D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy

Towards a sustainable Open Data ECOsystem D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy This project has received funding from the European Unionʼs Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 955569. The opinions expressed in this document reflect only the authorʼs view and in no way reflect the European Commissionʼs opinions. The European Commission is not responsible for any use that may be made of the information it contains. D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy Project Acronym ODECO Project Title Towards a sustainable Open Data ECOsystem Grant Agreement No. 955569 Start date of Project 01-10-2021 Duration of the Project 48 months Deliverable Number D4.3 Deliverable Title An approach to steer the behaviour of nongovernment data holders towards open data through a governance strategy Dissemination Level Public Deliverable Leader Katholieke Universiteit Leuven (KULEUVEN) Submission Date 30-09-2024 Author(s) Joep Crompvoets – KULEUVEN – Caterina Santoro KULEUVEN, Ahmad Ashraf Ahmad Shaharudin – TU Delft, Davide Di Staso - TU Delft, Giorgos Papageorgiou – Farosnet S.A., Alejandra Celis Vargas - AAU, Liubov Pilshchikova - TU Delft, Héctor Ochoa Ortiz - Unicam Document history Version # Date Description (Section, page number) Author & Organisation 17-04-2024 Kick-off Meeting Joep Crompvoets - KULEUVEN 13-05-2024 Second Meeting and brainstorming session with Miro Board Joep Crompvoets, Caterina Santoro – KULEUVEN, Ahmad Ashraf Ahmad Shaharudin – TU Delft, Davide Di Staso - TU Delft, Giorgos Papageorgiou – Farosnet S.A., Alejandra Celis Vargas - AAU, Liubov Pilshchikova - TU Delft, Héctor Ochoa Ortiz - Unicam V0.1 29-05-2024 First draft Caterina Santoro - KULEUVEN V0.2 26-06-2024 Input to Section 3-8 Ahmad Ashraf Ahmad Shaharudin – TU Delft, Davide Di Staso - TU Delft, Giorgos Papageorgiou – Farosnet S.A., Alejandra Celis Vargas - AAU, Liubov Pilshchikova - TU Delft, Héctor Ochoa Ortiz - Unicam 28-06-2024 Third Meeting based on V0.2 and discussion Joep Crompvoets, Caterina Santoro – KULEUVEN, Ahmad Ashraf Ahmad Shaharudin – TU Delft, Davide Di Staso - TU Delft, Giorgos Papageorgiou – Farosnet S.A., Alejandra Celis Vargas - AAU, Liubov Pilshchikova - TU Delft, Héctor Ochoa Ortiz - Unicam D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy Version # Date Description (Section, page number) Author & Organisation V0.3 22-07-2024 Input to Section 3-8 Ahmad Ashraf Ahmad Shaharudin – TU Delft, Davide Di Staso - TU Delft, Giorgos Papageorgiou – Farosnet S.A., Alejandra Celis Vargas - AAU, Liubov Pilshchikova - TU Delft, Héctor Ochoa Ortiz - Unicam V0.3 26-7-2024 Review Sections 3-8 + add text of Section 2 Background + methodology Joep Crompvoets V0.4 30-7-2024 Add text Section 1, Review Section 2, Review Sections 3-8, Add text of Section 9 and 10 Caterina Santoro V0.5 31-07-2024 Revision of V0.4 Ahmad Ashraf Ahmad Shaharudin – TU Delft, Davide Di Staso - TU Delft, Giorgos Papageorgiou – Farosnet S.A., Alejandra Celis Vargas - AAU, Liubov Pilshchikova - TU Delft, Héctor Ochoa Ortiz - Unicam V0.6 02-09-2024 Revision of V0.5 Anneke Zuiderwijk, Mélanie Dulong de Rosnay , and Bastiaan van Loenen V0.7 13-09-2024 Add executive summary, highlights, research methodology (in details), add conclusion in each section, review the introduction, problem statement, and conclusion Joep Crompvoets, Caterina Santoro – KULEUVEN, Ahmad Ashraf Ahmad Shaharudin – TU Delft, Davide Di Staso - TU Delft, Giorgos Papageorgiou – Farosnet S.A., Alejandra Celis Vargas - AAU, Liubov Pilshchikova - TU Delft, Héctor Ochoa Ortiz - Unicam V0.8 16-09-2024 Revision of V0.7 Anneke Zuiderwijk, Mélanie Dulong de Rosnay, and Bastiaan van Loenen V0.9 23-09-2024 Review the executive summary, the highlights, and the research methodology Joep Crompvoets, Caterina Santoro – KULEUVEN, Ahmad Ashraf Ahmad Shaharudin – TU Delft, Davide Di Staso - TU Delft, Giorgos Papageorgiou – Farosnet S.A., Alejandra Celis Vargas - AAU, Liubov Pilshchikova - TU Delft, Héctor Ochoa Ortiz - Unicam V0.10 26-09-2024 Approval Bastiaan van Loenen, TUD V1.0 30-09-2024 Final editing Danitsja van Heusden, TUD D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy Table of Contents Abbreviations ..................................................................................................................................................................... 7 Executive summary .......................................................................................................................................................... 8 Highlights ............................................................................................................................................................................ 9 1 Introduction .............................................................................................................................................................10 1.1 Problem definition .......................................................................................................................................10 1.2 Role of this deliverable in the ODECO project .................................................................................11 1.3 Outline ..............................................................................................................................................................11 2 Theoretical framework ........................................................................................................................................12 3 Research methodology .......................................................................................................................................17 4 As-is / to-be non-specialist actors .................................................................................................................21 4.1 Introduction ...................................................................................................................................................21 4.2 As-is: Non-specialist actors ......................................................................................................................21 4.3 To-be: Non-specialist actors ....................................................................................................................23 4.4 Conclusion ......................................................................................................................................................25 5 As-is/to-be elementary schools ......................................................................................................................26 5.1 Introduction ...................................................................................................................................................26 5.2 As-is: elementary school students ........................................................................................................26 5.3 To-be: elementary school students ......................................................................................................27 5.4 Conclusion ......................................................................................................................................................30 6 As-is / to-be Non-Profit Organisations (NPO)...........................................................................................31 6.1 Introduction ...................................................................................................................................................31 6.2 As-is: non-governmental intermediaries ............................................................................................31 6.3 To-be: non-governmental intermediaries ..........................................................................................35 6.4 Conclusion ......................................................................................................................................................36 7 As-is / to-be Journalists ......................................................................................................................................37 7.1 Introduction ...................................................................................................................................................37 7.2 As-is: Journalists ...........................................................................................................................................37 7.3 To-be: Journalists .........................................................................................................................................41 7.4 Conclusion ......................................................................................................................................................43 8 As-is / to-be commercial organisations .......................................................................................................44 8.1 Introduction ...................................................................................................................................................44 8.2 As-is: Commercial organisations............................................................................................................44 8.3 To-be: Commercial organisations ..........................................................................................................51 D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 8.4 Conclusion ......................................................................................................................................................52 9 As-is/to-be open data intermediaries ...........................................................................................................54 9.1 Introduction ...................................................................................................................................................54 9.2 As-is: Open data intermediaries .............................................................................................................54 9.3 To-be: Open data intermediaries ...........................................................................................................59 9.4 Conclusion ......................................................................................................................................................63 10 Summary of the results ..................................................................................................................................64 10.1 As-is situation in the open data ecosystem ..................................................................................64 10.2 To be situation in the open data ecosystem ................................................................................66 11 Discussion and Conclusion ...........................................................................................................................69 References .........................................................................................................................................................................71 List of figures Figure 1: The overall research methodology .......................................................................................................20 List of tables Table 1: Classification of governance instruments into structural and managerial instruments ....13 Table 2: Clusters of governance instruments strongly based on work of Verhoest and Bouckaert (2005) ..................................................................................................................................................................................15 Table 3: Matrix Enablers – Governance instruments .........................................................................................17 Table 4: Current governance instruments stimulating non-specialist actors to participate in open data hackathon ecosystems .......................................................................................................................................22 Table 5: Desirable governance instruments stimulating open data non-specialists actors to participate in the open data hackathon ecosystem .........................................................................................24 Table 6: Existing governance instruments stimulating elementary school students to participate in the open data ecosystem and share open data .................................................................................................27 Table 7: Potential governance instruments stimulating elementary school students to participate in the open data ecosystem and share open data ............................................................................................29 Table 8: Existing governance instruments stimulating NPOs to participate in the open data ecosystem and share open data ...............................................................................................................................33 Table 9: Potential governance instruments stimulating NPOs to participate in the open data ecosystem and share open data ...............................................................................................................................35 Table 10: Current governance instruments stimulating journalists users to participate in open data hackathon ecosystems .................................................................................................................................................39 Table 11: Potential governance instruments stimulating journalists to participate in the open data ecosystem and share open data ...............................................................................................................................42 Table 12: Existing governance instruments stimulating commercial organisations to participate in the open data ecosystem and share open data .................................................................................................47 Table 13: Potential governance instruments stimulating commercial organisations to participate in the open data ecosystem and share open data .................................................................................................52 Table 14: Existing governance instruments stimulating open data intermediaries to participate in the open data ecosystem and share open data .................................................................................................57 D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy Table 15: Potential governance instruments stimulating open data intermediaries to participate in the open data ecosystem and share open data (in addition to existing instruments) .......................60 Table 16: Summary of the results .............................................................................................................................64 D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 7 Abbreviations AND Automo�ve Naviga�on Data B2G Business to Government CGA Children's General Assembly C2G Citizens-to Government D Deliverable ESR Early Stage Researcher G2G Government-to-Government G2B Government-to Business G2C Government-to Community or Citizen M Managerial MS Milestone NGD Non-Government Data NPO Non-Profit Organizations OD Open Data ODECO Open Data ECOsystem OKB Open Knowledge Belgium OKF Open Knowledge Foundation OKFG Open Knowledge Foundation Germany OSM OpenStreetMap OSMF OpenStreetMap Foundation S Structural WP Work Package Nr Partner Partner short name Country Beneficiary 1 Technische Universiteit Delft TU Delft Netherlands 2 Katholieke Universiteit Leuven KUL Belgium 3 Centre National de la Recherche Scientifique CNRS France 4 Universidad de Zaragoza UNIZAR Spain 5 Panepistimio Aigaiou UAEGEAN Greece 6 Aalborg Universitet AAU Denmark 7 Università degli Studi di Camerino UNICAM Italy 8 Farosnet S.A. FAROSNET S.A. Greece Partner organisations 1 7eData 7EDATA Spain 2 Digitaal Vlaanderen DV Belgium 3 City of Copenhagen COP Denmark 4 City of Rotterdam RDAM Netherlands 5 CoC Playful Minds CoC Denmark 6 Derilinx DERI Ireland 7 ESRI ESRI Netherlands 8 Maggioli S.p.A MAG Italy 9 National Centre of Geographic Information CNIG Spain 10 Open Knowledge Belgium OKB Belgium 11 SWECO SWECO Netherlands 12 The government lab GLAB United States of America 13 Agency for Data Supply and Infrastructure ADSI Denmark 14 GFOSS Open Technologies Alliance GFOSS Greece 15 Inno3 Consulting IC France 16 Regione Marche RM Italy 17 Open Data Institute OCI United Kingdom D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 8 Executive summary In this research document, we explore governance stimuli that can enhance non-governmental open data sharing. Previous investigations have shown that many non-governmental actors are not sharing open data, which is problematic as the open data ecosystem depends on the participation of diverse contributors. The lack of inclusion of different perspectives in open data ecosystems restricts the possibilities of capturing the full potential of such ecosystems, and it is suspected to add to existing power imbalances at the expenses of vulnerable groups. For our analysis, we build on earlier findings that identified key enablers that can promote nongovernmental open data sharing. These enablers include training in data skills and literacy, access to suitable technical tools, alignment of private interests with open data sharing, adequate resources (financial, time, personnel), the presence of data-sharing communities, awareness of the social impact of open data, and the availability of engagement or enjoyment activities. To understand how governance mechanisms can foster open data sharing, we apply the hierarchy, market and network governance framework. We employed a multi-stepped research strategy, analysing perspectives from diverse actors such as non-specialists, elementary schools, non-profit organizations, journalists, commercial entities, and intermediaries through various methodologies including case studies, literature reviews, action research, and design-based research. This approach allowed us to identify governance instruments that are currently 1) enhancing nongovernmental data sharing, to which we refer as the ‘as isʼ situation or 2) have the potential to foster non-governmental data sharing, to which we refer as the ‘to beʼ situation. Our findings reveal that a wide range of governance instruments—from hierarchical measures like legal frameworks to grassroots network initiatives such as collaborative partnerships—can effectively stimulate open data sharing. Notable strategies include coordination through permanent bodies and the creation of consortia, which offer substantial benefits for various nongovernmental actors. However, market-oriented instruments, such as financial incentives, are currently underutilized, likely due to the non-commercial nature of many non-governmental actors we analysed in this study. Our results go beyond non-governmental open data sharing and highlight also how some governance instruments can facilitate other forms of participation from non-governmental actors, such as open data use in combination with open data sharing. This study establishes a solid foundation for understanding governance mechanisms in nongovernmental open data sharing and highlights areas for further research. While the study primarily focuses on specific actors and regions within the EU, extending these findings to a broader range of contexts could provide a more comprehensive view of the open data ecosystem. Additionally, further research could focus specifically on how to include the perspectives of vulnerable groups through non-governmental open data sharing. D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 9 Highlights ⇒ Different governance instruments can lead to further open data sharing by nongovernmental actors. Some instruments are already adopted and require further uptake, while others are not yet adopted and can improve open data sharing. ⇒ There is consensus among different actor groups on the potential of coordination through the creation of permanent bodies organizing, for instance, recurrent hackathons with longlasting effects or through the establishment of national or supra-national consortium for open data sharing by non-governmental organisations. ⇒ Legal Frameworks for Non-Governmental Open Data Sharing: The study suggests adopting or extending existing legal frameworks for stimulating non-governmental open data sharing. ⇒ Underutilized and under conceptualized Market Incentives: Financial incentives, such as tax credits, are underused and under conceptualized as viable strategies in stimulating nongovernmental data sharing, especially among non-profit and non-commercial actors, signaling the need for alternative or supplementary governance strategies. ⇒ Some governance instruments can also be used to stimulate other kinds of value to the open data ecosystem, such as the inclusion of actors by facilitating open data use along open data sharing. ⇒ Further research is needed to confirm and expand the results of our analysis to other perspectives within the same actor group, across different sub-fields (e.g., commercial organizations operating in various industries), scales (e.g., non-profit organizations of different sizes), and contexts (e.g., including those lacking the capabilities to engage with technology). D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 16 Instruments Hierarchy Market Network M2. Financial management: inputoriented performanceoriented up working and cooperation M5. Interorganizational culture and knowledge management M6. Capacity building D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 17 3 Research methodology In the previous section, we reviewed the theoretical background on governance instruments and their application in the open government context. To better understand how to encourage nongovernmental data holders to share open data, it is essential to assess the current use of governance instruments (the as-is situation) and their potential for promoting further open data sharing (the to-be situation). Ideally, implementing governance instruments in the to-be scenario would enhance non-governmental data holders' willingness to share data. To derive a comprehensive list of relevant governance instruments, we applied a multi-step research methodology summarized at the end of this section (Figure 1). We began by identifying enablers of non-governmental open data sharing from our previous work (Deliverable 4.1). These enablers include training in data skills and literacy, access to appropriate technical tools, alignment of private value and interests with open data sharing, sufficient resources (financial, time, people), the existence of data-sharing communities, awareness of the social impact of open data sharing, and the presence of engagement or enjoyment activities. Next, we conducted two online workshops attended by two researchers whose main expertise was in governance, and six researchers who have been studying different actorsʼ perspectives (i.e., non-specialist actors, elementary schools, NPOs, commercial organisations, and intermediaries) for more than one year. In the first online workshop (Workshop 1), we discussed the theoretical framework (i.e., hierarchy, network, and market governance through structural and managerial instruments) and collectively reflected on how different governance instruments can stimulate non-governmental open data sharing in both the as-is and to-be situations, achieving the goals set by the enablers. As a result of the first workshop, we developed a matrix of enablers and governance instruments (see Table 3), and we assessed the feasibility of analysing the different actorsʼ perspectives through a governance lens. Following the workshop, each researcher identified relevant governance instruments from various actor perspectives, including nonspecialists, elementary schools, not-for-profit organizations, journalists, commercial organizations, and open data intermediaries. Table 3: Matrix Enablers – Governance instruments Below, we outline the research methodology employed for each actor's perspective to develop our analysis. At the end of this section, we explain how these perspectives have been integrated into the results (Figure 1). D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 18 Research methodology applied for non-specialist actors The section on non-specialist actors is based on a review of existing literature on open data events, conducted in Di Staso, Mulder, et al (In press). We systematically searched for articles describing open data events on major academic databases (Scopus, Web of Science, ACM). The search terms included “open data” combined with “hackathon,” “game jam,” “design jam,” as well as synonyms for these terms. We only included English language journal and conference papers, describing in person open data hackathons or open data game jams, of least one day and no more than three days in duration (Di Staso, Mulder, et al., In press). We consolidated the search results into a single list of unique records, then filtered based on scanning the abstract or full text, and finally filtered based on the full text (Di Staso, Mulder, et al., In press). After all these steps, we obtained 20 unique articles. This section is also based on the following sources: • participant observation of open data hackathons in Europe, and one of its final reports (Rambøll Management Consulting, 2022). These include: the Nordic AI and Open Data Hackathon, which took place in March 2022 in Denmark, Sweden, and online, aimed at reusing open datasets of Nordic countries with AI; and the 3rd CASSINI Hackathon, which took place in May 2022 in several European countries, and was aimed at reusing space data (Copernicus, Galileo, etc.) to address challenges related to tourism and travel • lessons learnt from organizing three open data game jams with non-specialist participants, such as Masterʼs students, and non-specialist civil servants. A total of 101 participants attended the jams, which lasted approximately 8 hours. Participants were asked to express societal issues with open data and game-making tools. The events are described in (Di Staso, Christiansen, et al., In press). Research methodology applied for elementary schools A design-based research (DBR) methodological framework has been applied in the case of elementary schools. DBR is defined as a theoretical and practical approach for the development of new educational approaches (Bakker, 2018). Iterative cycles are developed, aiming at producing actionable knowledge that can be used to achieve some educational goal through design (Anderson & Shattuck, 2012). Each DBR cycle is a design experiment that develops in four phases: problem definition, design, intervention, and analysis (Anderson & Shattuck, 2012). This iterative process aims at having better and more concrete outcomes after each iteration. Three cycles including five interventions in Danish schools were conducted with the total participation of 117 pupils aged 14 to 16 years and nine teachers in 7th to 9th grade. The first cycle included a systematic mapping review of the skills associated to using open data in education and the learning approaches (Celis Vargas et al., 2023), semi-structured interviews with five teachers and an open data activity with 39 9th grade students to understand their current practices (Vargas et al., 2024). This first cycle was essential to define the as-is situation by exploring the current state of open data initiatives in education and the current practices of teachers and students to teach and learn open data competencies such as the analysis of data and the creation of data arguments to solve a real-world problem (Celis Vargas et al., 2023; Vargas et al., 2024). The second and third cycles focused on the systematic development of an authentic game called The Open Data Newsroom (Celis Vargas et al., 2024). The game is based on authentic open data practices to develop Data Literacy and Real-world problem-solving competencies (Celis Vargas et al., 2024). The Open Data Newsroom is a role-playing game where students play as data journalists to unravel an environmental mystery using and analysing open data. Four interventions in Danish schools were conducted with the participation of 78 students in 7th to 9th grade. The discoveries and outcomes of the second and third cycles help to define the to-be situation. D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 19 Research methodology applied for journalists The analysis of the governance of journalists as potential contributors consisted of three main components: First, a systematic literature review was conducted to identify key areas of focus regarding open data and journalism. Second, three semi-structured interviews were carried out with journalists and data analysts from small media organizations in the European Union that emphasize data journalism. Expert sampling was employed to select participants who are actively using data in their journalistic work, thus ensuring their expertise in the domain of data journalism. Specifically, one journalist from Eurologus in Belgium was interviewed online, along with one journalist and one data analyst from Divergent in Portugal, also interviewed online, and finally, the chief editor and a data analyst from Farosnet in Greece were interviewed in person. These interviews aimed to explore how journalists utilize open data and the challenges they face, guided by insights from the literature review. Lastly, ongoing action research is being conducted at Farosnet, where the researcher is embedded through the ODECO project, employing iterative cycles of planning, action, and reflection to identify and address specific needs and challenges in integrating open data into reporting. The collected data from the interviews and action research are being analysed through a qualitative content analysis approach, where audio recordings were reviewed to extract key concepts and themes using an inductive coding process, allowing for a comprehensive understanding of journalists' experiences and challenges in utilizing open data. Research methodology applied for NPOs The case studies of four NPOs were conducted: Open Knowledge Belgium (OKB), Open Knowledge Foundation Germany (OKFG), Open Knowledge Foundation (OKF) and CityLAB Berlin. These organisations are focused on openness of knowledge and data, and on innovation and digitalisation of public services. The selection criteria for the case studies were: 1) Non-profit organisations should have different missions/focuses/aims, 2) Each case should have more than one type of open data project, 3) The cases work on different levels, i.e., municipal/regional/national/international, and 4) The cases involve organisations and people willing and ready to cooperate in the research and share information required to conduct this research. Semi-structured one hour interviews were conducted online and in-person with employees of these four organisations. We interviewed thirteen employees who work on open datarelated projects within these NPOs. Additionally, we collected information from public web pages describing the open data projects of these NPOs. The deductive approach was used as collected data was analysed by using the codes based on the existing governance mechanisms theory, such as, instruments and enablers groups of codes. Research methodology applied for commercial organisations The analysis of commercial organisations focused on the case study of OpenStreetMap, a global crowdsourcing project for contributing and reusing open geospatial data where individual and organisational actors interact. Commercial organisations have willingly contributed value and data to the project since the early days of it, with a dynamic ecosystem flourishing around it. Hence it is a prime example to demonstrate the as-is and to-be governance instruments for commercial organisations. Data was collected through semi-structured interviews with employees of 25 companies (7 big corporations and 18 SMEs) who use and contribute to OpenStreetMap and through analysis of the OpenStreetMap Wiki and OpenStreetMap Foundation webpages. The companies in the Organised Editing Activities and Foundation Corporate members lists were contacted, and the final list was completed with companies met through OSM community events, and those proposed by the interviewees. Data analysis of the interviews was made with inductive reasoning, by extracting keywords, to find the common motivations and barriers to contribute open data. D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 20 Research methodology applied for intermediaries Insights on the existing (as-is) and potential (to-be) governance instruments to stimulate open data intermediaries to share open data are based on the case studies of Esri and OpenStreetMap (OSM). We conducted 53 in-depth interviews with representatives from these two organizations and other relevant stakeholders, such as open geospatial data providers and end-users. The interviewees are in charge in the managerial or technical aspects of open data. These two organizations were selected due to their significant contributions to the open geospatial data ecosystem in the last decade. We analysed the interviews through abductive approach, following the governance instruments in Table 1. Esri is a multinational geospatial software company that has long been an open data intermediary. It serves such a role in multiple ways, such as by collecting and pre-processing open data from various sources and offering the ready-to-use data in its software (called ArcGIS), providing consultation services to open data providers and users, and developing applications and visualizations based on open data. OSM is a geospatial data crowdsource platform. While the OSM Foundation (OSMF) provides leadership, OSM is run by the community who contribute, reuse, and build applications based on the open data on the platform. From the analysis of the different actorsʼ perspectives to the results After conducting an initial analysis from the perspectives of the various actors using the research methodology outlined in the previous sections (Draft analysis), we discussed the preliminary results in a second workshop (Workshop 2). The aim of the second workshop was to resolve any theoretical ambiguity that might have arisen from the application of the theoretical framework. In some cases, for instance, researchers were aware of the existence of mechanisms that could enhance non-governmental open data sharing; yet they could not clearly associate them with one of the governance strategies. Following this, the analysis from the actors' perspectives was finalized (Final analysis), leading to the derivation of key findings—namely, a list of governance instruments that currently enable or could potentially enable non-governmental open data sharing. These instruments are presented in Section 10 of this document. The following figure (Figure 1) summarizes the multi-step research approach adopted in this deliverable. Figure 1: The overall research methodology D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 21 4 As-is / to-be non-specialist actors 4.1 Introduction Non-specialist actors are those who do not possess specialised skills for data analysis, but who can still benefit from the reuse of open data. In recent years, governments and private entities have started engaging with non-specialist actors through open data hackathons (Purwanto et al., 2018), and other “accelerated design events” (Falk, 2022). At an open data hackathon, participants - including non-specialist actors - come together over a short period of time (usually 1-2 days) to create together solutions that make use of open datasets. As such, it is relevant to study how open data hackathons, as a system, can become more circular and inclusive. Different stakeholders come together to support the event, and new datasets are made available as open data both in the “pre-hack” and “post-hack” phases (Concilio et al., 2017). Throughout hackathons (and similar accelerated design events) participants can form new networks and learn to work together with open data. Non-specialist users can make an important contribution at these events, by sharing their contextual knowledge, and their “thick” (i.e. qualitative) data regarding issues that can be solved with open data. In this section, we go over the current (as-is) and desired (to-be) governance instruments that can stimulate non-specialist users (in the context of open data hackathons) to share their knowledge (data). 4.2 As-is: Non-specialist actors We summarised the current governance instruments stimulating non-specialist actors to participate in the open data hackathon ecosystem in table 4. They are based on a variety of mechanisms, which we discuss in this section. In the context of open data hackathons, the establishment of coordinating functions or entities (S1) is leveraged, as these events tend to form at least a temporary organising committee, dedicated to bringing together multiple stakeholders, from local organisations to local government and institutions, as well as private companies. Each stakeholder can contribute to the event by making technical tools available to participants (e.g. APIs offered by tech companies), open data (e.g. local government opening previously closed datasets), or with funding for the event in exchange for sponsorships. Open data hackathons also leverage systems for information exchange and sharing (S5). For example, IT companies often provide APIs and other technical tools for free to participants, in order to promote their product and introduce more people to their ecosystem. In terms of partnerships (S7), open data hackathons leverage this instrument by forging new relations among stakeholders both in their “pre-hack” and “post-hack” phases (Concilio et al., 2017). The “financial management: joined up working and cooperation” (M4) mechanism, is also leveraged in open data hackathons, as they commonly offer monetary rewards and post-event incubation (including facilitating business connections, offering shared workspaces and mentoring) to selected teams. Finally, the capacity building mechanism (M6) is adopted in two ways: (1) previous research has found that open data hackathons can do “community capacity building”, and (2) teambuilding activities and the overall “pressure-cooker” (Mulder & Kun, 2019) environment of these events creates an engaging atmosphere for participants. For example, as highlighted in the final report of the Nordic AI and Data Hackathon (Rambøll Management Consulting, 2022): “Participants demonstrated high levels of engagement across the three venues”. However, organisers also noted that in-person hackathons were more successful than online ones at “dynamic, interesting event.” D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 22 Table 4: Current governance instruments stimulating non-specialist actors to participate in open data hackathon ecosystems Instruments Enablers S1 S5 S7 M4 M6 Availability of training in data skills and literacy Availability of appropriate technical tools Provision of APIs and technical tools by organising partners Alignment of private value and interests with open data sharing New partnerships between stakeholders during the “pre-hack” and “post-hack” phases Availability of resources (financial, time, people/ workforce) Award funding and incubation resources to promising teams and products Community capacity building Existence of data sharing communities Living labs and hackathon organisational committees Awareness about the social impact of open data sharing Presence of engagement or enjoyment activities Teambuilding activities and “pressure-cooker” environment D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 23 4.3 To-be: Non-specialist actors We identified five structural governance instruments and one managerial instrument in an ideal, “to-be” scenario, which we summarized in table 4. Starting with the establishment of coordinating functions or entities (S1), governmental bodies at the EU level should consider establishing permanent hackathon organisation committees for open data hackathon. The EU Publications Office already maintains a public calendar of open data events (including hackathons), and, up until 2022, used to organise an annual EU Datathon. Given the benefits in terms of community building and skills development offered by hackathons, these efforts should continue and be further strengthened. Our second recommendation concerns the reshuffling division of competences (S2) . Organisers should adopt specific group formation strategies at the hackathon event, in order to ensure the cooperation of expert and non-specialist actors. A more flexible division of competences is needed to include non-specialist actors in decision-making processes concerning open data utilization, ensuring their insights and contextual knowledge are integrated into the final output. Our third recommendation is related to the establishment of a legal framework (S3). Open data hackathon participants should be invited to share their contribution and output under an open license. In the current state, hackathon contributions often have unspecified or proprietary licensing. Our fourth recommendation is on the systems for information exchange and sharing (S5). To facilitate the collaboration between expert and nonspecialist actors, all participants should be invited to use the same beginner friendly tools. In the current state, the use of advanced tools often excludes non-specialist actorsʼ contributions. Our final recommendation is about the financial management (M2). Organisers should aim at lowering financial barriers for non-specialist participants, such as offering grants or stipends to cover participation costs. Additionally, allocating funds to develop resources that specifically support non-specialist engagement, like educational materials or beginner-friendly datasets, can further stimulate their participation. D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 24 Table 5: Desirable governance instruments stimulating open data non-specialists actors to participate in the open data hackathon ecosystem Instruments Enablers S1 S2 S3 S5 M2 Availability of training in data skills and literacy Availability of appropriate technical tools Invite both expert and non-specialist participants to use the same beginner friendly tools Alignment of private value and interests with open data sharing Inviting hackathon participants to make their solutions available under an open license Availability of resources (financial, time, people/ workforce) Establishing a permanent coordination body for hackathon events, in connection to open data initiatives Lowering financial barriers for nonspecialist participants Existence of data sharing communities Awareness about the social impact of open data sharing D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 25 Instruments Enablers S1 S2 S3 S5 M2 Presence of engagement or enjoyment activities Use appropriate group formation strategies at open data hackathons 4.4 Conclusion Open data hackathons are moving away from meetups of exclusively specialist actors and are starting to include a broader audience. While they already leverage governance instruments such as the establishment of coordinating entities, as well as systems for information exchange, further steps should be taken to incentivize non-government data holders to share open data. This includes the creation of permanent open data hackathon organisation committees, the adoption of group formation strategies tailored for non-specialist actors, as well as lowering financial barriers for non-specialist participants. D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 32 also working together with other organisations under the non-profit foundation umbrella of Technology Foundation Berlin. This supports alignment of private value and interests with open data sharing. Another network instrument is financial management: joined up working and cooperation (M4). OKB is an umbrella organisation which supports open-data grassroots projects/communities by providing a legal structure and financial support. Thus, those communities can share the open data under the organisational umbrella while not having bureaucratic hurdles, which supports the availability of resources enabler . Inter-organisational culture and knowledge management (M5) is another network instrument. Openness is one of the values shared within the open data NPOs as part of the cultivated and shared organisational culture of these organisations, which supports the alignment of private value and interests with open data sharing. NPOs aim to create a social impact as part of the cultivated and shared organisational culture, which means they have an awareness about the social impact of open data sharing. Additionally, the organisational culture of NPOs is often less hierarchical, and employees can propose and take part in projects they enjoy, which means they have a presence of engagement or enjoyment activities to motivate open data sharing. The last network-based instrument is capacity building (M6). OKF and OKFG support other NPOs in gaining skills to publish or work with open data, for example, through providing training and advice. That helps with the availability of training in data skills and literacy to help NPOs publish the open data. D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 33 Table 8: Existing governance instruments stimulating NPOs to participate in the open data ecosystem and share open data Instruments Enablers S3 S5 S7 M4 M5 M6 Availability of training in data skills and literacy OKF created Open Data Commons and Open Definition OKF is working on a tool Open Data Editor that will allow users to work with data in a simplified way and publish their data easily OKFG works together with the “Code for …” community of volunteers. Volunteers provide technical expertise. OKF and OKFG support other NPOs in gaining skills to publish or work with open data Availability of appropriate technical tools Alignment of private value and interests with open data sharing OKFG is part of the F5 alliance of five NPOs around digital policy; CityLAB Berlin is working together with other organisations under the nonprofit foundation umbrella Openness is one of the values shared within the NPOs as part of the cultivated organisational culture Availability of resources (financial, time, people/ workforce) OKB is an umbrella organisation which supports open data grassroots projects/communities D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 34 Instruments Enablers S3 S5 S7 M4 M5 M6 by providing a legal structure and financial support Existence of datasharing communities OKFG helps open the data for the “Code for…” volunteering community Awareness about the social impact of open data sharing NPOs aim to create a social impact as part of the cultivated organisational culture and aims Presence of engagement or enjoyment activities The organisational culture of NPOs is less hierarchical, and employees can propose and take part in projects they enjoy D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 35 6.3 To-be: non-governmental intermediaries Table 9 shows the potential governance instruments to stimulate non-profit intermediaries to contribute open data. The network-based instrument of systems for information exchange and sharing (S5) can be used to create a common platform for NPOs to share open data, similar to open data government portals, and some NPOs voice interest in having such a portal. The existence of such a portal could help with the availability of appropriate technical tools and availability of resources that many NPOs lack and, thus, are unable to share their data as open data. Another network instrument is financial management: joined up working and cooperation (M4). Financial resources are often scarce, so to tackle the availability of resources to support the projects' execution and their long-term support, more financial cooperation is needed. There are existing examples of such cooperation, such as the OKF network, that can help other open knowledge organisations financially. Such a solution should be promoted, especially across the countries, to help distribute the resources to the NPOs with less support. The capacity building (M6) network instrument can also be used to ensure the availability of training in data skills and literacy and the awareness about the social impact of open data sharing of the NPOs. Some non-profit intermediaries that have valuable data do not have the needed technical skills to publish open data or are not aware of open data's potential impact. To help deal with that, NPOs or governmental organisations can promote training or workshops. Table 9: Potential governance instruments stimulating NPOs to participate in the open data ecosystem and share open data Instruments Enablers S5 M4 M6 Availability of training in data skills and literacy Promote training to improve technical skills for NPOs that lack them Availability of appropriate technical tools NPOs voice interest in having a common platform for open data sharing Alignment of private value and interests with open data sharing Availability of resources (financial, time, people/ workforce) NPOs voice interest in having a common platform for open data sharing Financial resources are often scarce, so to support projects long-term, more financial cooperation is needed Existence of datasharing communities D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 36 Instruments Enablers S5 M4 M6 Awareness about the social impact of open data sharing Promote training to improve open data awareness for NPOs that lack it Presence of engagement or enjoyment activities 6.4 Conclusion As-is situation shows that NPOsʼ are motivated to share due to mainly network governance mechanisms, with one hierarchical instrument and without any market-based instruments. Thereʼs mutual support between/within this type of organisations, together with similar mission or focus on openness, which motivates them and provides the space for them to share open data. The definition of being a non-profit or non-governmental organisation explains why network mechanisms are preferred, and no market-based ones are utilised, as the motivations do not lie in the profit-making lane but focused on the societal issues. To-be potential for open data sharing, similarly, shows that NPOs will benefit from network mechanisms as in as-is situation. Specifically, building an infrastructure in the form of NPOs shared open data portal together with receiving financial resources for projects like that, would improve the organisationsʼ ability to share their data even more effectively. It is important to note, this researchʼ case studies are with smaller sized non-profit organisations, which might present a limited set of results. Further research might explore NPOs of a bigger size as different results might be achieved due to bigger organisations having more resources, mainly financial ones. D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 37 7 As-is / to-be Journalists 7.1 Introduction Journalists investigate, collect, and present information as news stories to the public through various channels and formats. They play a crucial role in keeping the public informed and holding power to account (Kovach & Rosenstiel, 2007). The use of open data in journalism, or data journalism, involves using data to uncover, analyse, and craft compelling stories by merging traditional journalism techniques with data analysis and visualization. Although data journalism has been practiced since the 19th century, its popularity has increased in recent years due to the abundance of information available to journalists (Gray et al., 2012). In the past, the main struggle was to collect and compile data sets; now, the focus has shifted to data analysis due to the plethora of data available from various sources (Rogers, 2013). While open data has played a role in this shift and can be a valuable resource for journalists, the openness of the data is not their primary concern. Journalists are mostly focused on the accuracy, relevance, and usefulness of the data, and the impact they can achieve by using it in their stories, regardless of whether the data is publicly accessible or obtained through other means (Bradshaw & Rohumaa, 2013). Journalists are mostly users of open data. The most common use is republishing existing visualizations of data to inform their audience. However, they also use raw open data, analysing it and publishing their findings to the audience through various types of visualizations (tables, figures, maps, etc.) to support their stories. There are also cases where media organizations release datasets as open data, but these are exceptions and more common in large media organizations (Coddington, 2015) The research methodology consists of three main components: First, a systematic literature review was conducted to identify key areas of focus regarding open data and journalism. Second, three semi-structured interviews were carried out with journalists and data analysts from small media organizations in the European Union that emphasize data journalism. Specifically, one journalist from Eurologus in Belgium was interviewed online, along with one journalist and one data analyst from Divergent in Portugal, also interviewed online, and finally, the chief editor and a data analyst from Farosnet in Greece were interviewed in person. These interviews aimed to explore how journalists utilize open data and the challenges they face, guided by insights from the literature review. Lastly, ongoing action research is being conducted at Farosnet, where the researcher is embedded through the ODECO project, employing iterative cycles of planning, action, and reflection to identify and address specific needs and challenges in integrating open data into reporting. The collected data from the interviews and action research are being analysed through a qualitative content analysis approach, where audio recordings were reviewed to extract key concepts and themes using an inductive coding process, allowing for a comprehensive understanding of journalists' experiences and challenges in utilizing open data. Through the interviews and action research in the HuffPost Greece branch, it has been continuously revealed that journalists are reluctant to share their data without monetary compensation. 7.2 As-is: Journalists The current state of governance instruments that encourage media organisations to share data in the open data ecosystem is depicted in table 10. There is a lot of training available for journalists in the use of data, and some is provided by media organizations (Establishment of coordinating functions or entities, S1), indicating an interest in developing these skills in the community. D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 38 However, this training is not specifically focused on open data. The focus is mostly on the analysis of data and the outcomes that can be extracted rather than whether the data is open or from other sources. There is an increase in data-specialized teams in large news media and small data journalism organizations (Reshuffling division of competences, S2). This leads to new specialized roles that have to focus on the use of technology and combine it with traditional storytelling journalistic practices. Sophisticated tools for data collection, analysis, and presentation have also been developed, but there are no specialized tools for open data sharing (Systems for information exchange and sharing, S5). In the case of The Guardian, they are sharing some of the datasets they have compiled through Google Sheets so anyone interested can download and use them. It has been observed in academic publications that collaboration occurs between journalists and other specialists inside organizations. Some noteworthy examples are Boyles (Boyles, 2020), which examines journalism hackathons as spaces for collaboration between journalists and technologists, and Baack (Baack, 2018), who highlights the shared practices between journalists and civic hackers and their goals in using open data to inform and engage the public. There have also been academic publications (Handler & Ferrer Conill, 2016; Palomo et al., 2019) presenting cases where journalists reached out and collaborated with their audience to collect and analyse data, as well as to extract expertise to understand specialized topics or help them with the analysis of large amounts of data. In particular, La Nación Data (Palomo et al., 2019), an Argentinian news organization, has reached out to their active audience to help them collect and open data through there Vozdata platform, and subsequently released the collected data as open data. Another interesting publication is the presentation of the Guardianʼs analysis (Daniel & Flew, 2010) of MPs' expenses, where the public was invited to help in the data analysis (Partnerships, S7). From a managerial perspective, although the formation of data journalism teams is not a common occurrence in media organizations, the hiring and training of specialists to use data is a cornerstone of strategic planning towards a new direction (Strategic planning, M1, Interorganizational culture and knowledge management, M5). Capacity building initiatives such as workshops and hackathons help journalists not only improve their networks, locate and potentially recruit tech-skilled individuals like programmers and civic hackers but also help to align their goals with those of other actors in the open data ecosystem (Capacity building, M6). D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 39 Table 10: Current governance instruments stimulating journalists users to participate in open data hackathon ecosystems Instruments Enablers S1 S2 S5 S7 M1 M5 M6 Availability of training in data skills and literacy News organizations provide training in data skills to their journalists. Through workshops and online courses News organizations provide training in data skills to their journalists. Through workshops and online courses Availability of appropriate technical tools Using tools for data analysis and sharing The use of tools for the publication and popularisation of their work aligns with managerial goals Alignment of private value and interests with open data sharing Promotion of transparency and accountability is key for journalists Availability of resources (financial, time, people/ workforce) Hiring data analysts and researchers D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 40 Instruments Enablers S1 S2 S5 S7 M1 M5 M6 Existence of datasharing communities Collaborative projects and tools are promoted from international originations Partnerships between media organizations, tech specialist or academics help align interests towards the effective use of open data Awareness about the social impact of open data sharing Presence of engagement or enjoyment activities Hackathons and workshops for innovation Hackathons and workshops for innovation D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 41 7.3 To-be: Journalists The to-be state of governance instruments that can encourage media organisations to share data as open data in the open data ecosystem is depicted in table 11. As it has been mentioned above, in chapter 7.1, although there are available resources online, the training of journalists in open data is not considered a priority in many newsrooms. Most of the training programs are focusing on general data analysis and they are using open data as datasets for training, an important function. However, to enhance the importance of open data as a pillar of journalism, more focused training that covers the principles of openness and transparency is of vital importance (Establishment of coordinating functions or entities). Although there are many tools available, the main issue is not the lack of specialized tools for journalists to share open data. Instead, the real problem is the willingness of media organizations to prioritize and commit to sharing open data initiatives. Although the design of specialized tools could boost the limited endeavours that exist at the moment, it is mainly a conceptual impediment as media organizations often perceive data as a valuable resource that they are not willing to share. To alter this perception, the importance of increased transparency that they can achieve with the use of open data and the impact that can have on their audience and their engagement must be highlighted (Systems for information exchange and sharing, Capacity building). Furthermore, although there are cases where journalists are trained in open data or collaborating with experts, this is not a popular approach in media organizations. The importance of data incorporation into new articles must be highlighted through various means, including enhancing credibility, engaging the audience, facilitating transparency and providing improved storytelling. Management needs to recognize the commercial and business value of these benefits to justify allocating more resources to the use of open data (financial, time, personnel) (Strategic planning, Inter-organizational culture and knowledge management) D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 48 Instruments Enablers S1 S3 S4 S5 S6 S7 M1 M4 M5 Alignment of private value and interests with open data sharing Different forprofit companies established Overture Maps Foundation to pool resources to produce fitfor-purpose open geospatial data. Corporations donate to the OpenStreetMa p Foundation which, in turn, gives them access to a seat in the Advisory board. Open Licensing, which allows commercial reuse, so companies can add data, and reuse the merged product Different forprofit companies established Overture Maps Foundation to pool resources to produce fitfor-purpose open geospatial data Fugro is a publicly traded company that also has to maintain a sustainabl e image Different forprofit companies established Overture Maps Foundation to pool resources to produce fitfor-purpose open geospatial data. Corporations donate to the OpenStreetMa p Foundation which, in turn, gives them access to a seat in the Advisory board. Availability of resources (financial, Establishment of the OpenStreetMa p and Different forprofit companies established D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 49 Instruments Enablers S1 S3 S4 S5 S6 S7 M1 M4 M5 time, people/ workforce) Overture Maps Foundations Overture Maps Foundation to pool resources to produce fitfor-purpose open geospatial data Existence of data sharing communitie s Establishment of the OpenStreetMa p and Overture Maps Foundations Existence of Importing ang Organised Editing Guidelines for data insertion in OpenStreetMa p Existence of Importing ang Organised Editing Guidelines for data insertion in OpenStreetMa p Awareness about the social impact of open data sharing Establishment of the Humanitarian OpenStreetMa p Team and Missing Maps Existence of Importing ang Organised Editing Guidelines for data insertion in Establishment of the Humanitarian OpenStreetMa p Team and Missing Maps Establishment of the Humanitarian OpenStreetMa p Team and Missing Maps D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 50 Instruments Enablers S1 S3 S4 S5 S6 S7 M1 M4 M5 OpenStreetMa p Presence of engagemen t or enjoyment activities Creation of projects by commercial organisation s in the micro tasking tool Maproulette, for individuals to merge their data using gamification techniques, circumventin g import barriers D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 51 8.3 To-be: Commercial organisations In this section we explain the to-be, the desired situation for commercial organisations to contribute data to the open data ecosystem, in the use case of the OpenStreetMap (OSM) ecosystem. The barriers reported in ODECO deliverable D4.1 (Re et al., 2024) have to, in the desired situation, be lowered, with the establishment of governance mechanisms. As explained before, there is a pressing demand from community members to commercial organisations, especially big corporations, to provide accountability when contributing to the project. This is due to bad experience with imports in the early years of the project (Atakua, 2019; OpenStreetMap Wiki, 2023), paired with a fear of commercial organisations getting a dominant position and breaking the balance in the OSM governance. This situation, on the other hand, makes data imports, even if desired by commercial organisations, difficult to do and not worth it in several cases. While understanding the OSM Community position, the OSM project is also losing potential value in the missing contributions. A balance needs to be created, that makes companies feel more welcome to contribute, while still ensuring data quality, correctness, and correct integration. For the specific case of OSM, we propose the creation of a Working Group in the foundation, to serve as a single point of contact to help data producers, providers, and intermediaries to align their data and procedures to the Import Guidelines, and to ensure proper sustainability mechanisms for this data. This Working Group would have close collaboration with the existing Data, Licensing and Engineering WGs. An attempt to create such a group existed already back in 2010 but was ultimately never finalised2. This group encompasses the following governance instruments: (S1) Establishment of coordinating functions or entities, (S2) Reshuffling division of competences (with the aforementioned WGs), (S7) Partnerships, (M5) Inter-organizational culture and knowledge management. We also suggest that the Engineering Working Group supports the creation and maintenance of more Communities of Practice around tools, such as the one mentioned in the as-is, with MapLibre, to tackle the technical barriers to open data sharing, as well as to utilize the technical motivations. This relates to the instruments (S7) Partnerships, and (M5) Inter-organizational culture and knowledge management. The creation of more tools should also be accompanied by the ease of use of some of them. Commercial organisations interviewed did not complain about such issue, but this is because all of the interviewed organisations are tech-savvy. Looking as well at the existing lists of OSM commercial contributors, there is a technical gap that we are not addressing. Non-technical organisations may not be contributing as much as they could, but lack of tools, resources, and expertise may be keeping them away. In addition to it, mechanisms to improve motivations have to be considered. As seen as well with the interviewed organisations, and the lists of contributing organisations, more work is needed in aligning private value to the interests of open data sharing, with business models that work for commercial organisations. Most of the contributing companies are tech-savvy, and are in the information technologies (IT), social networks, transportation, or geographic information systems (GIS) businesses. There is a gap to be reduced by creating sustainable business models both using and contributing to OSM, that align to private values in domains outside those mentioned. We propose that the currently non-existent Import Support Working Group mentioned before, and which relates to the (S1), (S2), (S7), and (M5) instruments, also promotes OSM as a place where commercial organisations can keep up-to-date data of their business locations, similarly to how companies keep their business profiles in platforms such Google Maps. This would also help lower the gap in Point of Interest data quality (Klinkhardt et al., 2023; OpenStreetMap Community Forum, 2023). 2 https://wiki.openstreetmap.org/wiki/Foundation/Import_Support_Working_Group) D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 52 Table 13: Potential governance instruments stimulating commercial organisations to participate in the open data ecosystem and share open data Instruments Enablers S1 S2 S7 M5 Availability of training in data skills and literacy Creation of an Import Support Working Group Availability of appropriate technical tools Creation of an Import Support Working Group Creation of an Import Support Working Group Creation of an Import Support Working Group. Improvement of the Engineering Working Group. Creation of an Import Support Working Group. Improvement of the Engineering Working Group. Alignment of private value and interests with open data sharing Creation of an Import Support Working Group Availability of resources (financial, time, people/ workforce) Creation of an Import Support Working Group Creation of an Import Support Working Group Existence of data sharing communities Creation of an Import Support Working Group Creation of an Import Support Working Group Creation of an Import Support Working Group Awareness about the social impact of open data sharing Creation of an Import Support Working Group Creation of an Import Support Working Group Creation of an Import Support Working Group Presence of engagement or enjoyment activities Creation of an Import Support Working Group 8.4 Conclusion In conclusion, the OSM ecosystem has demonstrated that commercial organisations can take an important role in contributing to Open Data Ecosystems, including data contributions. Their D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 53 participation is motivated by a mixture of own-private values and social values. However, barriers exist in the technical and governance domains, including a great resistance by community members towards data imports and corporationsʼ role in the ecosystem, hindering their potential in contributing to the project. To overcome these barriers and encourage greater participation, we propose the creation of a dedicated working group in the OSMF to streamline and help with data imports, as well as to enhance collaboration between stakeholders. Furthermore, aligning business models with the goals of open data sharing can attract a wider range of commercial organizations beyond the currently dominating contributors: the tech-savvy industries, creating a more inclusive and robust ecosystem. This can be paired by fostering communities of practice and developing user-friendly tools that can address technical barriers and make it easier for non-technical organizations to contribute. D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 54 9 As-is/to-be open data intermediaries 9.1 Introduction Open data intermediaries are defined as “third-party actors who provide specialized resources and capabilities to (i) enhance the supply, flow, and/or use of open data and/or (ii) strengthen the relationships among various open data stakeholders” (Shaharudin et al., 2023). Examples of open data intermediaries are software providers that pre-process open data and include the ready-touse data in the software, platform providers that facilitate the sharing and reuse of open data, and app providers that integrate open data in the app functionalities. In most scenarios, open data intermediaries are not the original3 open data providers or end-users. However, they often preprocess open data (making it more readily usable), improve it (e.g., by rectifying errors), or augment it (by combining it with non-open data). Therefore, open data intermediaries could potentially share the pre-processed, improved, or augmented data back to the open data ecosystem for others to use. In this deliverable, insights on the existing (as-is) and potential (to-be) governance instruments to stimulate open data intermediaries to share open data are based on the case studies of Esri and OpenStreetMap (OSM). We conducted 53 in-depth interviews with representatives from these two organizations and other relevant stakeholders, such as open geospatial data providers and endusers. The interviewees are in charge in the managerial or technical aspects of open data. These two organizations were selected due to their significant contributions to the open geospatial data ecosystem in the last decade. We analysed the interviews through abductive approach, following the governance instruments in Table 1. Esri is a multinational geospatial software company that has long been an open data intermediary. It serves such a role in multiple ways, such as by collecting and pre-processing open data from various sources and offering the ready-to-use data in its software (called ArcGIS), providing consultation services to open data providers and users, and developing applications and visualizations based on open data. OSM is a geospatial data crowdsource platform. While the OSM Foundation (OSMF) provides leadership, OSM is run by the community who contribute, reuse, and build applications based on the open data on the platform. 9.2 As-is: Open data intermediaries Table 14 shows the existing governance instruments that stimulate open data intermediaries to share open data. Most of them are based on the network mechanism. On the other hand, only one hierarchy-based instrument and no market-based instrument to stimulate open data sharing were identified from Esri and OSM cases. The hierarchy-based instrument referred to is the establishing of coordinating functions or entities (S1) . This instrument supported the alignment of private values and interests with open data sharing and was observed through the case of the CEO of Esri Netherlands leading the Breakthrough Project Open Geodata from 2013 to 2017. This project was initiated by the Netherlandsʼ Ministry of Economic Affairs to identify and address issues around open geodata in the Netherlands. However, while this project exemplifies the potential form of coordination that could take place to stimulate the alignment of private (and public) interests to increase open data sharing, it had limited impacts in making the non-public sector share open data. The projectʼs primary outcomes were the release of the actual elevation map of the Netherlands and satellite data from the Dutch Space Office, both from public agencies. Hence, moving forward, similar types of coordination could be leveraged to stimulate open data sharing from non-public sector organizations. 3 “Original” here is used loosely to refer to organisations whose one of the core or expected responsibilities or activities is to provide open data. D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 55 Esri leveraged the network-based instrument of systems for information exchange and sharing (S5) to stimulate open data sharing through the availability of appropriate technical tools . Esri provides the technical infrastructure for open data dissemination and reuse. This infrastructure makes it easier for data holders to share their data as open data instead of developing the open data sharing platforms themselves and from scratch. It also allows Esri itself to share some of the data it pre-processed and produced as open data, which is useable even by non-ArcGIS users. Having said that, a more significant proportion of data pre-processed and produced by Esri is usable only to ArcGIS users. This is because Esri essentially wants geospatial data users to subscribe to its software in order to be able to use all of the data it offers. Therefore, the availability of more options of technical tools for open data dissemination and reuse, especially nonproprietary ones, may help further stimulate organizations to share open data. OSM represents an entity for collective decision-making (S6) that stimulates open data contribution by spurring the existence of data sharing communities. OSM provides the infrastructure and community support for open data sharing and reuse. OSM is governed by the OSMF, which coordinates collective decision-making on how OSM data should be provided and used, not only regarding the technical aspects but also the licensing. Beyond offering a platform for anyone to contribute open data, OSM is intrinsically communal. OSM community members (consisting of OSM data providers, users, and developers) often organize events and conferences, such as the annual State of the Map conferences. These events and conferences help expand OSM community members (not only among geospatial professionals but also students and hobbyists) and facilitate discussions on the development around OSM technologies and organization. OSM communities are diverse and multi-scale instead of a single community; for example, there are OSMF-registered local chapters, informal (or non-OSMF-registered) national and regional communities, and university-based OSM associations (e.g., YouthMappers). OSM presents an example of an entity for collective decision-making that could be emulated in other domains to stimulate the contribution of open data. The partnerships (S7) and financial management: joined up working and cooperation (M4) instruments are leveraged by Esri to stimulate open data sharing by mediating the alignment of private value and interests with open data sharing. Esri strengthens its position in geographic information ecosystems by partnering with various organizations to facilitate the availability and reuse of open data. A notable example is its partnership with other companies, including Microsoft, TomTom, Amazon Web Services, and Meta, to establish Overture Maps Foundation. Overture aims to provide high-quality and fit-for-purpose open geospatial data that anyone could use, especially developers within Esriʼs and other Overture partnersʼ ecosystems. Overtureʼs partners collectively contribute human resources and infrastructure support to generate and disseminate open data. Esri also leverages the partnerships (S7) instrument to share open data based on its awareness about the social impact of open data sharing. For example, Esri works with Microsoft and Impact Observatory to produce a high-resolution global land cover map based on the European Space Agency (ESA) Sentinel-2 satellite imagery and offers it as open data. This data is helpful for conservation efforts and sustainability projects. As the executive from Esri Inc. (the parent company of Esri headquartered in California) interviewed noted, environmental sustainability is a cause that Esri is historically passionate about, befitting its original name, Environmental Systems Research Institute (E.S.R.I). Hence, Esri sees the value of offering open data that could be used for sustainability purposes. Additionally, the Humanitarian OpenStreetMap Team (HOT), an organization that uses the OSM platform to coordinate the contribution of open data for humanitarian responses and community development initiatives, also leverages the partnerships instrument. HOT works with various international and local organizations, including the United Nations and Red Cross, and mobilizes a community of volunteers to help provide open data for D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 56 specific projects/purposes, such as to be used by save and rescue teams during the earthquake in Turkey and Syria in February 2023 and the tropical storm in Malawi in 2022. D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 57 Table 14: Existing governance instruments stimulating open data intermediaries to participate in the open data ecosystem and share open data Instruments Enablers S1 S5 S6 S7 M4 Availability of training in data skills and literacy Availability of appropriate technical tools Esri provides the infrastructure for open data dissemination and reuse Alignment of private value and interests with open data sharing Esri Netherlandsʼ CEO led the multistakeholder project initiated by the government aimed at improving (open) geodata in the country Esri, together with other forprofit companies, established Overture Maps Foundation to pool resources to produce fit-for-purpose open geospatial data Esri, together with other for-profit companies, established Overture Maps Foundation to pool resources to produce fit-for-purpose open geospatial data Availability of resources (financial, time, people/ workforce) Esri, together with other for-profit companies, established Overture Maps Foundation to pool resources to produce fit-for-purpose open geospatial data D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 64 10 Summary of the results In the previous sections, we examined the various governance instruments that currently promote open data sharing by non-governmental actors and explored potential instruments that could be developed or modified in the future to address the challenge of creating an inclusive open data ecosystem. In this section, we summarize the results of the current (as is) and desirable (to be) scenario. Table 16 summarizes the applicability of governance instruments across open data actors based on as-is and to-be scenarios. Table 16: Summary of the results S1 S2 S3 S4 S5 S6 S7 M1 M2 M3 M4 M5 M6 Nonspecialists A, T T T A, T A T A A Elementary schools T A A, T T A, T T T Non-profit organisations A A, T A A, T A A, T Journalists A, T A A, T A A, T A, T A, T Commercial organisations A, T T A, T A A A A, T A A A, T Open data intermediaries A, T T T A, T A, T A, T A, T T Note: A: As-is instruments T: To-be instruments If A and T are in the same box (A, T), it means that there could be potential new or additional ways of leveraging the referred governance instrument. 10.1 As-is situation in the open data ecosystem Summarizing the results of the previous sections, we see that the current (as is) situation is characterized by the presence of different governance instruments. Open data sharing has been enabled through the establishment of a coordination function (S1) in the case of non-specialist users, journalists, commercial organizations, and open data intermediaries. This instrument assumes the configuration of the temporary structures created for hackathons, training in data skills by news organizations, fit-for-purpose working groups, and ad hoc coordinating initiatives steered by governments that address issues around open data. In the case of journalists, we observe a reshuffling of the division of competencies (S2) with the rise of data-specialized teams in both large news organizations and small data journalism outlets. This shift results in the emergence of new specialized roles that prioritize the integration of technology with traditional storytelling practices in journalism. Legal frameworks (S3) supported open data sharing in elementary schools, non-profit organizations, and commercial organizations. Legal frameworks take the shape of the compulsory adoption of ‘project-based and playful learningʼ in elementary school, or the creation of legal tools and licenses to help publish and use open data. D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 65 An example of the creation of regulated markets (S4) that led to non-governmental open data sharing is illustrated by the Overture Maps project, which is tailored to the needs of commercial organizations and developers, in contrast to the community focus of OpenStreetMap (OSM). Systems of information exchange (S5) fostered non-governmental open data sharing in the case across all actors. Non-specialist users were introduced to the (open) data ecosystem through the provision of APIs, while teachers in elementary schools used and shared data for learning activities through open education tools. The provision of a technical infrastructure, as a simplified open data editor tool for publishing data created by NPOs, or the ones created for facilitating open data sharing by intermediaries. The existence of a platform for open data sharing influences data sharing practices of commercial users, as in the case of the OpenStreetMap database. The presence of entities for collective decision-making (S6) facilitated open data sharing for commercial users, and open data intermediaries. One critical example is the one of OSM that represents an entity for collective decision-making (S6) that stimulates open data contribution by spurring the existence of open data sharing communities. The creation of new relationships in the form of partnership (S7) increased the potential for open data sharing across all actors. In the case of non-specialist users, we see lasting changes in connections to pre-hack and post-hack phases. For elementary schools, we observe recurrent cooperation between schools and organizations to create extra-curricular activities. For NPOs, we see collaborations with community of volunteers that supports the availability of training in data skills and literacy, enabling the execution of technical projects for which NPO might not have had technical resources. Communities of practices and commercial organizations develop shared tools and to produce and share fit-for-purpose open geospatial data. Open data intermediaries partner with various organizations and companies to facilitate the availability and reuse of open data. Strategic planning (M1) initiatives are undertaken by news organizations to provide training in data skills to journalists and, thus, facilitate open data sharing. Financial management fostering joined up working and cooperation (M4) mechanism is considered as important for non-specialist users, NPOs, commercial organizations, and intermediaries. In open data hackathons, as they commonly offer monetary rewards and postevent incubation (including facilitating business connections, offering shared workspaces and mentoring) to selected teams. For NPOs, support is provided through the provision of a legal structure and financial support. Thus, those communities can share the data as open data under the organisational umbrella while not having bureaucratic hurdles, which supports the availability of resources enabler. For commercial users, we see the development of pooled resources to produce fit-for-purpose open geospatial data. For open data intermediaries, we see partnerships with that result in the contribution of human resources and infrastructure to support to generate and disseminate open data. In the case of commercial organizations, we see the example of rapid editor being used for the social good to rapidly map areas in need in humanitarian projects. Inter-organizational culture and knowledge management (M5) foster open data sharing in journalists, NPOs, and commercial organizations. Journalists utilize tools for publishing and promoting their work, while NPOs prioritize openness as a core value, fostering a culture that aligns private interests with open data sharing. NPOs aim for social impact and are aware of the implications of their data sharing practices. Their organizational culture is typically less hierarchical, encouraging employees to engage in projects they are passionate about, which enhances motivation for open data sharing. D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 66 Community capacity building (M6) enhances open data sharing across non-specialist users, journalists, and commercial users. Indeed, promotion of transparency and accountability, as well as hackathons for innovations, and the establishment of teams that enhance fit-for-purpose open data sharing (e.g., the establishment of the Humanitarian OpenStreetMap Team and Missing Maps), as well as the provision of trainings, are considered important instruments to favour open data sharing. We observe that two market instruments—namely, Financial Management (input-oriented, M2) and Financial Management (performance-oriented, M3)—were not identified among the different actors. 10.2 To be situation in the open data ecosystem Moving from the current (as is) scenario to the desirable scenario (to be), in this document we tried to identify the governance instruments that have the potential to further open data sharing. The establishment of coordinating functions or entities (S1) is considered as a relevant strategy to increase open data sharing of non-specialist users, elementary schools, journalists, and commercial organizations. A strong coordination function can take the form of Permanent Hackathons Governmental bodies at different levels, such as at the EU level. For elementary schoolsʼ coordination can be achieved through ad hoc projects of digitalization agencies and local municipalities, among other institutions. The establishment of a support working group for open data sharing is expected to increase open data sharing from commercial users. Coordination strategies can also be combined with other governance instruments to maximize impact. The analysis of the practices of intermediaries leads to a call for a mix of governance instruments that combine coordination (S1) with information exchange and sharing (S5) in combination with network-based instruments of having entities for collective decision-making (S6) and the financial management: joined up working and cooperation (M4) resulting in the establishment of a national or supra-national consortium for open data sharing by non-public organizations. This consortium could create and manage a non-proprietary technical infrastructure for open data sharing, governed by a committee of public, private, and civil sector stakeholders to facilitate collective decision-making. It would be supported by joint financial cooperation from multiple stakeholders. By utilizing various governance instruments, the consortium will aim at promoting open data sharing through technical tools, alignment of private interests, availability of resources, and active data sharing communities. It is important to target actions that enable open data sharing of non-specialist users and commercial organizations through reshuffling division of competences (S2). More specifically, those who organize hackathons events and intermediaries need to acknowledge mismatches between competences and Favor the establishment of specific groups that support open data sharing. This strategy will most likely result also in the hiring of data analysts, researchers, or more generally, open data enablers. The establishment of a legal framework (S3) is expected to enable non-governmental open data sharing of non-specialist users, intermediaries, and commercial organisations. Open data hackathon participants can be invited to share their contribution and output under an open license. Also, new and bolder legislative frameworks can extend the responsibilities and obligations for open data sharing to non-public sector organizations. For example, beyond public sector bodies, the EU Directive on open data and the re-use of public sector information (Open Data Directive) only applies to some non-governmental actors such as public undertakings and does not extend to companies and non-governmental organizations. D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 67 Moving from hierarchical instruments to market instruments (S4), governments can explore the adoption of financial incentives (such as tax discounts) to companies or civil organizations (including open data intermediaries) that share open data. The use of market instruments can counterweight the costs borne for open data sharing that heavily impact less resourceful organizations. Yet, the financial sustainability of such an approach needs to be careful assessed, as open government data already require sizeable investments. Systems for information exchange and sharing (S5) can facilitate open data sharing from intermediaries (as seen in the previous paragraphs), non-specialist users, elementary schools, NPOs, and journalists. Common beginner friendly tools can support the existence of data-sharing communities. Such systems can also mirror governmental open data portals or have additional features that guide a diverse range of users to overcome technical issues in encountered in open data sharing. These systems can also embed data analysis tools and features for collaborative projects to meet the needs of journalists. The establishment of entities for collective decision-making (S6) is expected to foster open data sharing by elementary schools by helping to align private values and interests with open data sharing. For instance, collective decision-making can find a balance between research-led initiatives and the interests of urban development companies. This governance instrument has been identified as promising also in the case of intermediaries, as seen in previous paragraphs. Lasting partnerships (S7) in elementary schools might enhance current open data sharing practices (e.g., through Coding Pirates, CCA and Green Schools) for the integration of open data providing tools and methods for using, creating and sharing it. Partnerships are also envisaged by commercial organizations and intermediaries as viable instruments to favour non-governmental open data sharing. The adoption of strategic planning (M1) can drive elementary schools and journalists to catalyse open data sharing through the creation of data-sharing communities and the provision of data skills. Financial management (M2): input-oriented aimed at lowering financial barriers for nonspecialist participants in the form of grants or stipends to cover participation costs is expected to favour the contribution of non-specialist users to the open data ecosystem. In elementary schools, investments are presumed to increase the availability of training in data skills and literacy and the availability of appropriate technical tools for schools, teachers and elementary school students. Another network instrument in the form of financial management: joined up working and cooperation (M4) can tackle the availability of resources to support the projects' execution and their long-term support in NPOs, especially across geographies. The same instrument is considered as promising by intermediaries, as seen in the combination of instruments described in relation with this actor in the previous paragraphs. Inter-organizational culture and knowledge management (M5) through the establishment or improvement of data sharing supporting groups can foster open data sharing by commercial organizations. This is instrument is considered as promising also in the case of journalists, in the shape of allocating dedicated resources (financial, time, personnel) to support open data initiatives within media organizations. The capacity building (M6) network instrument can also be used to ensure the availability of training in data skills and literacy and the awareness about the social impact of open data sharing of the NPOs. Some non-profit intermediaries that have valuable data do not have the needed D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 68 technical skills to publish open data or are not aware of open data's potential impact. To help deal with that, NPOs or governmental organisations can promote training or workshops. The capacitybuilding efforts (such as training and workshops) may be undertaken not only by public organizations and NPOs but also by others, including other for-profit companies. Journalists can be stimulated to share open data through showcasing successful case studies and examples where open data sharing has led to significant public benefits. The showcase can take of interactive and engaging content using open data to demonstrate its value and potential to the audience. In the analysis of the ‘to-beʼ results, we observe that Financial Management (performanceoriented, M3) is not mentioned as a viable governance instrument to stimulate nongovernmental open data sharing. D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 69 11 Discussion and Conclusion In this contribution we researched the potential instruments to enhance open data sharing by non-governmental actors by answering to the following research question: What governance mechanisms have the potential to foster open data sharing from non-governmental actors (or data holders)? In this section, we reflect on the key findings that emerge from this study and how they relate to the major challenges supporting the creation of an inclusive open data ecosystem. It is important to note that the theoretical framework applied in this study is not normative, meaning we do not express a preference for any particular governance mode (hierarchy, market, or network) or instrument. Instead, based on our findings, we suggest that a combination of governance modes and instruments can stimulate non-governmental open data sharing. Thus, while one governance mode or instrument may already enhance or have the potential to foster data sharing, this does not imply it should be preferred or applied in isolation. In other words, we do not rank governance modes or instruments. It can be noted that the variety of governance instruments capable of promoting open data sharing from non-governmental actors is extensive. These instruments range from hierarchical approaches, such as robust coordination and legal frameworks, to grassroots measures, including systems for information exchange and collaborative partnerships. Notably, the role of marketrelated instruments remains underrecognized in the current context. Therefore, from the observations in this report, there seems to be limited steering of the behaviour of open data actors through financial incentives. One possible explanation is that the sample of non-governmental actors in this study is driven by the democratic benefits of open data, rather than business-like goals, although they are not mutually exclusive. This would likely be the case for non-profit organizations, elementary schools, and non-specialist actors. Yet, further research is needed to understand if the effects of market governance are invisible to some actors but do contribute to non-governmental open data sharing indirectly. Despite the presence of various governance instruments that already support non-governmental open data sharing, we have also identified key elements that are currently lacking for both enhancing and establishing the conditions necessary for effective open data sharing. Some instruments, more than others, are considered as popular strategies for enhancing open data sharing. For instance, there is consensus among different actor groups on the potential of coordination for through the creation of permanent bodies organizing, for instance, recurrent hackathons with long-lasting effects or through the establishment of national or supra-national consortium for open data sharing by non-public organisations. Hence, open data sharing from hackathon events may be part of this consortium. The presence of such organizations can have positive spillovers on a range on other user groups, such as non-specialists, (elementary) schools, and commercial organizations. A further development of this research might consider mapping the existence of consortia for open data sharing across other domains (e.g., academia) and understand if they led to any improvements in open data sharing practices. Another top-down approach that is also suggested from the actorsʼ perspective is the adoption of new legal frameworks or the extension of existing legal frameworks to non-governmental actors. Yet, legislation tend to have a territorial focus. An interesting extension of these considerations requires an analysis of how and if such legal frameworks are in place across different geographies and where on the spectrum from openness to closeness. For instance, what are the main barriers to legal frameworks that mandate non-governmental open data sharing? At which level of openness is feasible to share data more openly than in current practice? What is currently preventing the adoption of business-to-government (B2G) or citizens-to-government (C2G) legislations in the EU and other countries? D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 70 From a market-oriented perspective, while we see tangible evidence of investments for governmental open data sharing, there is no uptake of financial incentives in the forms of tax credits or tax incentives more generally. While it is clear that such instruments need substantial budgeting, it is also clear that there is the tendency, among most of the observations in this report, to disregard this option or to consider it relevant only in the case of for-profit users including open data intermediaries and commercial organizations. On the contrary, network governance instruments are widely popular across different actor groups, as in the case of systems for data exchange and open data sharing. Such instruments are expected to meet the need for appropriate technical tool for open data sharing and use among NPOs, elementary schools, non-specialist users, and journalists. Also, both NPOs and open data intermediaries highlight capacity building tool to improve (open) data skills, literacy and awareness among their actor groups, thus overcoming serious barriers to open data sharing in user communities. Furthermore, we observed that some of the to-be instruments to stimulate non-government data holders to share open data can also be used to stimulate them in providing other kinds of value to the open data ecosystem. For example, the instrument of capacity building can also be used to provide training on using open data apart from training on publishing open data. This expands the results of our study to embrace also other contributions to the open data ecosystem, such as data literacy. This is a relevant result that calls for additional research on how governance instruments both enhance non-governmental open data sharing and other contributions to the open data ecosystem. In conclusion, different governance instruments can lead to further open data sharing by nongovernmental actors. Some instruments are already adopted and require further uptake, while others are not yet adopted and can significantly improve open data sharing. It is important to acknowledge several limitations of our study. These limitations can form the basis for further research. First, the research approach applied in the study, which relied on a combination of methodologies, did not cover the entire spectrum of actors in an actor group and often focused on a narrow geographical scope (the EU or specific EU countries). Further research is needed to confirm and expand the results of our analysis to other perspectives within the same actor group, across different sub-fields (e.g., commercial organizations operating in various industries), scales (e.g., NPOs of different sizes), and contexts (e.g., including those lacking the capabilities to engage with technology). Yet, it must be acknowledged that investigations on open data ecosystems that look at the range of actors presented in the study are scarce and our results offer a basis for further research. Second, while the tried to address information justice, we could not derive any findings that clearly connect inclusion of non-governmental data with vulnerable groups. As such, we call for further research to explore how the inclusion of diverse users reflects and amplifies the voices of most vulnerable groups. Finally, our research did not explore the effectiveness of the identified governance instruments in promoting non-governmental open data sharing, nor did it evaluate the probability of success for those governance instruments that appear promising in enhancing such sharing. Future studies could extend our findings by assessing—either qualitatively or quantitatively—how these instruments have facilitated (or currently facilitate) non-governmental open data sharing (as is), and how they might do so more effectively in the future (to be). D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 71 References Anderson, J., Sarkar, D., & Palen, L. (2019). Corporate Editors in the Evolving Landscape of OpenStreetMap. ISPRS International Journal of Geo-Information , 8 (5), Article 5. https://doi.org/10.3390/ijgi8050232 Atakua. (2019, June 14). Atakuaʼs Diary | On Land Cover Import . OpenStreetMap. https://www.openstreetmap.org/user/Atakua/diary/368829 Atenas, J., Havemann, L., & Priego, E. (2015). Open Data as Open Educational Resources: Towards transversal skills and global citizenship. Open Praxis , 7 (4), 377. https://doi.org/10.5944/openpraxis.7.4.233 Baack, S. (2018). Practically Engaged: The entanglements between data journalism and civic tech. Digital Journalism , 6 (6), 673–692. https://doi.org/10.1080/21670811.2017.1375382 Bouckaert, G., Peters, B. G., & Verhoest, K. (2010). The coordination of public sector organizations: Shifting patterns of public management . Palgrave Macmillan. Boyles, J. L. (2020). Laboratories for news? Experimenting with journalism hackathons. Journalism , 21 (9), 1338–1354. https://doi.org/10.1177/1464884917737213 Bradshaw, P., & Rohumaa, L. (2013). The online journalism handbook: Skills to survive and thrive in the digital age . Routledge/Taylor & Francis Group. Celis Vargas, A., Magnussen, R., Mulder, I., & Larsen, B. (2023). Towards a framework for Open Data literacy in education: A systematic mapping review of Open Data skills and learning approaches. Interaction Design and Architecture(s) , 57 , 133–151. https://doi.org/10.55612/s-5002-057-008 Celis Vargas, A., Papageorgiou, G., Magnussen, R., Larsen, B., & Mulder, I. (2024). The Open Data newsroom: A game approach for developing Open Data competencies in elementary school . 18th European Conference on Games Based Learning (ECGBL 2024), Aarhus University, Denmark. Chantillon, M., Crompvoets, J., & Peristeras, V. (2020). Prioritizing public values in e-government policies: A document analysis. Information Polity , 25 (3), 275–300. https://doi.org/10.3233/IP-190126 Coddington, M. (2015). Clarifying Journalismʼs Quantitative Turn: A typology for evaluating data journalism, computational journalism, and computer-assisted reporting. Digital Journalism , 3 (3), 331–348. https://doi.org/10.1080/21670811.2014.976400 Concilio, G., Molinari, F., & Morelli, N. (2017). Empowering Citizens with Open Data by Urban Hackathons. 2017 Conference for E-Democracy and Open Government (CeDEM) , 125– 134. https://doi.org/10.1109/CeDEM.2017.28 Concilio, G., & Mulder, I. (2018). Open4Citizens Policy brief . OPEN4CITIZENS - Empowering citizens to make meaningful use of open data. http://open4citizens.eu/wpcontent/uploads/2016/01/O4C_D4.8_30.06.2018_Final.pdf D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 72 Coughlan, T. (2020). The use of open data as a material for learning. Educational Technology Research and Development , 68 (1), 383–411. https://doi.org/10.1007/s11423-019-09706y Crompvoets, J., & Ho, S. (2019). Developing a framework for national institutional arrangements in geospatial information management. In Sustainable Development Goals Connectivity Dilemma (pp. 141–161). CRC Press. Daniel, A., & Flew, T. (2010). The Guardian Reportage of the UK MP expenses scandal: A case study of computational journalism . https://api.semanticscholar.org/CorpusID:102341640 Davies, T., & Perini, F. (2016). Researching the emerging impacts of open data: Revisiting the ODDC conceptual framework. The Journal of Community Informatics , 12 (2). Di Staso, D., Christiansen, L., Kleiman, F., & Janssen, M. (In press). Articulating Social Issues with Open Data: Exploring a Game Jam Approach. Proceedings - 8th International Conference on Game Jams, Hackathons and Game Creation Events . International Conference on Game Jams, Hackathons and Game Creation Events, Copenhagen, Denmark. Di Staso, D., Mulder, I., Janssen, M., & Wolff, A. (In press). Open Data Hackathons and Game Jams: A Systematic Literature Review. In Research Handbook on Open Government . Enaholo, P. (2017). Beyond mere advocacy: CSOs and the role of intermediaries in Nigeriaʼs open data ecosystem. In F. van Schalkwyk, S. Verhulst, G. Magalhaes, J. Pane, & J. Walker (Eds.), The Social Dynamics of Open Data (pp. 89–108). African Minds. http://www.africanminds.co.za/the-social-dynamics-of-open-data/ Falk, J. (2022). How Game Jams and Hackathons Accelerate Design Processes (arXiv:2205.04966). arXiv. https://doi.org/10.48550/arXiv.2205.04966 González-Zapata, F., & Heeks, R. (2015, May 20). Understanding Multiple Roles of Intermediaries in Open Government Data . Gray, J., Bounegru, L., & Chambers, L. (2012). The Data Journalism Handbook (1st ed.). OʼReilly Media, Inc. Handler, R. A., & Ferrer Conill, R. (2016). Open Data, Crowdsourcing and Game Mechanics. A case study on civic participation in the digital age. Computer Supported Cooperative Work (CSCW) , 25 (2–3), 153–166. https://doi.org/10.1007/s10606-016-9250-0 International Open Data charter. (2015). INTERNATIONAL OPEN DATA CHARTER - Principles . https://opendatacharter.net/principles/ Johnson, J. A. (2014). From open data to information justice. Ethics and Information Technology , 16 (4), 263–274. Scopus. https://doi.org/10.1007/s10676-014-9351-8 Klinkhardt, C., Kühnel, F., Heilig, M., Lautenbach, S., Wörle, T., Vortisch, P., & Kuhnimhof, T. (2023). Quality Assessment of OpenStreetMapʼs Points of Interest with Large-Scale Real Data. Transportation Research Record , 2677 (12), 661–674. https://doi.org/10.1177/03611981231169280 Kovach, B., & Rosenstiel, T. (2007). The elements of journalism: What newspeople should know and the public should expect (1., rev. ed., completely updated and rev). Three Rivers Press. D4.3 An approach to steer the behaviour of non-government data holders towards open data through a governance strategy 73 Ktistakis, M., Di Staso, D., Lopez Reyes, M. E., Papageorgiou, G., Celis Vargas, A., Pilshchikova, L., Santoro, C., Ochoa Ortiz, H., & Shaharudin, A. A. A. (2023). D3.1 Closing the cycle: Understanding potential contributions of open government data users to the open data ecosystem . https://cordis.europa.eu/project/id/955569/results Maron, M. (2020). An Incomplete History of Companies and Professionals in OpenStreetMap . http://2020.stateofthemap.org/sessions/RHDUV9/ Meijer, A., & Potjer, S. (2018). Citizen-generated open data: An explorative analysis of 25 cases. Government Information Quarterly , 35 (4), 613–621. https://doi.org/10.1016/j.giq.2018.10.004 Mulder, I., & Kun, P. (2019). Hacking, Making, and Prototyping for Social Change. In M. de Lange & M. de Waal (Eds.), The Hackable City (pp. 225–238). Springer Singapore. https://doi.org/10.1007/978-981-13-2694-3_12 OECD. (2015). OECD Guidelines on Corporate Governance of State-Owned Enterprises, 2015 Edition . OECD. https://doi.org/10.1787/9789264244160-en OECD. (2017). Recommendation of the Council on Open Government. Paris: OECD. Retrieved December , 10 , 2019. OpenStreetMap Community Forum. (2023, December 27). POI quality and usage – any corporates contributing? - General talk . OpenStreetMap Community Forum. https://community.openstreetmap.org/t/poi-quality-and-usage-any-corporatescontributing/107488 OpenstreetMap Foundation. (2024). Corporate Members—OpenStreetMap Foundation . https://osmfoundation.org/wiki/Corporate_Members OpenStreetMap Wiki. (2023). Import/Past Problems—OpenStreetMap Wiki . https://wiki.openstreetmap.org/w/index.php?title=Import/Past_Problems&oldid=25094 12 OpenStreetMap Wiki. (2024). Organised Editing/Activities—OpenStreetMap Wiki . https://wiki.openstreetmap.org/w/index.php?title=Organised_Editing/Activities&oldid=2 656488 Palomo, B., Teruel, L., & Blanco-Castilla, E. (2019). Data Journalism Projects Based on UserGenerated Content. How La Nacion Data Transforms Active Audience into Staff. Digital Journalism , 7 (9), 1270–1288. https://doi.org/10.1080/21670811.2019.1626257 Papadimitriou, C. (2023). All in? Identifying and tackling private sectorʼs barriers to data sharing: A Perspective on geospatial data in the Netherlands [TU Delft]. https://repository.tudelft.nl/record/uuid:0e21cba1-a238-43b0-935eef70ec9b160d#metadata Pellegrino, M., & Antelmi, A. (2023). At School of Open Data: A Literature Review: Proceedings of the 15th International Conference on Computer Supported Education , 172–183. https://doi.org/10.5220/0011747500003470 Pollitt, C. (2017). Public management reform: A comparative analysis - into the age of austerity (Fourth edition). Oxford University Press.