1 Semantic reusable web components: A use case in e-government interoperability 
Slavko Žitnik 1, Karmen Kern Pipan2, Miha Jesenko2, Dejan Lavbic 1 1 Univerza v Ljubljani, Vecna pot 113, Sl-1000 Ljubljana 2 Ministrstvo Republike Slovenije za javno upravo, Direktorat za informatiko, Tržaška cesta 21, Sl-1000 Ljubljana slavko.zitnik@fri.uni-lj.si, karmen. kern-pipan@gov.si, miha.jesenko@gov.si, de1an. lavbic@fri .uni-lj .si 
Abstract 
Advancements in technology and software engineering strive to efficiently build robust technologies and deliver them as fast as possible. It has been shown that both could be achieved by reusing existing implementations, libraries, components, and even fra­meworks. In the field of public administration, which needs to follow and implement national and EU regulations, it is essential to not only have compatible semantic data structures and processes but also to enable cross-sector and cross-border integrations. In this paper, we present the Oigital Europe for Ali project, more specifically, a use case of a reusable prototype -a web component based on semantic data representation. We show that the component can be easily integrated into an arbitrary website and supports ar­bitrary evidence types, represented using the project's ontologies. Reusing such components enables !aster creation of interoperable digital public services and opens new opportunities, better mobility, !aster processes, and reduced implementation costs. 
Keywords: Reusable components, building blocks, digital public services, reusability, software engineering, semantics 

Ponovno uporabljive semanticne spletne komponente: Primer uporabe za interoperabilnost digitalnih javnih storitev 
Izvlecek 

Napredek v tehnologiji in programskem inženirstvu si prizadeva graditi uci nkovit e in robustne tehnike, ki bi se jih dostavljalo cimprej. Dokazano je bilo, da je to možno doseci s ponovno uporabo obstojecih implementacij, knjižnic, komponent ali celo ogrodij. Na podro­c ju javne uprave, ki mora slediti in implementirati nacionalne in evropske predpise, je poleg združljivosti na podatkovnem nivoju in procesov pomembno zagotoviti cezsektorsko in cezmejno integracijo. V tem prispevku predstavljamo projekt "Oigital Europe for Ali," bolj natancno primer uporabe ponovno uporabljive spletne komponente, ki temelji na semanticni predstavitvi podatkov. Prikažemo, da je komponento možno enostavno vgraditi v poljubno spletišce in podpreti poljubne tipe dokazil, ki so predstavljena s pomocjo projektne ontologije. Ponovna uporabljivost takšnih komponent bo hitreje zagotovila interoperabilne digitalne javne storitve in bo odprla nove priložnosti, boljšo mobilnost, hitrejše procese ter bo zmanjšala stroške implementacij. 
Kljucne besede: ponovno uporabljive komponente, gradniki, digitalne javne storitve, ponovna uporabljivost, programsko inženirstvo, semantika. 
INTRODUCTION Nowadays, we use many software systems based on large frameworks used by developers to build uni­fying archi.tectures and easy-to-use graphical inter­faces for the end users. U sers leverage such software systems to support their business processes, and du­
ring that, they work on similar acti.vities -e.g., ente­ring data for multi.ple employees, showing diverse tables or updati.ng data. Building blocJ..s that enable business processes, for example, to enter the address of an employee or a company, should behave similar­ly. For software developers to achieve that, they need to define conunon c1iteria for their software system to unify them and the users. Building blocks are so-cal­led components, defined as follows: "An individual softvvare component is a software package, a web service, a web resource, or a n1odule that encapsula­tes a set of related functions ( or <lata). 1" 

Software is rarely built from scratch, but to a gre­at extent, some paits of code can be reused, copied and adapted to fit new requirements. For that reason, software process models were developed to provide guidance and create high-quality software at predic­table costs. Most of the new software is thus based on the systematic use of well-defined components designed for various projects. Software engineering researchers defined the need for components already in the 1960s (Mcllroy et al., 1968), while the subfi­eld Component-based software engineering (CBSE) more widely developed and provided architechual solutions in the 1990s (Sametinger, 1997). During that time, prominent commercial computing infrastruch1­re vendors followed mentioned ideas and presented their solutions for their progranuning frameworks, such as ASP.NET or Java (Kozaczynski and Booch, 1998). 
Some researchers argue that focusing on indepen­dent components for modem software engineering is a mistake. Stili, the focus should instead be on frame­works that even enable the existence of components (Wallace, 2010). l11e recent rise of Web prograinming frameworks showed precisely the opposite. These frameworks define components transpiled (i.e., tran­slation of a sotuce code to another language) into Ja­vaScript, cai1 be used completely independent of spe­cifically used Web technologies and allow for custom parameterized user interface styling (Gackenheimer ai1d Paul, 2015). 
Large <lata economy systems that we can find in the public sector deal with enom10us variety and amounts of <lata. This <lata is useless if it caimot be easily and correctly understood, accessed or combi­ned. As software components can achieve interope­rability ai1d reuse on a technical level, they caimot solve p roblems on a semantic level. In that sense, govemments pushed the development of open <lata initiatives where ali public <lata would be described in a machine-readable format. Semantic Web teclmo­logies describe the data, provide unifying scl1emas, and overcome problems of matching <lata fields only by short textual descriptions (Narducci et al., 2013). 
The teclmologies mentioned above are the main building blocks of future interoperable systems. For the needs of globalization ai1d cooperation of public administrations between EU member states, it is also crucial to enable semantic cross-sector and cross-bor­der <lata interchange. Of course, privacy, security, storage, ai1d other aspects of large software ecosy­stems need to be provided, but they are not the focus of this paper (GovStack, 2021). 
This paper presents a use case for building a Web­-based and semantically enabled reusable prototype -A web component for cross-border public admi­nistration interoperability. The rest of the paper is struchued as follows. We first review technologies that have already tried to standardize and intercon­nect different pub lic administrations in the EU. Then we continue with the brief presentation of the Digital Europe for Ali project 
This article delves into specific use-case dealing with semai1tic ai1d reusable prototype components as part of the field in the project group for Semantic interoperability solutions, where the authors of this article have been actively involved. We evaluate its adaptability and reusability and provide our view for further development. 
2 RELATED TECHNOLOGIES SUPPORTING DIGITAL PUHLIC SERUICES 
The previous section brietly overviewed general ideas for building reusable components in software engine­ering. A well-defined market segmented by Gartner already exists, and it is predicted that by 2024, 65 % of application development will be low code. In additi­on, recently, low-code ai1d no-code platforms are be­coming popular, leveraging reusable components. In this section, we focus on the state of provisioning the saine set of functionalities to public administration. 
Ideas for building reusable components in the public administration domain followed the deve­lopment of CBSE in the 1990s (Castano et al., 1995). At that time, <lata was more tightly coupled with ap­plication logic, and therefore first approaches focu­sed on identifying similar subschemas in Entity-Re­lational diagrams. Based on those, the development of common components was proposed. Also, most 
1 https://en.wikipedia.org/WikVComponent-based_software_engineertng (Accessed: July 29, 2022). 
of the work later focused on model-driven deve­
lopment for digital public services. There have been a lot of different models proposed by the UK, the EU and the USA to provide interoperability on various levels (Peristeras et al., 2008; Peristeras et al., 2009). 
The United Nations member states defined 17 su­stainable development goals representing the 2030 Agenda to connect ali countries in a global partner­ship, improve health and education, reduce inequa­lity, and spur economic growth. For the nations to even enable actions, their systems supporting digital public services must be interlinked. Apart from co­untry-specific initiatives, multi-stakeholder alliances have also been formed that try to provide open-sour­ce software, models, content, software architectures and standards -e.g., Digital Puhlic Goods Alliance, Go­vStack or eGovemments Foundation. 
One of the key players in enabling and supporting cross-border and cross-sector interoperability betwe­en countries is the EU. It defines standards, connects multiple countries or stakeholders, and funds pro­grammes and projects. Under the European Com­mission (EC) umbrella, several events are organized ( e.g., SEMIC2) to promote and share new knowledge. EC also leads some projects (e.g., eSENS3, TOOP4) to address common topics or provide general frame­works that multiple stakeholders in member states can reuse (e.g., the DE4A project5, SDG6) . In the last years, CEF Digital7, ISA, ISA2 8 and Interoperable Eu­rope9 have delivered many valuable building blocks and solutions for public administrations to build in­teroperability at all levels, following European Inte­roperability Framework10. 
In this regard, DE4A aims to establish a culture of co-creation, transparency, accountability and tru­stworthiness. Citizen and business-oriented pilots shall highlight chosen aspects of the teclmology ecosystem available for the Single Digital Gateway Regulation (SDGR11) implementation at the Euro­pean and Member State level, prove their teclmical viability and gauge the perfom1ance and degree to 
• lnteroperability Platform 
• Once-Only Technical System 
• 
Existing Building Blocks 
Figure 1: Stages towards a fully interoperable platform lsource: https://www.de4a.eu, accessed July 29, 20221. 
which non-functional requirements can be accom­modated12. We believe experiences, good practices, and infom1ation gained in the DE4A project will be of uhnost importance for the further development of cross-border digital public services in member states and Slovenia. 
3 DIGITAL PUBLIC SERUICES ANO CROSS-BDRDER INTERDPERABILITV 
This section presents the details of the Digital Eu­rope for All (DE4A) project, specifically related to the component we will more technically describe in the second part of the section. The D E4A project shows a possible path to full interoperability betwe­en different cross-border digital public services. The D E4A project started with the main building blocks supporting the SDGR's Once-Only Teclmical System (OOTS) and offer.s a way to interconnect cross-bor­der public services. Figure 1 shows the stages of buil­ding such a platform. 
3.1 The DE4A project 
The Digital Europe For All (DE4A) project builds on prior work toward EU Digital Single Market to rea­lize the Once-Only Principle. DE4A effectively puts forward a new Member State-driven large-scale pi­lot aimed at compliance with Single Digital Gateway and aligned with EU eGovernment Action Plan 2016­202013, Tallinn Declaration and EIF Implementation Strategy. lts overarching goal is to reinforce trust in 
2 hnps:/twww.de4a.eu (Accessed: July 13, 2022) 

3 hnps://single-market-economy.ec.europa.eu/single-markeVsingle-digital-gateway_en (Aocessed: October 12, 2022). 
4 hnps://ec.europa.eu/oefdigital/WikVdisplay/CEFDIGfTAUCEF+Digital+Home (Aooessed: July 14, 2022). 
5 hnps://ec.europa.eu/isa2/home_en/ (Acoessed: July 14, 2022). 

6 hnps://joinup.ec.europa.eu/collection/interoperable-europe/interoperable-europe (Aocesed: July 26, 2022). 
7 hnps://ec.europa.eu/isa2/eif_en/ (Aooessed: July 14, 2022). 

8 hnps://eur-lex.europa.eu/legal-contenVEN/TXT/?uri=uriserv:OJ.L_.2018.295.01 .0001.01.ENG (Aooessed:October 11, 2022). 
9 hnps://ec.europa.eu/archives/isa/library/documents/eu-egovemment-action-plan-2016-2020_en.pdf (Accessed: July 29, 2022). 10 https://eur-lex.europa.eu/legal-content/SL/TXT/?uri=CEL.EX%3A32016R0679 (Accessed: July 26, 2022). 11 https://eur-lex.europa.eu/legal-content/ENITXTn uri=uriserv%3AOJ.L_.2014.257.01 .0073.01 .ENG (Aocessed: July 26, 2022). 12 hnps://eur-lex.europa.eu/legal-<:antenVEN/TXT/?uri=oelex%3A32006L0123 (Aocessed: July 26, 2022). 13 hnps:/twww.de4a.eu/about-proJect, (Acoessed: July 21, 2022). 
public institutions and to unleash multiple measu­rable positi ve impacts in tenns of efficiency gains and reduction of current administrative burden and costs, rooted in a Toolkit for extended semantic in­teroperability and in the secure, privacy-preserving and trustworthy realization of fundamental once­-only, relevant-only and digital by default principles, through state-of-the-art, usable and high-quality ful­ly online procedures accessible through the Single Digital Gateway defined in SDGR. The DE4A takes reality as its starting ground -the needs and the capa­cities of the Member States (MS), ensures compliance with ali other relevant regulations (GDPR14, eIDAS15 , Services Directive16, among others) as well as EU and MS strategic eGovenunent and Single Market goals and aims to transfom1 governance towards open col­laboration and innovation, adaptable to depart from the variety of current environments17 . It will allow for full regulatory compliance, establishing a culh1­re of co-creation, transparency, accountability and trustworthiness. For a completely working digital single market, effectively enabling the cross-border exercise by citizens and businesses of their Single Market rights, EU MS must address several challen­ges in delivering better services. DE4A is a three-ye­ar pilot-led project started on January 1 2020, in su­pport of the SDG across Europe ((EU) 2018/1724) and helping to make the single digital market a reality. DE4A includes 22 parh1ers, including eight Member States (Austria, Luxembom·g, Netherlands, Portugal, Romania, Slovenia, Spain, and Sweden). The project establishes a culture of co-creation, transparency, ac­cotmtability and trustworthiness. Its goal isto facili­tate migration towards European Digital Public Ser­vices co-delivered across borders, across sectors and with different participants. The project has received ftmding from the European Union's Horizon 2020 re­search and innovation program tmder GA No. 870 63518. ATOS leads the DE4A project, and it consists of several different work packages (project groups) covering different areas as follows: Inventory of cur­rent eGovemment la.ndscape; Architechue vision a.nd framework; Semantic interoperability solutions; Cross-border pilots for citizens, business and evalu­
ation; Common component design & development; Sustainable impact and new goven1ance models; Le­gal and ethical compliance and consensus building; Stakeholder dialogue, dissemination and communi­cation; Project coordination and management and Ethics requirements. 
The project is validating a. comprehensive and scalable approach to develop and demonstrate the potential and benefits of securely and digitally sha­ring relevant-only and once-only data. DE4A aims to establish a culhue of co-creation, transparency, acco­untability and trustworthiness. Citizen and business­-oriented pilots shall highlight chosen aspects of the teclmological ecosystem ava.ilable for the SDGR im­plementa.tion at European and MS levels, prove their teclmical viability and gauge the performance and degree to which non-functional requirements can be accommoda.ted. The pilots comprise ftmctional use cases requiring interaction with real users -citizens, students, businesspersons and public serva.nts. The three pilots are as follows: 
• 
Studying Abroad Pilot will enable pa.perless pro­cedures for sh1dents' mobility and simplify cross­-border interactions for sh1dents engaging with processes, systems and pla.tfom1s used by higher education establishments. It will allow <ligi.tal (a) applications for higher education, (b) applications for study grants, (c) and diploma.s/certificates/ sh1dies/ professional recognition. 

• 
Doing Business Abroad Pilot will lower barriers for companies to starta business or conduct busi­ness in another EU MS. It will enable businesses to (a) retrieve and keep up-to-date company data from authentic sources, (b) starta business and (c) digital annual reporting. 

• 
Moving Abroad Pilot will enable EU citizens' mobility across the EU and support them in mo­ving from one MS to another by creating a digiti­zed process for exchanging infom1ation required to facilitate a cha.nge of residency and other onli­ne cross-border procedures relying on civil stah1s certificates.19 The processes will support (a) regi­stering change of address, (b) civil status certifica­tes, and ( c) pension claims and infom1ation. 


14 hnpsJlwww.de4a.eu/_files/ugd/2844e6_ 1e295e0aa15c4afa8f711baa52001189.pdf, (Accessed July 21, 2022). 15 hnpsJlwww.de4a.eu/_files/ugd/2844e6_ 1e295e0aa15c4afa8f711baa52001189.pdf, (Accessed: July 21, 2022). 16 hnpsJ/Wiki.de4a.eu/index.php/Birth_GanonieaI_EVidence (Accessed July 26 2022). 17 hnpsJ/github.corTVde4a-wp3/MOR (Accessed: July 15, 2022). 18 hnps:/lwww.de4a.eu/ _files/ugd/2844e6_ 1e295e0aa15c4afa81711 baa52001189.pdf, (Accessed: July 21, 2022). 10 hnps:/lwww.de4a.eu/ _files/ugd/2844e6_ 1e295e0aa15c4afa81711 baa52001189.pdf, (Accessed: July 21 , 2022). 
Data Data 
Pattem: 
Consumer 
III 

1 
DE4A Semantics DE4A Common Building Blocks Solutions Components 
User Service lnteroperability Solutions Toolbox Figure 2: The DE4A project Service lnteroperability Toolbox lsource: https://wiki.de4a.eu, accessed July 29, 20221. 
The project's main deliverables will compre­hensi vely document and validate technologies that must be implemented for the cross-border integra­tion of public services. Figure 2 shows the central and long-tem1 deliverable of the proposed architec­ture framework. Some of the final deliverables will define ali aspects, such as modelling requirements, processes, sen1antic solutions with nwdels, conm1u­nication messages and patterns and full framework implementation. The results will be demonstrated w ithin the production-level implementation of spe­cific cross-border and cross-domain pilots descri­bed above. 
The key business roles identified by the DE4A are llser, Data Consumer (DC), which consists of Data Evaluator (DE) and Data Requestor (DR), Data Provider (DP), which consists of Data Transferor (DT) and Data Owner (DO), and Registrar. User means a natural or legal person that accesses the system (uses the servi­ce). The registrar is responsible for accept:ing and sto­ring formal information according to rules of law or governance. Data Consumer is an organization/admi­nistration in demand of the data, while a Data Provi­der is a legal entity in charge of the <lata deployment. It is essential to understand their specialized roles: 
• Data Consumer (DC) -Data Evaluator (DE): organization authorized to receive and process data from a User via OOTS. -Data Requestor (DR): a technical role that sear­ches and requests data. If DR is a separate entity from DE, it carries the request under the mandate of the DE. 
• Data Provider (DP) 
-Data Owner (DO): organization which governs information about the User. DO is responsible for authorization approval, data extraction (also from secondary registries) and audit control. -Data Transferor (DT): a technical role respon­sible for the actual <lata transmission. It runs the OOTS-compliant data service. DT and DO can be different entities -e.g. when a national or sectorial intermediary is acting as an evidence broker. These business roles can exchange struchrred 
data (i.e., evidence) or related information following Once-Only Interaction Patterns. The DE4A Architec­hrre Framework defines five reference interaction pattems: 1. Intem1ediation Pattern, 
2. 
User-supported Intermediation Pattern, 

3. 
Verifiable Credentials Pattern, 

4. 
Lookup Pattern, and 

5. 
Subscription and Notification Pattern. 


As the reusable prototype component we descri­be in this paper uses only the User-supported Inter­mediation Pattern or Intem1ediation Pattem, we will present these only. Other patten1S are primarily used for communication between systems or credentials­-related needs. Figure 3 shows an illustration of the Intermediation Pattern. In this pattern, a user initi­ates a procedtrre through a public service request. The DC is missing the required evidence that can be automatically acquired from a known DP. After the evidence ( one or more) is retrieved, they are shown to the user, who can decide to complete the procedu­re. In some cases, the preview part might also allow 

Request service 
0 
® Request evidence ® Transfer evidence Preview evidence 
0 
® Submit eProcedure 

User Data Consumer Data Provider 
Figure 3: lllustration of the lntermediation Pattern. 
a user to upload signed evidence in his possession manually. 
In Figure 4, we show an example of a User-su­pported Intermediation Pattern. In tlus pattem, a user initiates a procedure through a public service request. In case when a DC needs additional eviden­ce, it notifies a user and redirects him to identify and authenticate himself with a selected DP. In the me­antime, the DC has requested the missing evidence, shown to the user via a preview component. If the user approves the evidence, DP sends it to the DC. If DC is missing more evidence, tlus process will be repeated. When ali evidences are acquired, the user can decide to submit the initiated procedure. 
3.2 Reusable prototype component use case: Multilingual Dntology Repository (MORJ 
In the scope of the DE4A project, we have designed and tested the MultilinguaI Ontology Repository (MOR) prototype component for possible further 
use in semantic data exchange among different EU 
member states. TI1e MOR component aims to regi­
ster and provide information on concepts and terms 
involved in the evidence exchange and support the 
automatic generation of customizable user interfaces 
(i.e., for explicit request, preview, and additional pa­
rameters request) that contain complex tem,s in any 
EU official language. TI1e component is developed 
as a prototype and is meant to be integrated into an 
arbitrary organization/administration portal that of­
fers ftmctionalities to users (e.g., evidence preview 
in step 5 of User-supported Intennediation Pattern 
communication). 
11,e D E4A defined clear semantic and ftmctional 

requirements for the component. Semantic require­
ments are highly related to MOR terms which are 
semantic elements (simple or complex) represented 
by a Unifom1 Resource Identifier (URI), syntax and 
multilingual meaning. Examples of complex types 
are canotli.cal evidence types ( e.g., birth certificate20) 
User 
@ Request evidence 
(2) Transfer evidence 

Data Consumer Data Provider 
Figure 4: lllustration of the User-supported lntermediation Pattern I* in a case when additional evidence is requestedl. 
that MOR needs to be able to process. Each tenn must be described using a label, description and an example in every EU official language. Suppose a de­scription and an example are not provided for the target language. In that case, the MOR component needs to translate values automatically -these values must be shown to the user using the "non-verified" label. Complex terms are modelled as a tree hierar­chy of tenns, and a path within the hierarchy unique­ly identifies them. If a complex term [X] is defined of type [Y], the [X] sub-tem1s have the same URI as the [Y] sub-tem1s except for the root element, so the root of [Y] tem1 paths is replaced by the [X] URI. For example, the sub-tem1 "Gender" of the complex tenn " BirthEvidence/Child" of type "Person" corresponds to the "Person/ Gender" su b-tem1 bu t wi th URI "BirthE vi­dence/Child/ Gender." MOR also contains code lists for enumeration types. Ali tem1s must be reusable. For example, the complex term "BirthEvidence/Child" of type "Person" may overload the multili.ngual mea­ning label and description for newborn persons in­stead of persons in general. 
Apart from semantic and syntax support, MOR must enable the automatic generation of customiza­ble user interfaces for any complex term in any EU official language. TI1ere are three cases where tlus functionality can be of help: 
• 
The explicit request functionality should infonn the user of the infom1ation to be requested as evi­dence, so the MOR can help to create a building block that generates such a user interface for any canonical evidence type and language. 

• 
The preview functionality should show the user the evidence to be incorporated into the proce­dure, so the MOR can help to create a building block that generates such a user interface for any canonical evidence type and language; audits can also use thls building block to help auditors to un­derstand any canonical evidence in any language. Titis is particularly interesting when the preview space is located on the evidence provider side. TI1e language on thls side can differ from the procedu­re requiring cross-border evidence, and the user may not understand the provider's language. 

• 
The additional parameters functionality requires the user to request some fields through a fom1, 


so the MOR can help create a building block that generates such a fonn for any additional parame­ters and language set. Tius is of particular interest when additional parameters have to be required on the evidence evaluator side because the evi­dence provider sets these parameters. In thls case, the type of the tem1s is the key to generating the appropriate input field in the fom1 ( calendar, se­lect list, text box, etc.) 


3.3 Technical implementation 
In thls section, we overview the implementation of the prototype -MOR component, show options for its possible usage, and comment on its reusability and extensibility. TI1e implementation is publicly available in a GitHub repository21. 
MOR implementation has two aspects: (a) the storage of the MOR tem1s according to the specifi­cation, and (b) the standard Grapltical U ser Interface (GUI) components to provide, in a multilingual way, a custontizable user interface for the explicit request, evidence preview and requesting additional parame­ters for the record matching. Within the DE4A pro­ject, GUI for requesting additional parameters has not been considered, and some simplifications have been adopted to implement the MOR storage to re­duce the technical complexity of its implementation. Since only the DE4A Moving Abroad pilot will use the component, MOR storage contains only canoru­cal evidence types for Birth (BirthEvidence) and Mar­riage (MarriageEvidence). 
3.3.1 Simplified MOR storage 
MOR needs to store ali MOR terms required for the three functionalities. As in the production enviro­nment, evidence types can be modified, or new types can be added; MOR should access these types via centralized services. 
In our case, MOR storage is implemented as JSON files, one per available language, offered by a REST API. After the modelling, we generated files that re­present ali supported canorucal evidence types (i.e., BirthEvidence and MarriageEvidence), complex tem1s used as types of other complex tem1s (e.g., Person, Location), and code lists (e.g., the EU Mem­ber States, administrative-territorial units, languages 

20 hnps://wiki.de4a.eU/index.php/Birth_Ganonical_Evidence (Accessed JUly 26 2022). 21 hnps://github.corn/de4a-wp3/MOR (Accessed: July 15, 2022). 
codes, currencies, gender). Available languages are the ones related to the Member States that participate in the Moving Abroad pilot: French (fr), Portuguese (pt), Slovenian (sl), Spanish (es), and, as the common language, English ( en). 
Each MOR term is comprised of five elements: 

l. Term URI: URI is the path corresponding to the XML element of the term in the XML schema of the complex. For instance, the root MOR tenn of the BirtlzEvidence:1.0 canonical evidence type would be "BirthEvidence", i.e., the root element of the corre­sponding XML schema. The rest of the sub-elements of such XML scl,ema are specified as separate MOR tem1S, sucl1 as " BirthEvidence/CertifiesBirth!Child." 
2. 
Origin: Source of the type of tem,, either DE4A or other XML namespaces, sucl1 as the ones used by ISA2. As the MOR component does not need to be aware of its origins, this field is not included in the MOR schema. 

3. 
Type: Depending on the nature of the term, accor­ding to the defined schema: -XML base type with prefix "xsd:" (e.g., xsd: 


string) for a simple tem1. -"xsd:enumeration" for a complex term defined as a code list. -"xsd:token" for a simple term defined as an ent­ry of a code list. -URI of another MOR term specified in a diffe­rent schema (e.g., MOR term LocationAddress as 
"BirthEvidence": { "t ype": "", "cardinality": "comment": "es": { 
"label": "Prueba de nacimiento", "description": "Informaci6n sobre "exampl e": "", "verified": "true" 
} 
}, 
a type of the MOR term BirthEvidence/Certifies­Birth/Child/ PlaceOfBirth). 
-No type for root MOR tem, s. A complex tem, whose sub-elements are included in the same schema, so their URI starts with the URI of the con1.plex te1m (e.g., BirthEvidence). 

4. 
Cardinality: Represented as two Boolean values. The first value represents if the term is mandatory, and the second represents if the tem1 can have mul­tiple values. Root MOR tenns have no cardinality. 

5. 
Comments: Optional text to describe the tem1 specification. 


3.3.2 Multilingual Dntology Repository Application Programming lnterface (MOR APIJ 
TI1e MOR API is defined to be the same for future im­plementations of the MOR storage with an existing database system. The MOR API retrieves ali MOR terms for a given language via one REST endpoint: 
"lmor/{lang}.11 
The JSON object retumed by the MOR API conta­ins MOR tenn values (presented in the previous sec­tion) and language-specific type values. These values are needed as metadata to present the retrieved sche­ma to the user. The "{lang}" JSON object consists of a label, a description, an example, and a ve1ified ilag. 
Examples of MOR temi. objects when Spanish is the selected language: 
un nacirniento que prueba su registra.", 
"BirthEvidence/CertifiesBirt h/ Child/PlaceOfBirth": { 
"type": "Person/PlaceOfBi rt h", "cardinality": "10", "cornment": "Country and place of birth", "es": { 
"label" : "Pais y lugar de nacimiento", "description": "Pais y lugar donde naci6 un nifto", "example": "Spain, Cercedilla", "verified" : "true" 
} } 

Note that "BirthEvidence" is the root tenn in Birth Canonical Evidence Type, so values of "type" a.nd "cardinality" are empty. 
TI1e JSON specification of the MOR API response is aimed at easing the finding of MOR tem1s when using a JavaScript JSON vaiiable. Because of this reason, the MOR specification of the tem1 with URI "BirthEviden­ce/CertifiesBirth/Olil.d/PlaceOfBirth" cai1 be easily obtai.ned by the expression javascriptJson Var[URJ]. 
3.3.3 Reusable and customizable MOR GUI 
Explicit request and evidence preview are required functionalities by every DE and DO when using a 
«subsystem» 
MOR Explicit request 
C] 
«component» 
Country 
selector 
C] 
User-supported Intem1ediation Pattem. Multilai1g11­age support is vital as a user from ai1y member state cai1 use services in any other member state. 
As evidence types were thoroughly defined using semantics, schemas make it possible to defi­ne automatically customizable GUI. The look and feel of the component ca.n be completely chai1ged by providing a customized Cascading Style Sheet file (CSS). Each HTML DOM object within MOR is identified by its id; therefore, a.n integrator ca.n completely redesig.n the component's user interfa­ce. The component is implemented as a.n AngularJS component and can be included into ai1y Web page 
o 
o 

«component»  «component»  
File upload  XMLpreview  
o «moda/,. Provislon retrleval and selection  «extemal Jnterface» Provislons per  o «extemal» IAL  
country and  
canonical  
C]  evidence type  
«component»  
Language  
selector  
«extemaJ interface»  
MOR Explict request  
input parameters  C]  
*extemal»  
Evidence  
MOR Expllclt request  retri eval  
outp ut parameters  
«moda/» Prevlew  MOR  
o «component» Language  Prevlew input parameters  
selector  
o  
<eCOmponent»  «intertace»  
XML prevlew  
nanslated  
content  
C]  
«component»  
Language  
selector  
«extemal interface»  
<J  
MOR Preview output parameters  

Figure 5: Multilingual Dntology Repository component diagram. 
using HTML placeholder <app-mor-er ... ></app­-n1or-er>. 
In Figure 5, we show the MOR component dia­gram. The blue denotes MOR implementation, while the red represents the external system (i.e., a specific Web page) where the MOR component could be in­tegrated. MOR component requires a definition of a few parameters to be placed within the target Web page. Results of the component are returned into a JavaScript variable within the embedded Web page (as parameterized during placement). Below, we describe the configuration and running of explicit request and evidence preview functionalities. 
<app-mor-er 
defaultlang = "en" 
requesterCountryCode = "BE" 


Explicit request 

A page of the DE portal opens the MOR GUI page for the explicit request. The parameters that DE needs to provide when embedding the component into a Web page are the following: (a) a default component language, (b) the DE country, (c) a list of Canonical Evidence Types identified by the MOR root tenns, and (c) the name of a JavaScript Map variable to fill by the MOR component (the hosted Web page can then use this data). An example of embedding a MOR component for the explicit request functionality: 
canonical EvidenceTypes = ''BirthEvidence,MarriageCertificate" output JSMapi d = "outputJSMapidMorEr" 

></app-mor-er> 

When the page with the embedded MOR compo­nent is open, the MOR API is called to retrieve the MOR JSON in the given language. Figure 6 shows a default MOR GUI for the explicit request functio­nality (with the default language set in Slovene). It allows for changing the default language and provi­sioning of the needed evidence. For each evidence it provides for: 
• Review of the schema and data fields that consti­h1te evidence. The review is available in multiple languages and provides a list of all labels and de­scriptions of the fields (see Figure 7). 
• Providing the evidence to the DE. The user can upload each evidence if he owns a relevant and verifiable digital document accepted by the DE (e.g., XML, PDF). The DE can automatically retri­eve evidence if the user selects the country that takes the role of a DP. The evidence retrieval pro­cess follows the predefined exchange patterns (described in the previous section). 
The selection made by the user per canonical evi­dence type is rehrrned by filling in the JavaScript Map variable specified. The keys in the map variable 
III Sloven šcina (si) • Izberi O državo ponudnika ali O naloži dokazilo 
~ 

• --•• WMMI !: Španija (ES) • O Izberi pošiljatelja 
Izberi O državo ponudnika ali O naloži dokazilo 


4$.j..,J.j.@j 

• •• 11 1 11•• Izberi državo• 
= Avstrija (AT) 
11 Belgija (BE) 
• Bolgarija (BG) :Z: Hrvaška (HR) 
~ Ciper (CV) llii. Ceška (CZ) 
Figure 6: Delault MOR component GUI lor the explicit request lunctionality, 
Visualizar Evidencia de nascimento esquema dentro a portugues (pt) y com 0 descric;:oes 
Evid8ncia de nascimento t-Documento oficial ou dados provando o nascimento de uma crian<;a. 
ldentificador <-Uma refer~ncia inequ/voca as evid~ncias de nascimento. Contelldo <-Conteudo -String Alfa-numerica Tipo de identificador +-O tipo de identificador emitido (se a autoridade emitir mais de um tipo de identificador). Data de em issao t-A data em que foi emWda Autoridade emissora t-O nome da agenda emissora 1D da autoridade emissora t-O ID da agenda emissora 
Data de emissao +-A data mais recente em que a instancia da evidencia de nascimento foi emitida. 
Autoridade emissora +-Uma entidade pUblica com autoridade oficial encarregada de cmitir a evidencia de nasdmento. ldentificador +. Muitas org::mizat;0cs s5o rcferid:Js por um Dcr0nimo ou algum outro idcntificador. Por exemplo, entro as instituir;0es da UE, o BCE 6 o identificador do Banco Central Europeu, o OLAF do Organismo Europeu AntiN/AFraud e assim por diante. Estes sao formalmente reconhecidos pela Comissao Europeia, que fornece uma lista dessas siglas. Listas anfllogas devem ser usadas em outros contextos. 
Contelldo <-Conteudo -String Alfa-numerica Tipo de identificador <-O tipo de identificador emitido /se a autoridade emitir mais de um tipo de identificador). Data de emissiio <-A data em que toi emitida Autoridade emissora E--o nome da agencia emissora 
Figure 7: Selected evidence type preuiew within a default MOR GUI. 
are the needed evidence type names. Each entry con­sists of one from two possible values depending on the user choice, that is, (a) the selected provider co­unh·y or (b) the ASCII representation of the uploaded file, which can be either an ASCII file such as an XML file or a binary file such as a PDF. 
Evidence Preview 
After issuing the requested canonical evidence, the MOR component GUI for the evidence preview is opened by a Web page at the DE or the DO side, depending on the type of the exchange pattem ( de­scribed in the previous section). 
When the page with the embedded MOR compo­nent is open, the MOR API is called to retrieve the MOR JSON in the given language (if not already ca­ched by the browser). Figure 8 shows a default MOR GUI for the evidence preview functionality. The parameters for embedding the preview part of the component include (a) the default component lan­guage, (b) a JSON array that contains JSON objects of specific evidence (key is the evidence type name and value is its representation), and (c) a JavaScript Map variable to fill by the MOR component with the result of the evidence preview. For instance: 
ai't English (en) y  ioj:@j§,1:HI  O include or  I exclude  
~ Mamage evidence  -.J lnclude or  O exclude  
IMHIH  


<app-mor-p 
defaultLang = "en" 
postActionValue = " [ {-.}, { ... }]" 
output JSMapi d = "outputJSMapidMorEr" 


></app-mor-p> 
The component allows for reviewing the retrie­ved evidence types from the DO. If the evidence is not provided in machine-readable XML fom1at fol­lowing semantic schemas, the evidence is available asa binary file (e.g., PDF, image). l11e preview of evi­dence is like the explicit request part (Figure 7) with the addition of actual values for the user's evidence. Users can exclude the evidence at tlus step, and wi­thin the process, the system will ask them again to provide the needed canorucal evidence (i.e., explicit request part). 
The selection made by the user per canorucal evidence type is rehrrned by filling the defined Ja­vaScript Map variable. TI1e key to each entry is an evidence type name, wlule the value is of Boolean type, depending on the user's cl1oice to include the evidence within the service. 
3.4 Reusability evaluation 
The MOR component building blocks (i.e., explicit request and evidence preview) have the advantage of being reusable and generic for any complex tem1 and 
language, so parties in the evidence exchange do not 

Figure 8: Oefault MOR component GUI for the evidence preview functionality, need to develop their equivalent components, and any 
modification in MOR is automatically available thro­ugh the MOR building blocks. For the best integration from the user experience perspective, CSSs are used. More specifically, we define the following aspects of adapting the component and its reusability: 
• 
Configuration and integration: The component is packaged as a self-contained HTML DOM object with JavaScript functionality. It can be included on any modem Web page. TI1e parameters for the component are set at its placement into an HTML DOM by providing values for its attributes. TI1e results of the component are outputted to a selec­ted JavaScript variable. 

• 
Support of additional evidence types: TI1e com­ponent can support an arbitrary evidence type and an arbitrary number. Tilis is enabled using se­mantic schemas and MOR terms JSON provided by the DE4A services. If the services update JSON responses, the MOR component wi.11 automatical­ly consider them and adapt the preview. 

• 
Multilanguage support: Multiple languages su­pport is enabled by MOR terms JSON files that are offered by the DE4A services. This JSON also includes translations for MOR component-related translations (e.g., names of buttons) wllich start with the root MOR term "GUI." For example, a 


~ '"" c~•J. ~ ~: E " 
Pladenj Preview 
Preview 
tut('•§f>uMk"..,1i.-.r,, .,,.,.,f>.., ,-,1f<,:,;;,-,,u,-,,r,i,11i.,-,,1 ty,l.!:M1lr,,'S',ff 
HOLDER Of ACJRVa1ENT 
label of a proceed button in a selected language is available in the "GUI/proceed" field. 

• Visual integration: One of the essential require­ments was to support fully custonlizable visual integration of the component. There can be many DEs or DPs from different countries, and they al­ready have their Web sites supporting digital pu­blic services. Therefore, the need is that the MOR component can be adapted for each one of those portals. h1 that sense, ali the HTML DOM com­ponents in MOR have an associated id, based on whlch visual characteristics can be defined via the MOR CSS during the integration. Furthermore, we would like to mention the Slove­
nian national "Once-Only" platform "Tray", whlch operates asa decentralized data exchange system su­pporting the implementation of the Government to Citizens (G2C), Government to Businesses (G2B) and Government to Government (G2G) e-procedures. TI1e system can also obtain data from many sources through a single-entry point, exchanging a mitlimum data set. TI1e central system, "The tray", takes care of optimal data load from ali the data sources with the help of the control and forecastit1g system based on continuous machine learnit1g. Based on thls tech­nology, a prototype for Slovenian evidence preview 
An.l.'1.l.NII 
"' ""T""'ocr,:-:;::, 
At;,~~"'rr r .... l),,n, ~ 

Figure 9: Evidence preview example by the Slovenian implementation of a pilot supporting digital puhlic services related to "Studying abroad"-Pladenj CTrayl Preuiew. 
has been developed in the DE4A project as one of the possible solutions. Figure 9 shows an example of evi­dence preview by the Slovenian implementation of a pilot supporting digital public services related to "Sh1dying abroad" (using Tray for data exchange). MOR component can be adapted to allow the same look and feel (in comparison to the default MOR evi­dence type preview setting, shown in Figure 7). 
MOR component m:ight therefore be used by coun­tries that do not yet implement such flmctionalities. Wi­th:in the DE4A project, these countries might be Spain, Portugal or Romania, which take part in the pilots. 


4 CONCLUSIDNS 
In this paper, we presented the main characteristics of the DE4A project aiming to support the delivery of cross-border digital public services. We showed the evolution of the development of reusable prototype components in software engineering and how it evol­ved for the needs of public administration systems ba­sed on OOTS princi.ples. We b1ieily reviewed progress in cross-border and cross-sector public admin:istrati­on interoperability. We focused on specifics of D:igital Europe for ali projects, especialiy those related to the use case of a reusable prototype component. In more detail, we presented a use case for building a semantic and Web-based reusable prototype component. The pilot component Multilingual Ontology Repository is a deliverable of the Digital Europe for ali project. We evaluated and showed how the component offers adaptability to any Website and is configurable to su­pport fuhue semantic models. 
We believe that using such prototypes and buil­ding blocks could support the EU Member States' pu­blic adrn:inistrations' interoperability and would open new opporhm:ities to support mobility, fasten proces­ses and be very cost-effective. Also, innovative tech­nologies will be able to serve with transfonnative im ­pact, for example, blockchain (for effective notariza­tion, fostering accotmtability and transparency in the distributed transactional environrnent) and rnachine learning ( o ver usage data to automate mon:itoring and irnprove effectiveness/quality of eProcedures). 

5 ACKNDWLEDGEMENTS 
The research and development reported in this paper have received ftmding from the Etuopean Union's Horizon 2020 Research and Innovation Prograrnme under grant agreement no. 87035 (DE4A: Digital Eu­rope for Ali). We would like to thank the rnernbers of the DE4A WP3 group, especialiy the WP leader, Thashmee Karunaratne, and Ana Rosa Guzrnan. We would also like to give credit to the authors of the DE4A deliverable D2.1, from where we used Figure 2 (ATOS is the leading partner of the DE4A project). 


6 LITERATURE 
(1) 
Sametinger, J. (1997). Software engineering with reusable components. Springer Science & Business Media. 

(2) 
Kozaczynski, W., & Booch, G. (1998). Component-based software engineering. IEEE Software, 15(5), 34-36. doi:https:// doi.org/10.1109/MS.1998. 714621. 

(3) 
Mcllroy, M. D., Buxton, J., Naur, P., & Randell, B. (1968, Oc­tober). Mass-produced software components. In Proceedin­gs of the 1 st intemationa/ conference on sottware engineering, 

Garmisch Pattenkirchen, Germany (pp. 88-98). 

(4) 
Wallace, B. (201 O). A hole for every component, and every component in its hole. Existential Programming. 

(5) 
Gackenheimer, C., & Paul, A. (2015). lntroduction to React (Vol. 52). Apress. 

(6) 
Narducci, F., Palmonari, M., & Semeraro, G. (2013). Cross­-language semantic matching for discovering links to e-gov services in the lod cloud. KNOW@ LOD, 992, 21 -32. 

(7) 
GovStack (2021). Accelerating the digital transformation of govemment services. https://www.govstack.global. (Acces­sed: July 13, 2022). 

(8) 
Castano, S., Antonellis, V. D., & Pernici, B. (1995, August). Building reusable components in the public administration domain. In Proceedings of the 1995 Symposium on Software reusability (pp. 81-87). 

(9) 
Peristeras, V., Tarabanis, K., & Goudos, S. K. (2009). Model­-driven eGovernment interoperability: A review of the state of the art. Computer Standards & lnterfaces, 31(4), 613-628. 

(1 OJ Peristeras, V., Loutas, N., & Tarabanis, K. (2008, March). Or­ganizational engineering in public administration: the state of the art on eGovernment domain modeling. In Proceedings of the 2008 ACM symposium on Applied computing (pp. 580­587). 

(11) 
Digital Europe for AII, https://www.de4a.eu/ (Accessed: July 21, 2022) 



Slavko Žitnik is an assistant professor at the Faculty of Computer and lnformation Science, University of Ljubljana. His main research interests are semantic Web and natural language processing, specifically text mining, information retrieval and information extraction. He is in active collaboration with the industry on his research topics and has active collaborations with Harvard University, University of South Florida, and University of Belgrade. 
Karmen Kern Pipan is the Head of Data Governance Department, Dffice for Development of Digital Solutions, lnformation Technology Directorate at the Ministry of Public Administration of Republic Slovenia. She is involved in data management, business intelligence, big data analytics, semantics and interoperability solutions to improve data-dr iven decision-making and user experience in public administration. She led an inter­-ministerial task force to prepare the Public Administration Development Strategy 2015-2020. 
Miha Jesenko has a legal-based background with Degree in Law, upgraded by Masters in lnformation Technology and Law. He works in Data Govemance Department, Office far Development of Digital Solutions, lnformation Technology Directorate at the Ministry of Public Administration of Republic Slovenia. His main areas of work relate to data governance and interoperability, especially semantic interoperability through work on ontologies, controlled vocabularies, and reference data. An important part of his work is cooperation in efforts to introduce new technologies used by public administration. 
Dejan lavbic is an associate professor at the Faculty of Computer and lnformation Science, University of Ljubljana. His research areas include lnformation Quality, Semantic Web and Web3, distr ibuted ledger technologies and self-sovereign identity. He collaborated in numerous EU and national-funded projects and has a long track of collaborations with the industry