is as follows: Section 2 includes an overview of the role of blockchain technology in the public sector and provides background information about the European Blockchain Services Infrastructure (EBSI) and the EBSI diploma use case. Section 3 describes the methodology and introduces the use case. Section 4 presents the results from the pilot. Section 5 discusses the use case requirements for wider implementation of the pilot. The conclusion section summarizes the main results and shares policy recommendations.

2. Conceptual Framework

2.1. Blockchain and Its Application in the Public Sector

Blockchain technology (BCT) is not a monolithic technology and covers several underlying and related technologies, such as smart contracts, token technology, decentralized applications, etc., which can jointly support a decentralized and automated information infrastructure. BCT has various applications in the public sector. At its most basic level, it can function as an information infrastructure to improve the management and exchange of public information between administrations. At an advanced level, BCT can allow decentralized information management and automated algorithmic decision-making through smart contracts, to create autonomous organizations that could potentially replace the traditional forms of public sector organizations [10].

There are currently four main areas where BCT is being used in the public sector [11]. The first is identity verification, where BCT-based solutions are used to support self-sovereign digital identity systems for individuals and businesses. Examples include the use of self-sovereign electronic identification, authentication, and trust services (eIDAS) by EBSI [12]; a blockchain-based identity system for refugees in Finnish camps [13]; and BCT-supported identity verification systems for more secure e-voting, such as in Zug [14] and in Barcelona [15].

The second area where BCT is used is in asset registries. BCT is currently used to improve the reliability, traceability, and security of registered data in public and private databases, reducing operational costs and verification times [16]. Examples include notary and diploma use cases for EBSI, the Exonum land title registry in Georgia [14], and the Equity Platform to register and validate individual energy transactions [17].

The third area where BCT is used is in automating and tracking high-risk transactions and improving the traceability and transparency of supply-chain management systems. There are various use cases in logistics, food tracking, and energy sectors for blockchain-based supply chain management systems [18].

Finally, the fourth emerging area of application for BCT is central bank digital currencies (CBDC), which are cryptocurrencies issued by central banks. Currently, Jamaica and the Bahamas are the only countries that have launched CBDCs, while several others have announced their intention to research and adopt them (For an overview of country cases, please see: https://cbdctracker.org/ (accessed on 31 January 2023)).

2.2. Blockchain Applications in the Verification of Education Credentials

There are several examples around the world of where blockchain-supported systems are used to verify diplomas and education credentials. One of the first active use cases was introduced by the University of Nicosia in Cyprus, which issues academic certificates with authenticity verification through the Bitcoin blockchain [19]. Hoburton School in San Francisco uses an intra-institutional blockchain for certificate authentication [20]. The University of Maryville in the US issues digital diplomas on a blockchain-compatible digital wallet (https://www.maryville.edu/digitaldiploma (accessed on 31 January 2023)). Similar solutions based on Ethereum and smart contracts are at different stages of being developed and deployed by the University of Zurich, University of Lisbon, Ho Chi Minh City University of Technology, the University of Southern New Hampshire, and Open Source University [5,21]. At the country level, several educational institutions in Singapore use blockchain and verifiable credentials for official documents and certificates (https://www.developer.tech.gov.sg/products/categories/blockchain/opencerts/overview.html (accessed on 31 January 2023)). Finally, in May 2021, the University of Lille announced plans to become the first university to achieve a true digital transformation of its academic department, by issuing 20,000 certificates of completion as digital certificates anchored in a low-cost blockchain. The University of Lille aims to scale up this project as part of the EBSI blockchain (The press release is accessible at: chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://www.univ-lille.fr/fileadmin/user_upload/presse/presse_2021/CPDemattestoklien.pdf (accessed on 31 January 2023)).

Gatteschi and colleagues [22] reported that blockchain applications can have various positive impacts on the education sector, by lowering barriers to the global mobility of students and academics through more effective and efficient accreditation of credentials. The self-sovereign identity systems, digital wallets, and digital credentials supported by the immutability and transparency of blockchain systems can curtail fraud and engender trust among parties during job applications and permit academic mobility. However, a systematic literature review by Mohammed and Vargas [4] suggested that the adoption of blockchain solutions in the education sector is affected by numerous technological, organizational, and environmental challenges. Their study suggested that technological challenges have received the most attention in the literature. For instance, poor usability, lack of scalability, limited interoperability and standardization, complexity of integration, security issues, privacy, immutability, lack of flexibility, and data unavailability are some of the reported technical challenges that prevent widespread adoption of this technology in the education domain.

2.3. European Blockchain Services Infrastructure (EBSI)

EBSI is an initiative of the European Blockchain Partnership (EBP), which is a cooperation between the European Commission, all EU Member States, and some countries of the European Economic Area. The goal of EBSI is to deliver EU-wide, cross-border public services with blockchain technology. Since 2020, a peer-to-peer network of distributed nodes has been set up across EU member states, Norway, and Lichtenstein. The EBSI is the first EU-wide blockchain infrastructure driven by the public sector. Each member of the network, as mandated by an EBP Member, hosts EBSI nodes at a national level. All nodes can create and broadcast transactions that update the ledger, and each node keeps an identical copy of this ledger.

As of 2022, there are seven EBSI use cases (i.e., the European Self-Sovereign Identity Framework (ESSIF), diploma, notarization, asylum process management, European social security pass, small-medium enterprise (SME) financing, and trusted data sharing) in different stages of development. Among these, the ESSIF and diploma use cases are at the most advanced level of deployment. In 2021, the EC and EBP created the “Early Adopters Programme” (EAP) to create an end-to-end experience and the first production implementation of the use cases. Considering the advancements made in the use cases, EBSI has prioritized the implementation of the diploma use case along with ESSIF as the first use case for deployment. From September to December 2021, alongside a series of milestones, guidelines, and technical solutions, the selected pilot projects were co-created with EBSI representatives.

In the EBSI framework, self-sovereign identity (SSI) refers to a decentralized identity model, where users can choose their identifier, identity and wallet provider, the information they want to disclose to verifiers, and the verification method. End users handle their digital identity and interact with the EBSI network through digital wallets. Digital wallets are blockchain-compatible applications that allow users to store identity information and engage in electronic transactions with third parties. Verifiable credentials (VCs) define the structure of the data, while verifiable presentations (VPs) provide the data model for presenting and exchanging the data. Decentralized identifiers (DIDs) represent the digital identities and unique identifiers of EU citizens and legal entities in the EBSI ecosystem (https://ec.europa.eu/digital-building-blocks/wikis/display/EBSIDOC/ (accessed on 31 January 2023))

Technically, the EBSI operates as a proof-of-authority consensus model supported by a pan-European network of nodes from the Member States. As of 2023, there are about 40 active nodes. Node operators are selected based on their operational capacity and their adherence to the specific security and availability standards of the EBSI governance rules. The current protocol of the EBSI is based on Hyperledger Besu. The EBSI provides three core technical services: APIs to allow outside applications to connect, smart contracts to link the APIs and the ledger, and an immutable ledger to record the transactions. (https://ec.europa.eu/digital-building-blocks/wikis/display/EBSI/Node+Operators (accessed on 31 January 2023))

The EBSI has developed its governance structure, lexicon, and concepts for verification services. The EBSI relies on W3C standards (https://ec.europa.eu/digital-building-blocks/wikis/display/EBSIDOC/Data+Models+and+Schemas (accessed on 31 January 2023)). The main components and entities that must be taken into account from an education verification perspective are described below:

2.3.1. Decentralized Identifiers (DID)

Decentralized identifiers (DIDs) are key elements that enable natural persons, legal entities, or things to interact with services provided by other entities. A DID is an identifier in itself and does not reveal anything about the owner. An entity may have multiple DIDs, and the owner can select which specific DID to use when interacting with other entities. Once an entity has a DID, the different verifiable attestations provided by third parties can be linked to it, with some attestations describing the DID owner’s identity and others simply providing data to the owner. DIDs are compatible with the self-sovereign identity (SSI) paradigm, allowing identities and data to be controlled, stored, and managed by the DID owner. A DID document typically expresses verification methods (such as public keys) and services relevant to interactions with the DID subject. A DID document can be generated on the fly when required, based on information in the ledger.

2.3.2. ESSIF Onboarding Service (EOS)

The ESSIF Onboarding Service (EOS) helps a legal entity onboard on the ESSIF ledger (DID Registry) and to become listed in an ESSIF Onboarding Service Registry (EOSR). The EOS must ensure the cryptographic control and ownership of the DID and protect the ledger from security threats, such as Sybil attacks and DLT resource exhaustion attacks, by registering very large sets of DIDs.

2.3.3. Trusted Issuer (TI)

A trusted issuer (TI) is a legal entity that is authorized to issue certain types of verifiable credentials (VCs). VCs are electronic documents that provide data about an entity and can be compared to certifying acts produced by authoritative sources, with a certain legal effect, depending on the context and use case. The relevant trusted issuer registry (TIR) contains information about TIs and their accreditations. Trusted accreditation organizations (TAOs) may authorize TIs to write their information and accreditations to the relevant TIR. Two types of VCs may be issued: (1) Verifiable IDs (V-IDs) for identification/authentication in a narrow sense (comparable to national eIDs or passports), and (2) verifiable attestations (VAs) as evidence of a permit (e.g., driving license, work permit) or as evidence of attributes/properties mainly not used for identification/authentication (e.g., diplomas, bus tickets, membership, bank account, postal address, email address, etc.). VAs are intended as a legal substitute for the certifying paper-based documents typically used to prove identity attributes.

2.3.4. Trusted Accredited Organizations (TAO)

A trusted accredited organization (TAO) is a legal entity that enables other legal entities to become TIs. TIs may require accreditations by TAOs for the legitimate issuance of certain types of VCs. TAOs are typically ministerial departments and public agencies, and their roles vary depending on the use case. Once a TI is accredited by a TAO, the information should be registered in a trusted accreditation organization registry (TAOR) that acts as the source of trust. The TAOR admin should be defined by use cases according to relevant regulations or other domain-specific requirements.

2.4. EBSI Diploma Use Case—Cross-Border Scenario

The EBSI diploma use case aims to develop a blockchain-based system that allows students, education institutions, and third parties to verify education credentials, without relying on centralized administrative procedures to authenticate documents. This reduces the administrative burden and minimizes potential fraud. As part of the EAP, a cross-border scenario has been developed to demonstrate that EBSI can implement cross-border verification of educational credentials based on GDPR-compliant ESSIF principles. The ESSIF principles require citizens to own and control their identities and data, initiate any actions themselves, accept any further provided credentials, and ensure that no personal data are stored on-chain.

The scenario seeks to test at least one university in an EU Member State (MS) issuing an education credential, which another university or third party from another MS can verify. The use case involves five roles: issuer (a university that identifies and issues educational credentials for the student), holder (a student who requests the issuance of educational credentials and shares them with a university or employer), verifier (a university or employer in another MS that verifies the educational credentials shared by the student), revocation authority (the issuer of a credential or any entity designed to revoke credentials), and registry administrator (any entity designed to administer the registry of entities, which will implement the MS governance rules). Figure 1 shows a layered view of the EBSI diploma use case.

3. Methods

3.1. Action Research

This research utilized an action research (AR) approach, which applies scientific methods to help organizations identify problems, discover their underlying causes, and generate new knowledge about organizations and change, which can be applied elsewhere [23]. AR is frequently utilized to assess and validate the adoption of technologies in information systems [24,25]. AR is highly collaborative, involving both practitioners and organizational members in the research and action process. Applications to organizational change generally involve the following cyclical events: (1) preliminary data gathering and diagnosis; (2) action planning; (3) implementation; and (4) assessment.

The research design for AR was formulated to support the onboarding and testing of a cross-border inter-university use case for the verification of education credentials with the EBSI network. The aim was to learn about the technical and policy requirements for moving EBSI diploma use cases into the production phase through action research. The participants of the AR were the researchers, the ICT units of the participating universities, wallet solution providers, the Belgian government, and EBSI officials. Data were collected through participant observations, individual interviews, focus group meetings, field notes, and document reviews.

The research was conducted over three months, from August to November 2021, and followed the milestones and roadmap set by the EBSI EAP. The roadmap was designed in stages, where each phase concluded with a reflection point with EBSI officials. Figure 2 summarizes the steps, objectives, and action phases described by the EAP.

3.2. Case Description

The focus of this research was on verifying student ID information and transcripts of records between two universities, KU Leuven and Università di Bologna, from the Una Europa network. Una Europa is an alliance of eleven European research universities from different countries. This case serves as a unique pilot as part of the EAP, because it provides a complete journey toward digitizing the exchange between two universities, placing the student at the center of the exchange. The pilot includes the transfer of both digital identity and non-diploma educational credentials (i.e., transcripts of records), with a focus on exchange students between KU Leuven and Università di Bologna (Unibo). As part of the pilot, two types of credentials were issued: (1) a verifiable student ID with some enrollment information, in addition to a standard verifiable ID issued by governments, and (2) a verifiable transcript of records. Figure 3 illustrates the two scenarios of the pilot.

4. Results

In this section, we present the results obtained from the action research. The results are presented of the diagnosis, action planning, implementation, and assessment stages.

4.1. Diagnosis

The action research began with data gathering and diagnosis of the needs and objectives for development of a proof of concept (PoC), to verify student transcripts/certifications in a cross-border setting with digital credentials. In the diagnosis stage, the project team focused on identifying the needs and objectives concerning three areas: (1) diploma verification at KU Leuven, (2) adoption of a digital credential framework, and (3) implementation of the pilot. Below, we elaborate on the findings for each of these areas.

4.1.1. Diploma Verification at KU Leuven

Before a student can register at KU Leuven, the admissions office must verify their prior degree(s), based on which they are admitted to study at KU Leuven. At this stage, Flemish diplomas can be requested by the admissions office via web services from the national authentic source (e.g., DHO—database higher education/LED—learning and experience database). If the prior educational institute(s) offers an online verification method, the admissions office can verify the degree(s) online, even before the student arrives in Belgium. In any other case, a PDF version of the certification can be uploaded in the application form, which can be used in the credential evaluation process. However, in that case, the student needs to bring the original diploma(s)/graduation certification(s) and the final transcripts to Belgium. At the beginning of the registration period for the upcoming academic year, the student receives instructions on how to make an appointment to check these documents.

If there is a need for verification of a KU Leuven diploma or certificate in general, KU Leuven students can request digital certification through a self-service application. The certificates are sent by mail to the student. The authenticity of the certificates can be verified using a QR code, which leads to the verification website.

Verifying diplomas, especially when there are no online verification methods available, takes a lot of time and money. Experts are needed per region to verify the authenticity of foreign diplomas. Moreover, even if the prior educational institute(s) offers online verification methods, there is no standardization in the verification. Certificates are often provided as a PDF document instead of a data schema in which the data can be automatically verified, interpreted, and processed.

4.1.2. Adoption of a Digital Credential Framework

Based on an impact map (see Figure 4), the Una Europa blockchain workgroup identified, together with the business stakeholders, the challenges and high-level requirements for a digital credential framework.

Through this exercise, the project team identified the following design principles/goals for a digital credential framework:

• Issuing digital credentials
• Verifying digital credentials, ensuring:

  o

  Authenticity of the issuer (who)

  o

  Authenticity of the owner/student (to whom)

  o

  Immutability of the content (not tampered)

• Providing legal value in a cross-border setting (eIDAS compliant)
• Allowing revocation of digital credentials
• No interaction required between the issuer and verifier of digital credentials
• Minimal manual verification needed by the verifier of digital credentials
• GDPR compliance for exchange of digital credentials, prioritizing the holder’s privacy
• Ensuring privacy through minimal personal data disclosure

The project group assessed that the BCT, and specifically the EBSI infrastructure, can help achieve these goals. Three key reasons were identified:

(1)

Regulatory complementarity: a European blockchain infrastructure ensures compliance with European digital certificate regulations.

(2)

Interoperability: the EBSI infrastructure uses W3C standards of verifiable credentials and the Europass learning data model.

(3)

Minimal environmental footprint: the EBSI is a permissioned blockchain network that uses proof of authority (PoA) as the consensus mechanism, ensuring a minimal environmental impact and sustainability, which are top priorities within the Una Europa universities.

4.1.3. EBSI Verifiable Credential Lifecycle

After the initial diagnosis, the project team concluded that adopting a digital credential framework and EBSI verifiable credentials could help Una Europa universities respond better to the challenges of a digital world, by ensuring high levels of security and authenticity. The team assessed that digital certificates could accelerate information processing, reduce administrative burdens for admission offices, and improve student–administration relationships.

The project team, along with officials, identified three stages in the EBSI verifiable credential lifecycle:

• Onboarding of actors
• Issuance and storage of VCs
• Presentation and verification of VCs

Figure 5 illustrates these stages and their specific objectives. Each stage poses particular challenges, which are further elaborated on in the implementation section (see Section 4.3).

4.2. Action Planning

In the action planning stage, the project team assigned functional roles for the Una Europa pilot. The planning process followed the EBSI diploma use case template in assigning specific roles.

Table 1. Description of roles.

Trusted Registries	What Is Legal Entity (LE) Responsible for?	Responsible LE in Diploma Use Case	What Are the Contents in the Diploma Use Case
DID Registry	ESSIF Onboarding Service (EOS)	e.g., Educational Institutions and/or accreditation bodies	DID and Pub Keys of legal entities (e.g., accreditation agencies, educational institutions)
ESSIF Onboarding Service Registry (EOSR)	EBSI (EOSR Admin)	e.g., Educational Institutions and/or accreditation bodies	ESSIF Onboarding Service (EOS) details and authorizations
Trusted Accreditation Organisation Registry (TAOR)	Domain/Jurisdiction specific authorities. (TAOR Admin)	e.g., MS Education and other MS entitled Entities	TAO details (incl. DID) and authorizations to accredit specific educational credentials
Trusted Schema Registry (TSR)	Domain/Jurisdiction specific authorities (e.g., TAOs)	e.g., DG EMPL for European Learning Data Model	Education specific schemes (e.g., diplomas)
Trusted Issuer Registry (TIR)	Domain/Jurisdiction specific authorities (e.g., TAOs)	e.g., Accreditation agencies for educational domain	Trusted Issuers (TI) and TI accreditations (Educational Institutions and its accreditation to issue Diploma Schemes)
Revocation & Endorsement Registry (RER)	TAOs, TIRs	e.g., Educational Institutions	Educational Verifiable Credentials status (e.g., Student card or Diplomas status)

presents the description of the functional roles assigned in the use case.

Figure 6 illustrates the allocation of roles and responsibilities for the two scenarios: (1) Applying for exchange at KU Leuven, and (2) Bringing back the obtained credits to their home institution.

4.3. Implementation

The EBSI architecture is divided into multiple layers, with external applications not having direct access to the lower layers, but instead accessing them through the interfaces of the core services layer. This principal layer leverages EBSI capabilities by defining external interfaces, to enhance the speed of delivery of use case applications, providing a consistent and granular interface, and securing the network by creating a separation between external applications and infrastructure systems.

For the pilot, Una Europa chose to work with the Spanish software company Gataca to interface with the EBSI. Gataca provided three tools: (1) a credential issuance tool that enables trusted authorities to issue verifiable credentials, (2) a mobile ID wallet to store identity credentials, and (3) single sign-on authentication and credential verification tools for service providers. The credential issuance tool was installed at KU Leuven and Unibo sites to issue credentials. The wallet is used by Una Europa exchange students, and the sign-on tools were installed at KU Leuven and Unibo sites to verify credentials.

The pilot implementation took place from September to October 2021 and lasted four weeks. Reflection points were set with the EBSI technical team, to discuss progress each week. At the end of the period, the following tasks has been achieved: (1) issuance and storage of a verifiable student identity by Unibo, (2) presentation and verification of the verifiable student identity by KU Leuven, (3) issuance and storage of a verifiable transcript of records by KU Leuven, and (4) presentation and verification of a verifiable transcript of records by Unibo. The pilot provided a user journey of an imaginary student, Eva, and the visuals from the pilot and the description of scenarios are included as Supplementary Material.

4.4. Assessment

The pilot was completed within the timeline described in the action plan, and the demonstration of the pilot took place in November 2021 during the quarterly meeting of the European Blockchain Partnership. The pilot showed that the efficiency of the verification process could be drastically improved through the use of technical services provided by the EBSI. The different scenarios tested during the pilot showed that the whole process of verification could be completed within minutes. However, the pilot also revealed several challenges concerning the wider adoption of the use case. In this section, we present our assessment of the onboarding of actors and the issuance and storage of VCs.

During the pilot, the students (certificate holders) needed to use an EBSI-compliant user wallet to store the VCs issued by KU Leuven and Unibo. At the outset of the pilot, there were four user wallets that complied with the EBSI: Danube Tech, Gataca, Poste Italiane, and Validated ID. The Gataca user wallet was utilized by the project team in the pilot, but for the broader implementation of the use case, users should not be limited to a particular user wallet. Nonetheless, the interoperability between different wallets remains a challenge that needs to be addressed.

Before universities can be onboarded as trusted issuers in the EBSI, they require accreditation from a national TAO, such as a Ministry of Education, for the issuance of a verifiable student ID (V-STID) and a verifiable transcript of records (V-ToR). For KU Leuven, the Accreditation Organisation of the Netherlands and Flanders (NVAO) functioned as a TAO in a production environment. NVAO safeguards the quality of higher education in the Netherlands and Flanders. Alternatively, since KU Leuven is also registered on the European Quality Assurance Register for Higher Education (EQAR), EQAR could act as a TAO and issue a verifiable accreditation to KU Leuven. However, one challenge is to increase awareness among national authorities and clarify their future role within EBSI. In the pre-production environment, AHOVOKS (Flemish Agency for Higher Education, Adult Education, Qualifications, and Scholarships) was registered in the TAOR and acted as TAO (source of trust) for KU Leuven. Currently, Unibo is not registered on EQAR, and the Italian Ministry of Education acted as a TAO. In the pre-production environment, the TAO for Unibo was still to be determined.

Another challenge in the onboarding process was the lack of data schemes for the V-STID and V-ToR. These data schemes must be registered in the trusted schemas registry (TSR) before they can be issued by a trusted issuer. New schemes, or proposals to expand existing ones, must be requested, reviewed by domain experts, and approved by use case representatives, to ensure interoperability. Data schemes for educational credentials are also aligned with the EDCI (Europass Digital Credential Infrastructure) data model and ELMO/EMREX standard.

During the pilot, the EBSI specifications were still evolving, and thus, the onboarding process was not straightforward. Figure 7 illustrates the integration setup during the pilot. The project team experienced onboarding as the main challenge in the pilot. To overcome this challenge, it is important to convince stakeholders and governments of the added value of the EBSI infrastructure over other (European) infrastructures. Bilateral and multilateral exchanges with governmental and stakeholder organizations suggest that not many organizations are convinced about the added value of decentralized infrastructures such as EBSI and blockchain-based identity systems. The EC and the EBP are creating an entire ecosystem around the EBSI blockchain, but it must be widely adopted to prove its value.

For a wider adoption, the use case needs a way to authenticate the students before issuing a V-STID. An easier way to authenticate a student is through SSI-compliant digital wallets. The upcoming version of eIDAS foresees the issuance of digital wallets that are compatible with SSI elements to every European citizen (see Digital Europe Working Programme 2021–2022). This would require a verifiable id (V-ID) issued by the Belgian government being stored in the digital wallet of the student. The Belgian government has the necessary components to facilitate this: the Federal Authentication Service (FAS) coupled with the Belgian eID (https://dt.bosa.be/en/identificatie_beveiliging/federal_authentication_service (accessed on 31 January 2023)). For this, the Belgian government (i.e., BOSA) needs to be onboarded on EBSI as an issuer of V-IDs. This adds, however, an extra level of complexity to the use case. In the pilot, the project team chose to authenticate the student ID based on his/her student account.

The project team concluded that in the future, their preference would be to work with V-IDs issued by the respective governments, since universities are not the owner of the personal information of citizens. The V-ID could then be used as a means of authentication, especially for diplomas, but also to prefill admission forms or to provide other services with minimal data disclosure (e.g., that a student is more than 18 years old). In the Belgian context, the BOSA needs to examine this in the context of the hybrid EU digital identity wallet, which will also include VC support. BOSA is currently evaluating its role as an issuer of VCs and the wider adoption of the use case for both education and social security. In the absence of BOSA taking on the role of V-ID issuer, the project team can use the FAS to identify the students. Based on a DHO data service, the admission office can consult the authentic source of the personal data of the students and “simulate” a V-ID. Figure 8 illustrates the ideal system for the verification of education credentials in Belgium.

5. Discussion

During the pilot, a pre-production deployment environment was provided by EBSI. This environment allowed us to verify that the framework set by the EBSI and ESSIF provided the necessary conditions to fulfil the design principles and goals outlined in Section 4.1.2. Additionally, the pilot helped us assess the institutional and technical conditions required to scale up the pilot for actual implementation. However, to transition to the production phase, extra capabilities are needed. Most importantly, the EBSI requires the engagement of a large ecosystem, to have a significant impact on digital interactions between citizens and organizations. Specifically, the following dimensions require further attention:

• Onboarding in the EBSI ecosystem governance
• Issuance of EBSI-compliant digital wallets and data schemes for transcript validation
• Interoperability issues concerning digital identity systems

We will now discuss the main implementation challenges in each dimension.

5.1. Onboarding in the EBSI Ecosystem Governance

At present, EBSI governance and wallet providers such as Gataca are accredited to onboard legal entities as trusted anchor organizations (TAOs). During the pilot, KU Leuven registered a fake decentralized identifier (DID) for AHOVOKS using Gataca’s services. However, in the future, accreditation of TAOs should be carried out by authorized authorities in the European education ecosystem. For wider adoption, all relevant organizations must be included, and their roles should be clarified and communicated. In Belgium, these organizations are either NVAO or EQAR. Ideally, EQAR could serve as a hub for European universities, while NVAO could serve as the TAO for Flemish and Dutch universities. For now, a lot of organizational work needs to be done to facilitate the onboarding of European and national legal entities in the education domain. Further administrative and political efforts are needed to convince legal entities to fulfil their roles in the EBSI ecosystem governance.

Moreover, it is essential to include entities authorized to issue e-identities in the EBSI ecosystem, to ensure that verifiable identities (V-IDs) can be used as a means of authentication. Before KU Leuven can issue a verifiable student transcript ID (V-STID) to an active student, the student needs to authenticate themselves. Ideally, the student would provide a V-ID stored in their wallet. This would increase the level of assurance and comply with the forthcoming EU eID. BOSA is the only institute accredited to issue EU electronic IDs and that can provide a V-ID in Belgium. BOSA needs to provide the means to support the digital registration of a person, either as a classical eID or a verifiable identity credential, to prove the core identity, whereby the University can issue the V-STID.

During the pilot, the issuance of a V-STID went through the student portal of KU Leuven and Unibo. Alternatively, students at KU Leuven could authenticate themselves with their Belgian ID through the Federal Authentication Service (FAS). The pilot showed that accreditations could be created and registered on EBSI, but there was no way of validating them. The EBSI is required for the verification of the signature of verifiers, whether the issuer is in the trusted registry, and whether the organizer is in fact accredited by the authorities.

The EBSI should add additional functions, for a complete verification process (a complete verification process will be part of the Digital Europe Programme). KU Leuven can issue V-IDs, but they are not accredited for this. During the piloting phase, KU Leuven performed all the onboarding processes, but the verification functions were not yet active. For now, we lack information about the verification process, but Figure 9 illustrates how onboarding would take place in the implementation phase.

Finally, performance tests conducted by the EBP during the pilot phase suggested that the previous implementation of the software stack may require technical adjustments to support upscaling. When testing an EBSI node directly, the maximum rate was 120 requests per second, before a light degradation in service was observable. The response time for EBSI nodes was partially affected by various security and networking measures, including an external firewall, web application firewall, and internal node firewall. Considering that the EBSI network is a full peer-to-peer mesh of nodes sharing the same ledger, its scalability is directly proportional to the number of nodes in the network. At present, 35 nodes are fully operational, decentralized, and distributed across all member states. Multiplying 35 nodes by 120 requests per second, the current EBSI pilot environment can handle 4200 requests per second. Since many improvements were implemented at the API level, an increase in the previously measured rate is expected in the production phase. This performance measurement is comparable to other public permissioned blockchains such as Hyberledger Fabric [26] and Exonum [27], which reported 3000–5000 transactions per second. Although these numbers are comparatively lower than centralized storage systems that can handle up to 50,000 transactions per second [28], in the public sector domains, a higher security and privacy of transactions might be preferable over a high transaction speed.

In the design of the production environment, the EBSI is considering the balance between throughput, scalability, and privacy, and avoiding single points of failure or centralization. Previous research suggests that permissioned blockchains can be re-engineered to support 20,000 transactions per second through a series of optimizations, focusing on input/output, caching, parallelism, and efficient data access [29]. However, increased performance in blockchain often comes at the expense of security or transparency [30]. The EBSI governance reports that measures will be put in place to enforce anonymity of requests from relying parties, to verify information using the trust model provided by the EBSI network. While a trade-off between performance and anonymity in education verification might be acceptable, different technical solutions can be sought for upscaling in other use cases (e.g., social security), where the anonymity of transactions is more crucial.

5.2. Issuance of EBSI-Compliant Digital Wallets

In September 2021, the Belgian Federal Minister of Digital Transformation, Mathieu Michel, announced that by 2023, every Belgian would receive a digital wallet to manage identifications (https://www.brusselstimes.com/belgium/189705/belgium-to-simplify-administrative-processes-with-digital-wallet (accessed on 31 January 2023)). However, it is unclear how this digital wallet mechanism will be linked to the EU digital identity wallet specification and whether it will be compatible with the EBSI system. Furthermore, on 3 June 2021, the EC proposed a framework for a European Digital Identity, which will be available to all EU citizens, residents, and businesses in the EU. Citizens will be able to prove their identity and share electronic documents from their European Digital Identity wallets (https://www.enisa.europa.eu/news/enisa-news/moving-forward-in-securing-online-trust-via-the-digital-wallets (accessed on 31 January 2023)). The European digital identity wallets will be compatible with SSI solutions (i.e., hybrid wallets). To move the use case to the production phase, it is crucial to have clarity on the compatibility of Belgian digital wallet solutions with the EBSI and European digital identity wallet solutions, and how the future of digital identity systems in Europe will support ESSIF and EBSI solutions. The current political statements suggest that EBSI wallet solutions will be compatible with the next generation e-identity framework system in Europe (eIDAS 2.0). However, decision-makers in member states need to support the widespread adoption of ESSIF/EBSI solutions.

Currently, there is some hesitancy about providing political support for EBSI solutions in Belgium. There are two possible solutions for issuing EBSI-compliant wallets in Belgium. First, the BOSA has the Federal Authentication Service (FAS), and they can easily issue education and social security credentials through an EBSI-compliant wallet, which could be created or procured by the BOSA. Alternatively, the BOSA could implement a simple web app, such as in Spain (https://demo-id.aplicaciones.fnmt.es/ (accessed on 31 January 2023)), protected by the FAS (for ID verification using eID) and then issue a V-ID. For both options, the BOSA needs to be registered as a trusted issuer of V-IDs. Eventually, the system should work such that the student goes to a wallet, presents the ID, and obtains the V-ID.

5.3. Interoperability Issues concerning Digital Identity Systems

In June 2021, the EU Commission presented a draft of the eIDAS regulation, aimed at creating a new EU digital ID scheme (EUeID). The goal is to provide all citizens and businesses in the EU with digital wallets to identify and authenticate themselves, store a variety of other documents (such as diplomas), and present them in a verifiable manner (https://ec.europa.eu/commission/presscorner/detail/en/IP_21_2663 (accessed on 31 January 2023)). EBSI and ESSIF are important initiatives backed by the European Commission, to create a new digital identity ecosystem in Europe. The Belgian government is currently reflecting on different approaches in Europe and waiting for a definition of a hybrid approach (a hybrid approach refers to a system that can leverage both decentralized technologies and centralized PKI solutions), but has some reservations about participating in these initiatives, primarily due to the legacy context of the digital identity system in place. Further efforts are needed by the EBSI team to discuss the benefits and costs of implementing a decentralized identity system with the Belgian federal government.

Additionally, in October 2021, the Flemish government announced a new partnership with SOLID, to create data safety for the Flemish people (https://inrupt.com/flanders-solid (accessed on 31 January 2023)). SOLID is an open-source and modular solution that links open data with decentralized storage, but with a centralized consent and verification architecture. SOLID is still at an early stage of exploration, it does not comply with W3C standards, and usage of VC is only possible in limited experiments (https://lists.w3.org/Archives/Public/public-solid/2022Nov/0001.html (accessed on 31 January 2023); https://ontochain.ngi.eu/content/solid-verif (accessed on 31 January 2023)). The technical description (https://solid.mit.edu/ (accessed on 31 January 2023)) of SOLID states that users can choose to switch between apps and personal data storage servers without losing any data or social connections. SOLID can supposedly be used on a blockchain or in centralized databases. However, it is currently unclear on what data governance platform the Flemish government plans to use SOLID, but the project aims to develop its own solution in four years. For now, it is not clear how this initiative will be interoperable with the EBSI/ESSIF systems.

The EBSI is trying to synchronize efforts in the context of the EU-eID wallet with the EU digital identity framework. The challenge is to create an EBSI-compliant and interoperable wallet that can bring together issuers, holders, and verifiers. To achieve this, there is a need for a synergy with the European digital identity framework. Some non-formal agreements have been made by European counterparts to synthesize the digital identity systems in Europe with ESSIF, but for the moment, it is still unclear how the digital identity systems in Europe will evolve. Clarification of these issues is crucial for the adoption of ESSIF-compliant solutions. On a broader scale, EBSI systems also need to be interoperable with other cross-border infrastructures and concepts such as the single digital gateway, Europass, and GAIA-X.

Furthermore, there are interoperability issues between wallet providers. Currently, 17 wallet providers have passed the EBSI wallet provider conformant tests (https://ec.europa.eu/digital-building-blocks/wikis/display/EBSI/Conformant+wallets (accessed on 31 January 2023)), but there are still issues when an issuer uses a wallet provider with another user wallet (e.g., standardization of notification when a verifiable credential is available for a wallet user, and making them eIDAS/EU-eID compliant).

6. Conclusions

The EBSI is a significant European initiative that supports the expansion of blockchain and blockchain-backed solutions in public services. This article presents the results of a pilot project that aimed to verify education credentials through the EBSI blockchain in a cross-border setting. This pilot was the first case to utilize this new blockchain infrastructure in a cross-border setting and tested the institutional, technical, and user-specific requirements for wider adoption. It assessed new digital technologies, such as digital wallets, verifiable credentials, and decentralized identifiers within this blockchain infrastructure.

All EU citizens are entitled to move and work in another EU country, without restrictions. To facilitate this, national and European public sector organizations issue various credentials and documents that need to be verified (e.g., portable documents, European Health Insurance Card—EHIC). Most of these credentials are provided in paper format, and the verification process is complex, time-consuming, and error-prone. The synergy of EBSI verifiable credentials with ESSIF presents a promising area, where blockchain technology can significantly improve the efficiency and traceability of official documents in cross-border settings. The research suggested that the EBSI blockchain could be a key European infrastructure to support cross-border services, but its widespread adoption depends on onboarding national and international institutions and private service providers in the EBSI ecosystem, as well as tackling interoperability and compatibility issues within the broader EU digital identity framework. Addressing these issues requires the collaboration of national, European, and market actors.

Studies have shown that, despite their potential, the public sector faces significant bottlenecks in moving their blockchain use cases into the production stage [7,14]. The findings of this research reconfirmed that the scaling up and wider implementation of the pilot would require coordinated institutional, organizational, technological, and regulatory adjustments. One significant challenge that identified in the research is ensuring the interoperability of the various systems in digital verification and overcoming the institutional complexities surrounding the issuance of the identity of legal persons and organizations.

To fully implement the use case, creating synergies and interoperable systems between different national and supranational strategies is crucial. Furthermore, integrating digital education credentials with other European mobility education systems (e.g., Europass, Erasmus without paper) is necessary. However, there are technical, legal, and organizational gaps that need filling to create a fully integrated digital education system in Europe that can fully leverage SSI solutions. It is also important to explore the compatibility of EBSI with other cross-border infrastructures and concepts, such as single digital gateway and GAIA-X. Further research is recommended, to explore how seamless and interoperable cross-border information systems can be created between these different decentralized and centralized infrastructures. With the advancement of other EBSI use cases, it would be possible to draw conclusions across other public sector domains. The implementation of social security credentials is another use case announced by the EBSI network (https://ec.europa.eu/social/main.jsp?catId=1545&langId=en (accessed on 31 January 2023)). Future similar research focusing on the social security domain could help us to gain further insight into the comparability of implementation challenges in different sectoral domains.

Finally, the authors note that the institutional framework created by the European Union and the complex governmental relations in the Belgian federal structure create certain idiosyncrasies. They acknowledge that these idiosyncratic conditions may limit the generalizability of the findings, especially regarding the impact of institutional and interoperability dimensions. Moreover, blockchain as a technology requires a holistic assessment of institutional, organizational, technological, and regulatory dimensions for digital transformation [31]. The complexities across these different dimensions have not been fully explored in the existing literature. Therefore, replication of similar studies in and across country cases could give us further insight into the generalizability of the findings.