FIDELIS landscape survey analysis

FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 1 | Page Project Title FIDELIS: Establishing A European Network of Trustworthy Digital Repositories Project Acronym FIDELIS Grant Agreement No. 101188078 Start Date of Project 2025-01-01 Duration of Project 36 months Project Website https://eden-fidelis.eu/ FIDELIS landscape survey analysis Work Packages WP5 - Building common understanding of TDRs and a harmonised matrix of repository capabilities and characteristics WP8 - Repository training and support: Initiating Network cohesion and expansion Lead Authors (Org) Thomas Jouneau (INRAE, Université de Lorraine, 0000-0001-5986-8128), Maaike Verburg (KNAW-DANS, 0000-0001-9408-3190) Contributing Author(s) (Org) Laurence Horton (UEDIN-DCC, 0000-0003-2742-6434), Geert van Geest (SIB, 0000-0002-1561-078X), Sanni Tujunen (TAU-FSD), Joel Kallio (TAU-FSD, 0009-0003-5076-9985), Tobias Paulsen (Sikt, 0009-0004-5123-6782), Severine Duvaud (SIB, 0000-0001-7892-9678), Jonas Recker (GESIS, 0000-0001-9562-3339), Åse Jorun Holthe-Tveit (Sikt, 0009-0005-8185-8854), Deborah Thorpe (KNAW-DANS, 0000-0002-2307-8770), Philipp Conzett (UiT, 0000-0002-6754-7911), Mari Kleemola (TAU-FSD, 0000-0001-8855-5075), Vasso Kalaitzi (KNAW-DANS, 0000-0001-8337-120X), Robert Huber (UBremen, 0009-0007-3843-2385), Clement Jonquet (INRAE, 0000-0002-2404-1582), Fernando Aguilar (CSIC, 0000-0002-5345-1716), Ann-Kristin Forshaug (Sikt, 0009-0004-1823-8604), Tuomas Alaterä (TAU-FSD, 0000-0002-3448-3448), Eduardo Esteves (ELIXIR, 0000-0001-5458-4978) Due Date 2025-06-30 Date 2025-06-27 Version V1.0 DOI https://doi.org/10.5281/zenodo.15744996 Dissemination Level X PU: Public PP: Restricted to other programme participants (including the Commission) RE: Restricted to a group specified by the consortium (including the Commission) CO: Confidential, only for members of the consortium (including the Commission) Versioning and contribution history Version Date Author(s) Notes 0.1 2025.03.11 Maaike Verburg (DANS), Philipp Conzett (UiT), Mari Kleemola (TAU-FSD), Åse Holthe-Tveit (Sikt), Ann Kristin Forshaug (Sikt), Tobias Paulsen (Sikt), Joel Kallio (TAU-FSD), Sanni Tujunen (TAU-FSD), Tuomas Alaterä (TAU-FSD) Survey design and launch 0.2 2025.03.25 Maaike Verburg (DANS) Set up ToC of the report 0.4 2025.06.03 Maaike Verburg (DANS), Laurence Horton (UEDIN-DCC), Geert van Geest (SIB) Drafted chapter 2 and repository characteristics results 0.5 2025.06.06 Maaike Verburg (DANS), Laurence Horton (UEDIN-DCC), Geert van Geest (SIB) Draft ready for internal review WP8 0.6 2025.06.16 Thomas Jouneau (INRAE, Université de Lorraine), Sanni Tujunen (TAU-FSD), Joel Kallio (TAU-FSD) Draft ready for internal review WP5 0.7 2025.06.23 Tobias Paulsen (Sikt) Draft ready for internal review WP2 0.8 2025.06.25 Deborah Thorpe (DANS), Fernando Aguilar (CSIC), Severine Duvaud (SIB), Philipp Conzett (UiT), Laurence Horton (UEDIN-DCC), Mari Kleemola (TAU-FSD), Jonas Recker (GESIS), Vasso Kalaitzi (DANS), Eduardo Esteves, Robert Huber (UBremen), Clement Jonquet (INRAE), Åse Jorun Holthe-Tveit (Sikt) Internal review provided 0.9 2025.06.25 Maaike Verburg (DANS), Laurence Horton (UEDIN-DCC), Geert van Geest (SIB), Thomas Jouneau (INRAE), Sanni Tujunen (TAU-FSD), Joel Kallio (TAU-FSD), Tobias Paulsen (Sikt) All internal review comments incorporated 1.0 2025.06.27 Maaike Verburg (DANS), Thomas Jouneau (INRAE) Version ready for publication FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 2 | Page Table of contents 1 Executive summary 8 2 Rationale and method 10 2.1 Rationale of the survey 10 2.2 Creation of the survey 11 2.3 Dissemination of the survey 12 2.4 Analysis methodology 13 3 Results 15 3.1 Respondent characteristics 15 Retention, curation, and preservation 18 Certification 20 3.2 Repository capabilities and standards 22 Digital Object Management 22 Repository platforms and software 31 Persistent Identifiers (PIDs) 33 Metadata schemas 34 Semantic artefacts 37 Shared services 39 Organisational Infrastructure 41 Licensing 49 Continuity 54 Funding 54 Technology and Security 56 Storage and integrity 61 Other standards, best practices, and solutions for technology and security management 62 3.3 Legal challenges 65 Additional legal questions 67 3.4 Training and support needs 69 3.5 Expectations of the FIDELIS Network 76 4 Conclusions and next steps 80 5 Appendices 84 Appendix A - Survey glossary 84 Appendix B - Additional tables and lists relevant to repository characteristics 91 Disclaimer FIDELIS has received funding from the European Commission’s Horizon Europe funding programme for research and innovation programme under the Grant Agreement no. 101188078. The content of this document does not represent the opinion of the European Commission, and the European Commission is not responsible for any use that might be made of such content. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 3 | Page List of tables Table 1. TTRAM Activities and Functions (A|F): Deposit & Appraisal 24 Table 2. TTRAM Activities and Functions (A|F) : Curation, Quality & Compliance 26 Table 3. TTRAM Activities and Functions (A|F) : Discovery & Identification 27 Table 4. TTRAM Activities and Functions (A|F) : Access 28 Table 5. TTRAM Activities and Functions (A|F) : Interoperability 28 Table 6. TTRAM Activities and Functions (A|F) : Preservation 29 Table 7. TTRAM Activities and Functions (A|F) : Provenance & Authenticity 29 Table 8. TTRAM A|F: User Support 30 Table 9. Other ideas (not explicitly related to digital objects but mentioned) 30 Table 12. TTRAM Activities and Functions (A|F) : Mission & Scope 44 Table 13. TTRAM Activities and Functions (A|F) : Governance 44 Table 14. TTRAM Activities and Functions (A|F) : Resources 45 Table 15. TTRAM Activities and Functions (A|F) : Continuity of service 45 Table 16. TTRAM Activities and Functions (A|F): People & expertise 46 Table 17. TTRAM Activities and Functions (A|F): External engagement 46 Table 17. TTRAM Activities and Functions (A|F): Legal & Ethical 47 Table 18. TTRAM Activities and Functions (A|F) : Third-party dependencies 48 Table 19. Additional ideas expressed by respondents linked to TTRAM A|F 49 Table 20. Licences for metadata 51 Table 21. Licences for data 51 Table 22. Licences for software 51 Table 23. Licences for other digital objects 52 Table 24. Licences for named “other” digital objects. 53 Table 25. TTRAM Activities and Functions (A|F): Security 58 Table 26. Sensitive data management 59 Table 27. Tech watch 59 Table 28. TTRAM Activities and Functions (A|F) : Technical infrastructure 60 Table 29. Other technology & security challenges 60 Table 10. Purposes mentioned and metadata schemas employed for them 92 Table 11. Purposes of specific catalogues, registries, resources and services mentioned by respondents. 94 List of figures Figure 1. An overview of the different Pillars of the FIDELIS project. 10 Figure 2. The geographical spread of the survey responses. 16 Figure 3. Overview of the different types of repository in the sample. 16 Figure 4. The domains served by the repositories in the sample. 17 Figure 5. The levels of retention, curation, and preservation provided by the repositories in the sample. 19 FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 4 | Page Figure 6. The levels of retention, curation, and preservation provided by discipline-specific versus generalist repositories. 20 Figure 7. Certification status and opinion on some formal certification schemas. 21 Figure 8. Challenges and needs that repositories face in terms of digital object management activities and functions. 22 Figure 9. Challenges and needs that institutional repositories face in terms of digital object management activities and functions. 23 Figure 10. Challenges and needs that national repositories face in terms of digital object management activities and functions. 23 Figure 11. Number of respondents having expressed challenges in each topic. 24 Figure 12. Top 10 platforms or software environments used by repositories. 31 Figure 13. Technologies and services used to deploy repository software or platforms. 32 Figure 14. Use of persistent identifiers in repositories for digital objects and metadata elements. 34 Figure 15. The 15 most commonly used or currently being implemented persistent identifiers. 34 Figure 16. Repartition of respondents by number of metadata schemas cited (n=159) 35 Figure 17. Ranking of metadata schemas cited. 36 Figure 18. Most commonly cited purposes for metadata schemas. 36 Figure 19. Implementation of the metadata schemas in the repository. 37 Figure 20. Use of semantic artefacts in repositories. 37 Figure 21. Implementation of the semantic artefacts in the repository. 38 Figure 22. Ranking of catalogues, registries, resources and services most cited (n>1). 40 Figure 23. Purposes most cited (n>1) for catalogues, registries, resources, services, or federated systems. 41 Figure 24. Challenges and needs that repositories face in terms of organisational infrastructure. 42 Figure 25. Challenges and needs that institutional repositories face in terms of organisational infrastructure. 43 Figure 26. Challenges and needs that national repositories face in terms of organisational infrastructure. 43 Figure 27. Repartition of types of licences by category of digital object. 50 Figure 27. State of continuity agreements. 54 Figure 28. Stakeholders of continuity agreements. 54 Figure 29. Duration of sustainable funding for repositories. 55 Figure 30. Main funding sources of repositories. 56 Figure 31. Challenges and needs that repositories face in terms of technology and security. 57 Figure 32. Number of respondents having expressed challenges in each topic. 58 Figure 33. Measures used to ensure the storage and integrity of digital objects and metadata. 61 Figure 34. Main other standards, best practices and solutions for technology and security management. 62 Figure 35. What is the legal status of your repository? 65 Figure 36. What would you say are the biggest legal challenges your repository faces when sharing data between repositories or with scientists from other research organisations? 66 Figure 37. Would your repository be interested in collaborating on frameworks and guidelines regarding GDPR and/or Intellectual Property Rights with others in the Network? 66 Figure 38. Does your repository have clear legal frameworks, policies and guidelines for handling personal data in accordance with GDPR and other relevant data protection laws? 67 FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 5 | Page Figure 39. Does your repository have clear frameworks and guidelines for sharing data within Europe? 67 Figure 40. Does your repository follow one or several specific legal metadata standards for data governance and compliance? 68 Figure 41. Would your repository be open to change to/adopt/add another core metadata schema if that would improve legal interoperability? 68 Figure 42. Where do you currently (or have you previously) find/receive training on topics related to Trustworthy Digital Repositories? 69 Figure 43. Received (blue) and desired (orange) training and support topics. 71 Figure 44. Received (blue) and desired (orange) training and support formats. 72 Figure 45. Desirability judgement of some training and support characteristics. 73 Figure 46. Judgement of importance of some training and support qualities. 74 Figure 47. How likely do you currently think it is that you would join or apply for each activity? 75 Figure 48. What do you see as the most important benefit(s) this Network should provide to you or your organisation? 76 Figure 49. What format of collaborative mechanisms would you be most likely to join? 77 Figure 50. Expected barriers to joining the FIDELIS Network. 78 FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 6 | Page TERMINOLOGY Note #1: Many standards, schemas, and softwares are mentioned in the results. The clarifications and links to these will not be included in full in this report, but rather accompany the data when it is published. Note #2: Appendix A presents the glossary that accompanied the survey, which presents more information about the terminologies and terms used. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 7 | Page Terminology/Acronym Description AI Artificial Intelligence API Application Programming Interface CESSDA ERIC Consortium of European Social Science Data Archives CLARIN ERIC Common Language Resources and Technology Infrastructure COAR Confederation of Open Access Repositories DARIAH Digital Research Infrastructure for the Arts and Humanities DMP Data Management Plan ELIXIR Distributed infrastructure for life-science data EOSC European Open Science Cloud EOSC EDEN EU Project for Enhancing Digital preservation strategies at European and National level FAIR Findable, Accessible, Interoperable, Reusable GDPR General Data Protection Regulation PID Persistent Identifier R&D Research and Development RDA Research Data Association RDM Research Data Management TDR Trustworthy Digital Repository TTRAM Transparent Trustworthy Repository Attributes Matrix TTRAM A|F Activities and Functions, the elements in which the TTRAM is structured. These Activities and Functions are derived from a range of existing standards, criteria, and requirements. 1 Executive summary The FIDELIS project aims to establish a European Network of Trustworthy Digital Repositories (TDRs) that will support their development and growth, foster harmonisation and interoperability, and strengthen relevant skills through training and support. For this project to successfully achieve its goals, it was deemed essential to gather input from the community on their current practices, challenges, and needs. The outputs of this survey will be used to shape the activities of the project in the next years, and subsequently help shape the FIDELIS Network as it is designed and implemented. This activity was a collaborative effort across the project, with each part of the project focusing on their own sections of interest. This one report presents the full analysis of the survey results, covering the two associated Milestones (M11 survey analysis ready and M19 Analysis of survey results with regards to support and training) and additional analyses on other relevant topics. The survey covered the following topics: ● Repository Characteristics ● Activities & Functions ○ Digital Object Management ○ Organizational Infrastructure ○ Technology & Security ● Training & Support Needs ● Legal Challenges ● Network Interest & Expectations Responses were obtained from 159 individuals representing 144 different repositories (counting unique names). This report covers all results and findings from the survey. The main findings are considered to be: ➢ Most repositories offer Deposit Compliance as their level of care to their objects. However, many repositories (especially discipline-specific ones) are currently working on implementing Active Preservation as a more extensive level of care. ➢ Most repositories show good adoption of common artifacts, already interfaced or interoperable, such as persistent identifiers, commonly adopted metadata schemas (while retaining more confidential, disciplinary or “niche” artifacts where necessary), or licenses. The result is an international repository landscape that is increasingly converging in terms of common standards and practices. ➢ Lack of funding (particularly in a context where project-based funding is increasingly the norm), and lack of human resources or people with the right expertise, are the most commonly cited challenges. These two big concerns can be linked with another one, the difficulty to keep a repository up to date with the new technologies or the new demands and challenges emerging almost every year. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 8 | Page ➢ While only few respondents seem to rely on third party services for some core services of their repository, many of them expressed this reliance as a challenge rather than an opportunity. ➢ The survey reveals a fragmented legal landscape, with many repositories lacking formal frameworks for GDPR compliance or cross-border data sharing, responsibility for GDPR compliance is often placed on researchers. The FIDELIS Network can help by promoting shared templates, peer-to-peer learning, and working groups, but success depends on active member participation. ➢ Topics that are valued most to receive more training and support on are repository certification, long-term preservation, developing and sustaining a repository, access conditions and sensitive data, and stakeholder engagement and communication. ➢ Other preferences expressed about training and support activities are for them to be practical and specific, have a clear value and return on investment (time and effort), and for participants to have the opportunity to connect with, learn from, and network with fellow participants, organisers, (guest) speakers, and other experts during the activity. ➢ Respondents showed a general enthusiasm about the potential benefits of the FIDELIS Network. It was suggested that granular and practical work on tangible outcomes that can then be adopted throughout the Network and communicating in a combination of (online) meetings and a forum would be the most effective way to establish the Network. Financial constraints were the most envisioned barrier to participating in the Network. The results of this survey have provided valuable insights into the current activities, challenges, and needs of the community when it comes to a wide variety of topics, as well as their initial opinions on the FIDELIS project and our aims. The findings will be used to shape the activities of the project, identifying where support and training is needed and in which ways we can bring people together to jointly tackle challenges and advance their developments. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 9 | Page Figure 2. The geographical spread of the survey responses. Responses from countries outside of Europe: Armenia: 1, Canada: 2, Tunisia: 1, Uganda: 1, United States: 1. Figure 3. Overview of the different types of repository in the sample. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 16 | Page The largest portion of responding repositories identified as a discipline-specific repository serving one or more specific domains (n = 105, compared to 54 generic repositories), with most of them serving the domain of the Natural Sciences (see Figure 4) and all domains as defined in the survey (according to the Frascati fields of R&D classification11) represented to some extent. Most repositories hold data as their primary content type (n = 139), with literature (n = 42), software/code (n = 30), and other content types (n = 16) much less represented. Respondents most often indicated their role(s) as being repository manager (n = 126) or curator (n = 55). It can be challenging to capture job descriptions with generic titles, and respondents were able to indicate all roles that they felt matched their work. In general, we received some different perspectives in the survey responses, also including those of board members and system engineers. Figure 4. The domains served by the repositories in the sample. Respondents were asked to include all the persistent identifier(s) their repository can be identified by. As a result, we collected 83 RORs12, 110 re3data identifiers13, 74 FAIRsharing DOIs14, and a handful of other identifiers, such as OpenDOAR15 and ISNI16. This indicates that although the identifiers are in good use, there is still some improvement that can be achieved for repository findability when it 16 https://isni.org/ 15 https://opendoar.ac.uk/ 14 https://fairsharing.org/ 13 https://www.re3data.org/ 12 https://ror.org/ 11 OECD (2015), Frascati Manual 2015: Guidelines for Collecting and Reporting Data on Research and Experimental Development, The Measurement of Scientific, Technological and Innovation Activities, OECD Publishing, Paris, https://doi.org/10.1787/9789264239012-en (p.59) FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 17 | Page comes to using persistent identifiers as well as to synchronise the information presented across the different schemas. Respondents were asked which communities they consider their repository to be a part of. The communities mentioned most often were RDA, ELIXIR, CESSDA, CLARIN, and DARIAH. Other notable communities mentioned were COAR and EOSC. This information is important to take into account when designing a new community for repositories. How can the FIDELIS Network be a unique addition to the kind of membership and support that is already available to the targeted repositories? Retention, curation, and preservation Respondents were asked about the levels of retention, curation, and preservation provided by the repository. The levels defined were in line with the ones proposed in the position paper from the CoreTrustSeal board17: ● As deposited - Digital object content and supporting metadata are distributed to users exactly as they are provided by depositors. ● Deposit compliance - Digital object content and supporting metadata deposited are checked for compliance with defined criteria, e.g. digital object formats, metadata elements, and compliance with legal and ethical norms. ● Initial curation - The digital objects are curated by your repository to meet defined criteria, which may exceed those defined for Deposit Compliance. ● Active preservation - In addition to Deposit Compliance and Initial Curation, your repository takes long-term responsibility for ensuring that the digital objects and metadata can be understood and rendered as required by the designated community for reuse. Repositories were asked to indicate for each level whether it was not provided, planned to be provided in the future, in the implementation phase to provide, implemented to provide, or whether the respondent was unsure about this information. For most levels respondents indicated that it was currently implemented and thus actively provided by their service. The ‘as deposited’ level of retention was most often indicated as not being provided at all (see Figure 5). Active preservation was indicated as planned (n = 20) or in implementation (n = 17) the most of all levels, suggesting this level of care is being seen as desirable by more repositories or their stakeholders. 17 CoreTrustSeal Standards and Certification Board. (2024). Curation & Preservation Levels: CoreTrustSeal Position Paper. Zenodo. https://doi.org/10.5281/zenodo.11476980 FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 18 | Page Figure 5. The levels of retention, curation, and preservation provided by the repositories in the sample. Considering the two main types of repositories (discipline-specific and generalist), it can be seen in Figure 6 that all levels of retention, curation, and preservation are most often already implemented in either type of repository. Two noticeable observations are that discipline-specific repositories seem to not implement ‘as deposited’ level of retention more often, and are more often working on offering active preservation than generalist repositories. This may help identify more specific audiences for certain topics of training or harmonisation. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 19 | Page Figure 6. The levels of retention, curation, and preservation provided by discipline-specific versus generalist repositories. Certification Respondents were asked about four formal certifications: CoreTrustSeal18, Nestor Seal (DIN31644)19, ISO 1636320, and ISO 2700121. For each certification, one of the following options could be indicated: ● Do not have, not interested in obtaining ● Do not have, interested in obtaining ● In progress of achieving ● Currently certified ● Expired, in process of renewing ● Expired, not planning to renew ● Do not know /unsure. The largest part of respondents indicated not being interested in obtaining the Nestor Seal or either of ISO certifications, or being unsure about them. The exception was CoreTrustSeal, with more 21 https://www.iso.org/standard/27001 20 https://www.iso.org/standard/87472.html 19 https://www.langzeitarchivierung.de/Webs/nestor/EN/Zertifizierung/nestor_Siegel/siegel.html 18 https://www.coretrustseal.org/ FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 20 | Page interest in obtaining the certification and also the most repositories indicating having achieved the certification or being in the process of (re)certifying (see Figure 7). Besides these four included formal certification schemes, respondents could also indicate other certifications that their repository has or that are of specific relevance or interest to their repository. Some recurring mentions include the community mandates, guidelines, and services, such as the ELIXIR Deposition Database, CESSDA Service Provider obligations, and CLARIN B-centre, as well as national initiatives such as the German DINI22, and global resources such as RDA and OpenAIRE guidelines. Figure 7. Certification status and opinion on some formal certification schemas. 22 https://dini.de/dienste-projekte/dini-zertifikat/english/about-the-certificate FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 21 | Page 3.2 Repository capabilities and standards The three subsections of this part of the survey were presented in a similar way: an indication of the challenges and needs, followed by a series of additional questions which were only shown if the respondent agreed to answer them23. Digital Object Management Challenges and needs Respondents were first asked to indicate the difficulties and needs they encounter in a controlled matrix: on the y-axis, a list of eight categories derived from the TTRAM (Transparent Trustworthy Repository Attributes Matrix)24 ; on the x-axis, a 5-point scale ranging from “no challenges” to “critical challenges”. Then, an open-ended question (“What are the major challenges and needs your repository faces in terms of digital object management?”) allowed them to identify key challenges and provide more details, if they wished. For each of the eight TTRAM Activity and Functions (A|F) Figure 8 shows the distribution of responses on a five-point-scale, illustrating the respondents’ assessment of the severity of the challenge that the particular A|F poses. Figure 8. Challenges and needs that repositories face in terms of digital object management activities and functions. 24 L'Hours, H., Kleemola, M., Parkes, O., Recker, J., Duvaud, S., van Horik, R., Alaterä, T. J., Liberante, F., Conzett, P., Kaartinen, H., Bäckman, S., & Esteves, E. (2025). FIDELIS TTRAMatrix v00.02 Guide (v01.00). Zenodo. https://doi.org/10.5281/zenodo.15676189 23 It had been considered to systematically aggregate the multiple entries from one repository. This has however been abandoned because of the mixed nature of the questions: technical, so repository-focused (such as the metadata schemas or PIDs used); and subjective, so respondent-focused, such as the questions about challenges. In the figures, the numbers represent the respondents, except otherwise specified, and for the whole 3.2 section. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 22 | Page Clearly, three challenging activities and functions are emerging from Figure 8: “Curation, Quality & Compliance”, “Interoperability” and “Preservation”. Without overinterpreting these results, it is possible that the least problematic activities and functions(namely, “Discovery & Identification”, “Access”, and “Provenance & Authenticity”), are those for which commonly used repository software tools provide built-in support, such as metadata schemas, DOI minting, or harvesting capabilities. It also could be linked to the lack of expert human resources found across the survey. A simple cross tabulation was also conducted, offering a focus on institutional (Figure 9) and national (Figure 10) repositories. These focuses do not significantly differ from the global picture, with “Curation, Quality & Compliance”, “Interoperability” and “Preservation” remaining the primary issues. It is however worth noting the even higher degree of difficulty attributed to interoperability issues for national repositories, compared with the general graph. Figure 9. Challenges and needs that institutional repositories face in terms of digital object management activities and functions. Figure 10. Challenges and needs that national repositories face in terms of digital object management activities and functions. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 23 | Page Figure 11 shows the result of mapping the responses to the open-ended question to the eight TTRAM A|F also applied above. These 78 answers are rich and deserve a deeper dive. Even if the focus of the question is operational or technical, we find already a number of concerns that appear throughout the survey: limited human resources for curation, preservation, and user support; insufficient automation and integration of tools; and the need for recognised guidelines across deposit, appraisal, and preservation workflows. Metadata quality, standardization, and richness are recurring concerns, particularly in enhancing discovery and long-term usability. Legal topics (e.g., GDPR, licensing) and access to high-volume or sensitive data also are present. Figure 11. Number of respondents having expressed challenges in each topic. For each of the TTRAM A|F considered here, Tables 1-9 show the number of mentions of different categories of ideas and challenges induced from the responses. In addition, other TTRAM A|F are cross-referenced if it is relevant to the expressed ideas and challenges. In the subsequent comments and key points, the number of mentions for each idea is indicated in brackets. Table 1. TTRAM Activities and Functions (A|F): Deposit & Appraisal Idea expressed Mentions Other TTRAM A|F High volume data 4 Storage & Integrity Automated support during deposit 4 FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 24 | Page Idea expressed Mentions Other TTRAM A|F Recognised guidelines for appraisal 2 Legal: GDPR 2 Legal & Ethical Capacity to expose machine-friendly deposit guidelines 2 Automated mechanisms for appraisal (during the deposit) 2 Support during deposit: legal appraisal 1 Data: focus on raw data while user interest is in derived data 1 Metadata: lack of user expertise or attention 1 Data complexity 1 Lack of documentation for the end user 1 Training Interoperability with platforms and tools used by researchers 1 Interoperability Lack of manifest to describe "completeness" 1 Lack of user expertise 1 Training, External engagement Learning how to deal with qualitative and relational data 1 Metadata user guidance 1 No appraisal because of repository missions to accept all long-tail data from the institution 1 Mission & Scope Scalability (ever growing datasets) 1 Support depositors and make them keep the data consumer in mind 1 User training Key points: Even if the lack of user expertise is only cited explicitly once, most remarks regarding deposit seem to address this problem indirectly: ● While the most traditional way to address it, user documentation, is also mentioned only once, other ideas emerge. Several respondents suggested integrating automated solutions during the depositing process, be it for communicating data appraisal criteria (2), metadata completion (4), or more traditionally by integrated guidance (1). ● The need to make the depositors aware of the needs of the user who is going to receive or download the data is interestingly raised (1). ● To ease the deposit, a better interoperability with platforms and tools used by researchers (such as automated dataset creation) is also suggested. ● The problems raised by high volume data are naturally met at that “early” stage (4) as well as those induced by complex data (1) or legal matters (1). ● Some challenges can be repository-specific: one respondent mentions the scope of their institutional repository (long-tail data from the institution) does not allow them to be too picky regarding the data uploaded. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 25 | Page GitHub; GitLab; Zenodo” in responses. The 30 “Other” mentions come on their own (18) or in conjunction with other solutions (12). For the latter, the entries fall into the shared tools or infrastructure such as Microsoft Azure25, Nextcloud26, or EMBL-EBI27 for example. The former focuses more on the dedicated repository software: ● seven custom (in-house) developments, mostly based on open source software (except one) ● nine specific solutions (OntoPortal28 ; Reposis29 ; Samvera Hyrax or Samvera custom build30 (2); Bespoke; ICAT31; a commercial AWS Infrastructure; STAR; GeoNetwork32). ● a mention of NodeJS which seems out of scope. Only one respondent mentions Figshare33 as a repository solution in this particular question. However, Figshare is mentioned by at least two respondents in other questions (on organisational challenges). The next question asked was “What technologies or services do you use to deploy the platform/software on which your repository operates?” -and was a multiple choice question offering eight options. The responses are summarized in Figure 13. Figure 13. Technologies and services used to deploy repository software or platforms. 33 https://figshare.com/ 32 https://geonetwork-opensource.org/ 31 https://icatproject.org/ 30 https://samvera.org/ 29 “Repository Service of the GBV (Gemeinsamer Bibliotheksverbund) based on MyCoRe” 28 https://ontoportal.org/ 27 https://www.ebi.ac.uk/ 26 https://nextcloud.com/ 25 https://azure.microsoft.com/ FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 32 | Page Beyond the observation that “traditional” deployment (i.e., as on-premise installation on local servers) is the most favoured option, it is also possible to divide the numbers between the respondents who declare solutions self-hosted or hosted by a partner institution (52), and the ones mentioning solutions managed by a third-party provider, commercial or otherwise (28 if we include the hybrid deployments). When respondents checked “Other”, they reported slight variations on the main choices. ● Cloud deployment, all systems managed within the infrastructure apart from Colectica. ● two wanted to check both “host institution” and “3rd party” for their cloud; ● “Hybrid deployment hosted by a partner institution and managed by the same partner institution and a third-party provider” ● “cloud deployment, all systems managed within the infrastructure apart from us”; ● “on-premises deployment on virtual machines managed within the infrastructure of host institution” ● work in progress: migrating from On-premises deployment to institutional partner’s cloud deployment. Persistent Identifiers (PIDs) The respondents were asked whether persistent identifiers (PIDs) were applied to digital objects in their repositories, and if so, to tick in a controlled list34 every type of PIDs which applied. A note was added to avoid confusion with the previous question asking for “persistent identifiers [their] repositories can be identified by”. No distinction was made in the question between digital objects actually held in the repository, and metadata elements pointing to other resources (persons or institutions registries, publication repositories, journals, etc.). Most repositories (84 out of 91) apply PIDs, 3 are implementing them. Only one respondent checked “No”. Regarding the persistent identifiers that are used, DOIs (provided through DataCite or other agencies) or, more broadly, Handles, make up the largest part. RORs and ORCIDs are probably used for metadata alignment. The additional identifiers not present in the listbut mentioned by respondents include IdHAL; MorpheusModelID, ProteomeXchange, w3id, Wikidata ID, and finally, internal identifiers. Figures 14 and 15 show the implementation status of PIDs stated by the respondents as well as the 15 most commonly mentioned PID systems. 34 The proposed list was rather extensive, with some entries (URLs, URIs, notably) not actually endorsed as PIDs by most recommendations or policies, including EOSC PID policy. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 33 | Page Figure 14. Use of persistent identifiers in repositories for digital objects and metadata elements. Figure 15. The 15 most commonly used or currently being implemented persistent identifiers. Metadata schemas The respondents were asked to cite up to five main metadata schemas used for digital object management, and specify the purpose for each of them in a free text matrix. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 34 | Page Not surprisingly, there is a large variety of schemas mentioned by respondents (56 in total) with a handful of well-known ones that dominate this section. The most common schemas that are mentioned are ubiquitous. Dublin Core (54) is commonly used as a minimal format for metadata exchange and the DataCite metadata schema (29) is automatically used for datasets registered at that agency. One could infer that these schemas are so commonly used that they are implicitly assumed to be present, and that many respondents did not think of mentioning these, even if they are applied by the repository. The large number of references to DSpace internal metadata (see Figure 17) comes from a series of DSpace instances used in the same community. As in other questions, it is advised to not overstate these numbers. Putting this aside, next in the ranking come DDI (24) and schema.org (12). In turn, only a small number of responses mentioned DCAT (5). Some respondents cited tools, protocols (SKOS), formats (such as JSON), thesauri, etc., which are not metadata schemas strictly speaking (19 occurrences). While some of these artefacts may embed metadata, we left them aside35. Among purposes cited, the identifier registration (DataCite) is also found in the questions about catalogues, registries and services. As for other free-text matrices, both metadata schemas and purposes have been coded from the answers. The question received 86 answers, with the distribution as presented in Figure 16: Figure 16. Repartition of respondents by number of metadata schemas cited (n=159) 35 The entries not taken into consideration are as follows: CF conventions, CIF (Crystallographic Information File), Computational Workflow, DIF (Directory Interchange Format), ELSST (European Language Social Science Thesaurus ), GDAL/OGR, GeoJSON, JSON, MPEG-21 schema file, OWL, RDF, RDFS, README, RO-Crate, SKOS. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 35 | Page The short list of the most cited schemas shown in Figure 17 does not bear any surprise. The 13 mentions of DSpace internal metadata are due to the presence of the repositories from Serbia. Figure 17. Ranking of metadata schemas cited. If we break each one of these schemas into the purposes cited, we get a list which can be found in Appendix B. Figure 18 shows the ranking of most commonly cited purposes. Table 10 links these purposes to the metadata schemas mentioned by the respondents for the given purpose. Figure 18. Most commonly cited purposes for metadata schemas. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 36 | Page Table 10 matches the identified purposes with the schemas cited by the respondents. To ease the reading, it can be found in Appendix B. Respondents were then asked to specify how the metadata schemas were actually implemented. Figure 19 shows the answers. Figure 19. Implementation of the metadata schemas in the repository. The nine respondents who checked “Other” expressed the following ideas: ● A user interface is provided, in the repository software or elsewhere (e.g., AIMS) (3) ● Configuration at the repository level (Dataverse) (2) ● Depends on the metadata schema (PREMIS/DNX hard-encoded; DC is configurable; mappings on-demand…) (1) ● Interfaced to external thesauri and metadata schema (ELSST) (1) Semantic artefacts Respondents were asked to specify if their repository uses semantic artefacts (e.g., controlled vocabularies, ontologies, taxonomies) to annotate, describe or index digital objects, and to indicate the implementation status of the artefacts (see Figure 20). Figure 20. Use of semantic artefacts in repositories. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 37 | Page These responses show the importance of relying on semantic artefacts within data repositories is generally well perceived, 70 respondents over 90 having implemented some form of semantic artefact integration within their data repositories. Seven respondents, however, have not yet taken any action, nor made plans, regarding semantic artefacts, suggesting an area for improvement in the FIDELIS initiative. The responses regarding implementation status of the semantic artefacts highlight that half of these implementations are done in a way that can possibly be improved to ease their use and maintenance (see Figure 21). FIDELIS in the future could promote the use of external semantic artefact catalogues, e.g. through the use of a standard API such as the MOD-API developed in FAIR-IMPACT36, rather than an embedding or a hard-coding. Thirty respondents mentioned relying on an external tool, service or semantic artefact catalogue. These responses might need a further inquiry to qualify the different practices (services involved, technical specificities) more precisely. Figure 21. Implementation of the semantic artefacts in the repository. 36 A key takeaway of FAIR-IMPACT was the recommendation to use external vocabulary services or semantic artefact catalogues to manage vocabularies. The reasons are multiple. First, versioning and maintenance: external services specialize in maintaining vocabularies over time, including updates, corrections, and deprecated terms, which ensures long-term semantic consistency and guarantee to have the latest version of the vocabulary. Second, a separation of concerns : decoupling vocabulary management from the repository architecture makes the system more modular, maintainable, and adaptable to change. Third, an access to APIs and tools, since external services often provide API (possibly standard ones), and tooling that enables them to go beyond the use of controlled terms e.g., semantic annotation, vocabulary recommendation, term proposal, etc. (See : Wilson, Antony, et Clement Jonquet. «D4.3 - Specification of Shared Metadata Description of Semantic Artefacts and Their Catalogues Including Common Reference API» 2024. https://doi.org/10.5281/ZENODO.12579779) FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 38 | Page From the responses of participants who ticked "Other", the following details can be noted: ● Manual entry (2) ● Manual entry in a dedicated Dataverse metadata field which is mapped to DDI (1) ● “Can be used and searched for via the metadata profile generator, more terminologies need to be added (coscine)” (1) ● “Controlled and linked” (1) ● “Harvesting external services into local data storage for optimised search and discovery” (1) ● “Our own software and own vocabulary server that is fed from external authoritative servers when we are not the authoritative source for the vocabulary” (1) ● “We have captured project metadata” (1). Shared services Respondents were asked about the shared catalogues, registries, resources, services, and federated systems they used. For ease of reading, the term "service" is used generically to refer to the five object categories addressed in the questions. The respondents were asked to cite up to five main catalogues or registries used for digital object management, and specify the purpose for each of them in a free text matrix. The same was asked regarding resources, services or federated systems. The answers to these two questions reveal a certain confusion regarding their respective focus, which some respondents explicitly expressed (“Maybe I don't understand this question, what do you mean by catalogue or registry?”; “I don't quite understand the question above”, etc.). In fact, some catalogues or services are mentioned under both questions, while many entries in response to the second question actually refer to vocabularies or registries (e.g., MeSH, CESSDA vocabulary service, registries of institutions or people, etc.). The mention of DataCite for minting DOIs (coded below as “Identifier registration”) is ubiquitous across the different answers. This confusion may not be particularly problematic, in the sense that the services listed in both matrices commonly act as third parties supporting various core or additional repository functionalities. In this regard, the top ten list and the indicated purposes are meaningful. For this reason, we present the data for these two questions together. To do so, we combined the answers to both questions in a single list. The free text answers were coded, both for the items listed and their associated purposes. We checked that there were no identical answers from the same repository. A large number of services were mentioned in the answers; many of which had a limited scope (e.g. discipline-specific or limited to a particular institution or country). Together, the two questions received responses from 49 repositories. One service (NomadLite, a local service from the University of Belgrade) was mentioned twice in 12 answers to cover two different purposes. To correct this, we reduced these occurrences by half in the following graph. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 39 | Page The number of different services mentioned was quite large. In Figure 22 below, we present the distribution of mentions only for the services mentioned more than once37. Figure 22. Ranking of catalogues, registries, resources and services most cited (n>1). A breakdown of these responses by purposes mentioned results in a list which can be found in Appendix B. Figure 23 shows the ranking of most commonly indicated purposes for the services mentioned. 37 The resources and purpose cited FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 40 | Page Figure 23. Purposes most cited (n>1) for catalogues, registries, resources, services, or federated systems. This list of purposes largely summarises some of the functionalities of repositories that are not provided by the repository software alone, but rely on third parties to operate, for example identifier registration or everything that pertains to semantic interoperability (such as ontologies and vocabularies). Table 11 matches the identified purposes with the services cited by the respondents. To ease the reading, it can be found in Appendix B. Organisational Infrastructure Like for Digital Object management, respondents were first asked to indicate the difficulties and needs they encounter regarding organisational infrastructure in a controlled matrix: on the y-axis, a list of sixteen activities and functions derived from the TTRAM; on the x-axis, a 5-point scale ranging from “no challenges” to “critical challenges”. Then, a free text question asked the respondents to list the major challenges they face and add any clarifications they wished. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 41 | Page Critical areas include managing sensitive data, applying appropriate rights and licensing schemes, and ensuring regulatory compliance at organisational and file levels. Repositories also emphasised the importance of bespoke legal and technical support for implementing Traditional Knowledge (TK) labels and managing data privacy. Additional concerns raised included coordinating legal responsibilities within consortia and aligning with national and international standards. Overall, there is a clear demand for actionable guidance that integrates human legal expertise with technical implementation in order to address the diverse legal and ethical scenarios encountered in digital repository management. For the TTRAM A|F Analysis & Impact, no table is necessary. Six respondents mentioned the need for impact measurement or meaningful indicators. A respondent wishes there would be a “framework including meaningful indicators of mission achievement for TDRs”, while another one deplores the lack of clear metrics or indicators measuring the reuse of datasets. Challenges related to training were mentioned twice in this survey section, but training challenges are also apparent in other sections of the survey. Table 18. TTRAM Activities and Functions (A|F) : Third-party dependencies Idea expressed Mentions Other TTRAM A|F Coordination between a partner institution and a third party provider 1 Development of and cooperation with the university's CRIS system, which is currently under construction. 1 Third party dependencies 3 Third party dependencies (Duraspace) 2 Third party dependencies (Ex Libris Rosetta) 2 Third party dependencies [Figshare] 1 Key points: Third-party dependencies for a repository service can be a liability, if the agreement with a provider covers core or vital features of the repository service. This advocates in favour of a greater autonomy in managing critical infrastructure components. Some cases outlined here involve vendors of software-as-a-service solutions for repositories by their names. Of the three mentions of third parties dependencies, at least one is also in this kind of situation (“We are relying on 3rd party software, so there would be a challenge if the University stopped funding the subscription, or if the company went out of business”). This reliance may also imply limitations in flexibility, FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 48 | Page responsiveness, long-term planning, or access to core or basic services (“we struggle to get any meaningful usage statistics about the data held on our repository”)38. Some other ideas that do not necessarily pertain to this question are listed in Table 19 below. Table 19. Additional ideas expressed by respondents linked to TTRAM A|F Idea expressed Mentions Other TTRAM A|F Assessing data quality across new formats and research methods 1 Deposit & Appraisal Data integrity 1 Storage & Integrity Framework for depositing data linked to publications 1 Discovery & Identification Having research data appraisal guidelines for research organisations 1 Deposit & Appraisal “I cannot legally comment on them.” 1 No challenges 1 Preservation 1 Preservation Semantic interoperability 1 Interoperability Server maintenance (power outage) 1 Technical infrastructure Support for reproducibility of data and software, also in relation to Machine Learning (ML) and Artificial Intelligence (AI). 1 Interoperability Tool integration with the repository for data and software 1 Interoperability Additional questions regarding organisational infrastructure After the two questions regarding challenges and needs, respondents were asked: “These were the core questions on organisational infrastructure. Would you like to answer some additional questions or continue to the next part of the survey?”. 78 respondents checked “Yes”, triggering the following questions. Licensing The respondents were asked to specify which licence(s) they used for four categories of digital objects: metadata, data, code/software, and other digital objects with a request to precise which digital object the licence applies to. More specifically, the questions were as follows: 38 Another respondent, however, mentions that [they] “do not have any issues with the challenges listed above as we use Figshare's infrastructure, which is already highly developed and geared towards a generalist repository paradigm.“. As no challenge is listed, it does not figure in the numbers here. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 49 | Page ● For the reuse of metadata, which licence(s) does your repository apply? ● For the reuse of data, which licence(s) does your repository apply? ● For the reuse of code/software, which licence(s) does your repository apply? ● For the reuse of digital objects other than metadata, data or code/software, which licence(s) does your repository apply? Please specify the type of digital object and the licence(s). Respondents were allowed to provide any number of licences for any type of content. Although the answers were in free text, they were relatively easy to aggregate, as most of the mentioned licences are widely recognized and commonly used. Figure 27 shows the distribution of different licences used for different types of digital objects, e.g. metadata, data, code/software and other. Figure 27. Repartition of types of licences by category of digital object. These numbers should be interpreted with caution. Some respondents simply copy-pasted a long list of licences in response to each question without refining their own specificity. This may explain the apparent application of CC licences to software, which could be however explained otherwise by the embedding of code in the dataset. The abbreviations we used for licences are derived from the Linux Foundation Project webpage39. We did not keep the version numbers, as respondents most often did not specify them. Licences for Metadata For metadata, i.e. contents usually freely displayed and disseminated, CC0 dominates the answers (see Table 20). 39 https://spdx.org/licenses/ FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 50 | Page Table 20. Licences for metadata (n=66) Licence Metadata Total CC0 CC0 46 46 Creative Commons licence requiring attribution CC-BY or compliant 13 18 CC-BY-SA 3 CC-BY-NC 2 CC-BY-NC-SA 0 Licences for Data As shown in Table 21, Creative Commons licences, (licenses demanding attribution), dominate the picture. The two responses containing six software licences each seem to contain copy-pasted identical lists of licenses across all types of items. It is therefore uncertain whether these entries correspond to an actual application. Table 21. Licences for data (n=59) Licence Data Regrouped CC0 CC0 10 10 Creative Commons licences requiring attribution CC-BY or compliant 28 58 CC-BY-SA 14 CC-BY-NC 14 CC-BY-NC-SA 2 Software "libre" licence BSD 2 12 Apache 2 LGPL 2 AGPL 2 ODC 2 MIT 2 Commercial licence 1 1 Licences for Software As expected, software-dedicated licences predominate in this section (see Table 22). Table 22. Licences for software (n=49) Licence Software Total CC0 CC0 4 4 Creative Commons licences requiring attribution CC-BY or compliant 9 29 CC-BY-SA 8 FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 51 | Page Licence Software Total CC-BY-NC 6 CC-BY-NC-SA 6 Software "libre" licence BSD 8 55 Apache 9 LGPL 12 AGPL 4 ODC 1 MIT 13 EUPL 5 MPL 2 Eclipse Public License 1 Commercial licence 2 Other "Public domain" or "free use" without a (known) license 2 All rights reserved 0 5 Scientific use file Licence 0 Other: ToS, Internal licenses... 2 Custom license, chosen by user 2 ODbL 1 CERN Hardware licences 0 Open Government License 0 Licences for Other Digital Objects There were few respondents for this question, many of whom stated that they do not possess digital objects other than those already discussed, or simply just provided a list of licences, which we summarise below (Table 23). Table 23. Licences for other digital objects (n=20) Licence Other digital objects Total CC0 CC0 4 FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 52 | Page Licence Other digital objects Total Creative Commons licences requiring attribution CC-BY or compliant 12 27 CC-BY-SA 6 CC-BY-NC 5 CC-BY-NC-SA 4 Software "libre" licence BSD 0 6 Apache 2 LGPL 1 AGPL 1 ODC 1 MIT 1 Commercial license 1 1 Other All rights reserved 0 3 Scientific use file Licence 0 Other: ToS, Internal licenses... 0 Custom license, chosen by user 1 ODbL 1 CERN Hardware licences 1 Open Government License 0 While the question prompted respondents to specify the type of digital objects to which a given licence is applied, few respondents gave information. Table 24 contains a summary of these responses. Table 24. Licences for named “other” digital objects. Object Licence(s) applied MorpheusML model descriptions CC-BY 4.0 Open Hardware CERN Open Hardware licences: CERN-OHL-P, CERN-OHL-W, CERN-OHL-S Grey literature Sometimes no licence, but everything we produce is CC-BY 4.0 FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 53 | Page Continuity The respondents were asked whether their repository has a continuity agreement with another organisation to ensure ongoing access to and preservation of their digital objects and metadata, and to indicate the implementation status of the agreement as shown in Figure 27. Figure 27. State of continuity agreements. If the respondents answered that they have implemented or are implementing a continuity agreement, they were further asked to specify in free text what type of organisation the agreement is with(see Figure 28). Most respondents mentioned public entities such as national infrastructures, consortia of research performing organizations (or a combination or variation). Only one respondent mentioned contracts with commercial entities. Figure 28. Stakeholders of continuity agreements. Funding Regarding funding, the respondents were asked: “To what extent does your repository have sustainable funding to carry out your mission?” As shown in Figure 29, the majority of respondents FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 54 | Page (n=29) indicated that they have sustainable funding for 2-5 years. However, 11 respondents answered that they were unsure about their repository’s funding and nine respondents indicated that their repository has funding for one year at a time. So, while many repositories have a moderate level of sustainable funding, there is still uncertainty and short-term funding among a notable portion of respondents. On the other hand, 19 respondents indicated they have sustainable funding extending beyond five years (i.e., nine responses for “6-10 years” and 10 responses for “More than 10 years”). Figure 29. Duration of sustainable funding for repositories. The following details were identified from the responses of participants who selected “Other” (see Figure 29). One respondent reported receiving joint funding from the government and the university for a period of 2-5 years, with the university committed to continuing support if government funding were reduced or discontinued. Two respondents described funding models based on annual budgeting through their host institutions, without clearly allocated long-term resources. One repository is actively working toward a more sustainable model in collaboration with other national or international partner institutions, while another operates without any funding and is maintained on a pro bono basis. One noted that the repository is currently supported by new, dedicated hardware expected to meet operational needs for the next few years. However, future upgrades and additional staff may be required if data deposits or user support demands increase. One respondent mentioned that the funding is mainly private including sales of resources. Next, respondents were asked to specify the main funding sources of their repository, with the option to select all that applied (see Figure 30). Based on the responses of 76 participants, the majority (n=53) reported receiving funding from their host or partner institutions. The next most common sources of funding were structural funding (n=33) (i.e. central funding or contract from a research or infrastructure funder that is in the form of a longer-term, multi-year contract) and support through contract services or project funding (n=20) (i.e. charges for contract services to other parties or for research contracts), which indicates a moderate reliance on more temporary FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 55 | Page funding. Less commonly mentioned funding sources were data deposit fees (i.e. in the form of annual contracts with depositing institutions or per-deposit fees) and funding from access charges (i.e. charging for access to standard data or to value-added services and facilities) (n=3 each), which may indicate that user-based funding models are relatively uncommon. Five respondents were unsure of the main funding sources of their repository. Respondents could also indicate other funding sources that their repository has. Figure 30. Main funding sources of repositories. Four respondents indicated their repository’s funding source as “Other” (see Figure 30). Three of these respondents refer to national public funding (national research infrastructure, Ministry of Education and Culture, state budget). One respondent states that the process of receiving payments through annual contracts with the relevant depositing institutions is currently in the planning or preparation phase. Technology and Security Like for Digital Object Management and Organizational Infrastructure, respondents were asked to indicate the difficulties and needs they encounter regarding Technology and Security on a controlled matrix : on the y-axis, there was a list of five activities and functions derived from the TTRAM, and on the x-axis, a 5-point scale ranging from “no challenges” to “critical challenges”. Following this, respondents were again presented with an open-ended question, asking them to list the primary challenges they encounter. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 56 | Page Figure 31 shows for each of the five derived activities and functions40 how responses were distributed on a five-point-scale, illustrating the respondents’ assessment of the severity that the particular A|F poses in relation to technology and security. Figure 31. Challenges and needs that repositories face in terms of technology and security. While no single subject clearly stands out among the entire group of respondents, the overall impression is that of a cautiously concerned community. The slight dominance of the “Sensitive data management” topic is worth noting, although there is a lack of significant comments from the respondents who selected the “critical” or “significant challenges” options. “Tech watch” is another topic which seems to raise concerns. These concerns can be associated with issues expressed in the free text comments about keeping the repositories up to date and aligned with changing technical requirements. The number of mentions for each of the five TTRAM A|F considered here is shown in Figure 32, with “Other” receiving the most mentions. 40 Note that “Sensitive data management” and “Tech watch” are not standalone TTRAM A|F FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 57 | Page Reference document (6) We found several mentions of internal, public documents constituting a summary of the practices of the repository. These documents would typically be publicly accessible and cited in the course of a CoreTrustSeal application: ● Disaster recovery plan (2) ● File naming (1) ● File types (1) ● Long-term data preservation reference document (1) ● Preservation plan (1) People & Expertise : Staff training (3) ● Staff training: IT security (mandatory) (1) ● Staff training: Records management and archives (1) ● Staff training or certification: FitSM (1) Other ● Data deposits Compliance Checks (1) FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 64 | Page 3.3 Legal challenges Legal challenges pose a significant barrier to effective data sharing between repositories and organizations. To gain deeper insight into these issues, the survey included three key questions: the legal status of the repositories, the specific challenges they encounter, and their willingness to collaborate on addressing these challenges. An additional set of questions further explored the maturity and variation in legal frameworks across repositories. Of the respondents, 81 indicated that their repository is part of a research performing organization (see Figure 35), such as a university or research institute. This indicates that the majority are tied to established academic or research institutions, meaning that they have access to legal and administrative support, but also that they may be subject to stricter compliance requirements. It will be important that the Network focuses on addressing the needs and legal challenges specific to participants within these institutional frameworks. Figure 35. What is the legal status of your repository? In response to the question about the biggest legal challenges for data sharing, there is a fairly even distribution among the options (see Figure 36). However, the most commonly reported issues were Copyright and Intellectual Property Rights (75) and GDPR (65), followed by data governance and data security (53). These results are aligned with those of another question, where many respondents indicated that their institutions hold expertise in these domains. However, qualitative responses revealed inconsistencies in the implementation and operationalization of this expertise, indicating a need for improved policy alignment and practical guidance. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 65 | Page A recurring comment from respondents highlights a critical aspect of GDPR compliance: the responsibility for ensuring data compliance falls on researchers before the data is submitted to repositories. This means that repositories are not solely accountable for GDPR compliance, as the primary responsibility lies with the researchers who generate and handle the data prior to submission. However, repositories still play an important role in providing clear guidelines, ensuring that their processes align with legal requirements, and supporting researchers in understanding and adhering to GDPR standards. Figure 36. What would you say are the biggest legal challenges your repository faces when sharing data between repositories or with scientists from other research organisations? When asked about collaborating on frameworks for GDPR and Intellectual Property Rights, 44 respondents expressed interest, 22 declined, and 7 cited lack of time despite their interest (see Figure 37). While there is clear support for collective solutions, practical constraints may limit participation. However, the Network should facilitate the establishment of working groups focusing on these legal issues. Figure 37. Would your repository be interested in collaborating on frameworks and guidelines regarding GDPR and/or Intellectual Property Rights with others in the Network? FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 66 | Page Additional legal questions Following the three initial questions, respondents were asked if they wished to answer a set of additional questions regarding legal challenges. 24 respondents chose to answer additional questions. Of the respondents, 38 reported having clear GDPR-compliant frameworks, while 12 did not (Figure 38). This highlights the need for network collaboration to share best practices and ensure all members meet a common minimum standard. Figure 38. Does your repository have clear legal frameworks, policies and guidelines for handling personal data in accordance with GDPR and other relevant data protection laws? Of the respondents, 36 indicated that their repository has clear frameworks for sharing data within Europe, while 14 do not (see Figure 39). Similar to the previous question, this highlights an opportunity for the Network to facilitate the reuse of existing frameworks and establish working groups to support members in developing and implementing effective guidelines. Figure 39. Does your repository have clear frameworks and guidelines for sharing data within Europe? When asked whether their repository follows specific legal metadata standards for data governance and compliance (Figure 40), respondents most frequently indicated the use of Dublin Core (36), FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 67 | Page followed by DDI (14). PREMIS and DCAT were less commonly used indicating that these are used in more specialised contexts. This highlights an opportunity for the Network to explore how metadata standards are applied, and a discussion on how they can promote greater interoperability. Figure 40. Does your repository follow one or several specific legal metadata standards for data governance and compliance? Out of 40 respondents, 32 expressed openness to adopting or changing metadata schemas to improve legal interoperability (see Figure 41), highlighting strong community interest in enhancing interoperability and aligning with common standards. Figure 41. Would your repository be open to change to/adopt/add another core metadata schema if that would improve legal interoperability? FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 68 | Page 3.4 Training and support needs FIDELIS will offer different support and training mechanisms throughout the project. Moreover, training and peer-to-peer support is envisioned as an essential part of the long-term future of the FIDELIS Network. To craft a training and support strategy in line with the community needs, the survey included questions regarding topics for training as well as other characteristics such as format and delivery. Respondents were asked where they find and/or receive training and support on topics related to Trustworthy Digital Repositories (see Figure 42). The most frequently selected options were specific communities (with CESSDA, ELIXIR, RDA all mentioned in a free-text follow-up question), training initiatives from European projects (such as EOSC-Nordic42, FAIRsFAIR43, FAIR-IMPACT44), and online training offerings. These responses indicate rich sources of existing training initiatives and materials that FIDELIS should not aim to duplicate, but rather to highlight, enrich, or collaborate with. Interestingly, receiving peer support from colleagues at the host institution and from (international) peers in other institutions were both also often selected. This shows an existing habit of peer-to-peer support in the community, which FIDELIS can harness and uplift with our Network initiative. Figure 42. Where do you currently (or have you previously) find/receive training on topics related to Trustworthy Digital Repositories? 44 https://fair-impact.eu/ 43 https://fairsfair.eu/ 42 https://eosc-nordic.eu/ FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 69 | Page Of specific interest are both the topics and the formats in which the community would like to receive training and support. For topics, respondents were asked separate questions about what they currently experience and what their desires are for the future. The data shows that most training and support has already been received on topics related to FAIR-enabling qualities, persistent identifiers, licensing and reuse, and FAIR assessment (see Figure 43, the blue bars). On average, repositories have already received training and support on 4.5 different topics, with answers varying from one to thirteen different topics. Other topics that respondents already had received training and support on ranged from train-the-trainer capability building to different research skills. Respondents were asked whether the training and support they previously received was sufficient for their needs. Of the 46 responses received, 28 were generally satisfied with the training and support received. However, in many cases these answers also included the additional experience that there are many different topics of relevance for repositories and important tools, standards, and practices develop very quickly, making it difficult for training to remain sufficient for a longer period of time. Repositories with fewer staff and less effort experience challenges in dedicating time to training. When these repositories do dedicate time, it is hard to improve the practical skills of their staff when training is focused on high-level or theoretical topics. When we examine respondents’ interests for receiving (more) training (see the orange bars in Figure 43), development and sustainability of a repository ranks ties with long-term preservation as the leading options on 81 responses each. Interest in long-term preservation training matches the finding presented in section 3.1 that many repositories are trying to implement active preservation in their repository. Other topics that many respondents were interested in include repository certification, the development and sustainability of a repository, and access conditions and sensitive data. The largest gaps between having received training and holding an interest in training were found with persistent identifiers (87 to 43) and FAIR-enabling (103 to 67). In these cases it seems there has been sufficient training with not much interest for (more) training in that area. Other topics that were suggested by respondents included metadata standards, repository interoperability, and repository governance and management. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 70 | Page Figure 43. Received (blue) and desired (orange) training and support topics. Seminars and webinars are the leading format for training and support respondents have experienced. The high prevalence of peer support already noted above can also be observed here, with many respondents indicating that they have received peer support to learn (new) skills. It provided the largest gap between a support format previously experienced and of interest, a difference of 41 respondents. Personal mentoring, open hours, and financial support have been experienced least often. When asked which formats of training and support respondents would like to receive, seminars, webinars (106), and interactive workshops (97) led the responses (see Figure 44). Expert guidance and peer support were also valued highly, with 81 interested in this. Sixty-three respondents would like to receive financial support, whilst only 19 had received this before. This was the largest gap between interest in a form of support and having received this form of support. Personal mentoring and dedicated time to work independently were indicated to be the support approaches with the least support, though still indicated with interest by around 40 respondents. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 71 | Page Figure 44. Received (blue) and desired (orange) training and support formats. Respondents were asked about the importance of different qualities and characteristics of training and support in two sets of statements. In this first set of answers (see Figure 45), it can be seen that achieving a specific objective or outcome by the end of a training is valued by most respondents (116 agree or strongly agree). This matches the previous finding that higher-level and theoretical training often do not feel sufficient to justify the effort spent on the activity. Other highly valued characteristics, are when training and support include opportunities for peer engagement (97 agree or strongly agree), when training and support is broken up into multiple shorter activities instead of one longer activity (83 agree or strongly agree), and when learners can train at their own pace (76 agree or strongly agree). Another positive result that can be observed is that many respondents indicate that their organisation encourages them to join training and support activities (98 agree or strongly agree). Responses shift more towards indifference or disagreement when it comes to creating a tangible product in training, receiving digital badges or formally recognised certificates for training, and learning on your own. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 72 | Page Figure 45. Desirability judgement of some training and support characteristics. In the next set of questions, respondents were asked to indicate the importance of a set of statements about different characteristics of training and support (see Figure 46). There is a general trend of emphasis on the importance of all elements. Deemed most important are the ability to connect with, learn from, and network with fellow participants, organisers, (guest) speakers, and other experts during the activity, and for the activity to be at a more advanced level. This aligns well with previously observed answers where peer support and granular, specific, and practically-focused training and support is most desired. Respondents also find it important that the training does not take up too much of their time and that it covers new and cutting-edge topics, following closely the quickly-developing landscape. Slightly less important are the ability to join an activity with a team of colleagues, which could be due to the fact that not all repositories have bigger teams of people on their staff for this to be applicable. A mix of different teaching formats is also deemed somewhat less important, which matches with the previous findings on training formats, where information sharing seminars and webinars were most valued. Beginner level training was still deemed important by 82 of the respondents, with around thirty respondents indicating this is not what they are looking for at all. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 73 | Page 4 Conclusions and next steps The aim of the FIDELIS landscape survey was to collect community input that would inform the different parts of the project and the building of the FIDELIS Network for Trustworthy Digital Repositories. Considering the results from the previous sections, this last chapter describes how FIDELIS will take the findings on board in its developments. Repository capabilities and standards The survey results confirm that the repository landscape is very diverse. Sometimes this diversity can be explained by specific disciplinary, governance or other factors; sometimes there are possibilities to converge practices, streamline processes, leverage shared resources and even achieve cost efficiencies. General information on this variety combined with the detailed knowledge about the actual repository practices, as well as the challenges that repositories are facing, will be used by FIDELIS Pillar 3 Trustworthy Digital Repository Harmonisation and Interoperability to plan for the next two years. The survey results provide information about the gaps where FIDELIS can provide recommendations or solutions, and enable Pillar 3 to identify and select standards, frameworks, policies and practices that can be used, established or developed further to enhance interoperability and federation in the context of EOSC. Common areas for cooperation and alignment between the repositories in the Network can range from common policy templates to reducing burden, challenges, and risks through shared services. The survey results will also be used to improve the TTRAM model and its activities and functions, and to inform our work on a tiered approach to capability and maturity of repositories. In terms of repository capabilities and standards, the survey analysis revealed the following key conclusions: 1. Curation, Preservation, and Metadata Quality: These areas emerged as critical challenges, with repositories facing resource constraints (particularly in terms of qualified personnel), desirable but scarcely present automation, and a lack of standardized practices. Metadata quality and standardization are particularly pressing concerns for enhancing discovery and long-term usability. 2. Interoperability and Semantic Alignment: Repositories face challenges in integrating external platforms, aligning metadata standards, and synchronizing metadata records. While common metadata schemas and persistent identifiers (PIDs) are widely adopted, there is a clear need for richer, multilingual metadata to enhance discoverability and usability. Additionally, repositories often embed or hardcode semantic artefacts within their systems rather than integrating with centralized semantic artefact catalogues. This limits FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 80 | Page standardization and complicates the management of artefact complexity, such as versioning and heterogeneity. Addressing these issues will be critical for fostering interoperability and enabling seamless networking across repositories. 3. Governance and Sustainability: Governance fragmentation, funding instability, and legal compliance (e.g., GDPR) are significant difficulties. Repositories require clear governance models, sustainable funding mechanisms, and shared policies to ensure long-term success. 4. Technical Infrastructure: Robust, scalable infrastructure is essential for managing sensitive data, ensuring preservation compliance, and maintaining repository reliability. Non-commercial solutions and automated integrity checks are particularly needed. Legal Challenges A significant outcome of the survey is the fragmented legal landscape within which repositories operate. Many repositories lack formal frameworks for GDPR compliance or cross-border data sharing, and responsibility for legal compliance is often delegated to researchers without sufficient institutional support. Intellectual property rights (IPR) and inconsistent licensing practices also emerged as recurring challenges, limiting interoperability and data reuse. While a few respondents described promising workflows involving intake appraisal, consent templates, and use of open licences, the majority indicated a lack of formalised processes for legal interoperability. The FIDELIS Network can provide valuable support by promoting shared legal templates, facilitating peer-to-peer learning, and coordinating working groups on GDPR, licensing, and metadata standards. However, the success of these initiatives will depend on the willingness of members to actively contribute to shared solutions and engage in collaborative development. Creating a sustainable, legally interoperable data-sharing ecosystem will require collective investment from across the community. Repository Training & Support The Repository Training & Support pillar will use the findings from the survey to help create our training strategy. This informs choices around the design and formats of the training offered, the topics covered, and how to communicate effectively with the community about the activities on offer. Aside from the topics indicated in the questions in the dedicated section, we also take away interesting findings from the other sections to potentially craft training and support around important standards and challenges experienced in the different topics. Such training and support efforts will be a collaborative effort with the other parts of the project and potentially external projects and organisations. A clear desire for more advanced, practical, and granular activities was expressed across different parts of the survey, which we will take on board in designing our training offerings. Seminars, webinars, and workshops were most appreciated and respondents’ desires for technical and repository management topics map well to the areas of focus that we already envisaged for our training and support activities. FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 81 | Page Peer support was highly valued and desired, while there was also some uncertainty around the mentoring programme where this type of support will be showcased most. It is possible that these uncertainties originate from a lack of understanding about what mentoring entails. This could be alleviated by explicit information and instruction around the type of support, which we will be able to provide on our website during the launch of the call. The survey has been useful in building our awareness of some of the worries and assumptions that we should address in advertising the call. In general, it is important that we provide clear, extensive, and complete information around the support offered to allow repositories to accurately judge the value they will get out of each activity, compared to the expected time and effort to spend. The survey results also highlighted again the existence within the repository landscape of a wealth of existing training resources as well as communities which already offer training and support to repositories. Harnessing what already exists instead of duplicating effort is essential to meet the current needs of our potential members. Contributing to the rich landscape through creation and collaboration was the aim of FIDELIS from the outset, but the survey answers have pointed to some specific communities and resources to investigate. When using the results of the survey to inform the drafting of the FIDELIS training strategy, we should remember that the responses were skewed towards more “EOSC-ready”47 countries. It will be important to continue to consider the other perspectives in the community in the development of the training strategy, for example, repositories that have a need for beginner-level training, and the opportunities that are created by financial support for repositories with limited resources. FIDELIS is committed to targeting less developed repositories and countries where lower rates of certification are attained to enable them to become active members of our Network. Expectations of the FIDELIS Network The final part of the survey returned results that inform us about community expectations regarding the FIDELIS Network. Respondents’ answers showed a general enthusiasm for the potential benefits of the Network, and contributed a high number of ideas for such benefits, collaboration mechanisms, and topics of relevance. The already-stated aims of the FIDELIS Network align well with the survey respondents’ desire for connection, advocacy, and knowledge exchange. Respondents stated that they believed that the most effective way to establish the Network would be through granular and practical work on specific topics with tangible results, which can then be endorsed or adopted throughout the Network. Communicating through, for example, a forum, alongside opportunities to attend (online) events and meetings is most highly appreciated. Repositories deal with barriers to participation in a membership-based network, most notably if there is a membership fee. FIDELIS’ proposed free provisional membership provides a useful intermediate solution for members to participate in the creation of the Network. 47 https://www.eoscobservatory.eu/eoscreadiness/2023/practices/nationalMonitoring/all FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 82 | Page This report, the data, and the relevant analysis scripts will become available for public reuse. The outcomes will also be shared with the EOSC EDEN project48, relevant EOSC Association Task Forces49, existing and forthcoming EOSC Nodes50, and other relevant partners in the EOSC ecosystem to share this extensive snapshot of information on the current repository landscape. These different initiatives may use these insights to inform their own activities and areas of focus for the future. 50 https://eosc.eu/eosc-about/building-the-eosc-federation/ 49 https://eosc.eu/eosc-association/eosc-task-forces/ 48 https://cordis.europa.eu/project/id/101188015 FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 83 | Page 5 Appendices Appendix A - Survey glossary This glossary accompanied the survey to support respondents in understanding the terms used. Access: Activities/Functions that include determining the appropriate access routes (e.g. download, secure remote environment, etc.) and mediating the process of providing access to users. Access activities/functions are defined by digital object characteristics and user characteristics that comply with Rights Management (see below). (Adapted from L’Hours & Bell 2023) Activity/Function: An activity or function provided by a (research) repository. In this survey, Activities and Functions are grouped into three thematic areas: Digital Object Management, Organisational Infrastructure, and Technology and Security. Analysis & Impact: Activities/Functions that include the analysis of internal business information and externally available information to ensure and demonstrate that the organisational entity is fulfilling its mission (see Mission & Scope). (Adapted from L’Hours & Bell 2023) Archive: Curated collection or repository containing physical or digital static records, objects, metadata and data deemed suitable for permanent retention, set up and managed to established standards and models, such as ISAD(G), CoreTrustSeal, and the OAIS reference model, that ensure long term integrity, security, authenticity and accessibility of the records, objects, metadata and data. (Source: CODATA Research Data Management Terminology. Archive. https://terms.codata.org/rdmt/archive-noun) Authenticity: The degree to which a digital object is what it appears to be, with evidence to prove it. Also see Provenance. Catalogue: Index describing, indicating the location of, and recording other details of resources, materials works, etc. Curated and organised using a formal metadata schema such as MARC, ISAD(G), Dublin Core, DataCite etc. (Source: CODATA Research Data Management Terminology. Catalogue. https://terms.codata.org/rdmt/catalogue) Community Watch: Activities/Functions that cover engagement with (meta)data reusers to ensure their current needs (see ReUse above) and future needs (see Preservation) are met. (Adapted from L’Hours & Bell 2023) Conceive, Create, Collect: Activities/Functions that cover varied pre-repository phases of the research project lifecycle, e.g., engaging with (meta)data producers and (meta)data owners during FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 84 | Page the research project planning phase - whether or not they become (meta)data depositors - to help them improve the quality of deposited digital objects and communicate the value of services, including preservation and reuse. (Adapted from L’Hours & Bell 2023) Continuity of Service: Activities/Functions that ensure defined business continuity in line with the expected ‘business as usual’. This includes all non-technical organisational, digital object management (see ‘workflows’) and security processes. It also encompasses provisions to recover from interruptions to the business, including disasters, as well as plans for succession when a service is discontinued. (Adapted from L’Hours & Bell 2023) Controlled Vocabulary: List of standardised terminology, words, or phrases, used for indexing or content analysis and information retrieval, usually in a defined information domain. (Source: CODATA Research Data Management Terminology. Controlled vocabulary. https://terms.codata.org/rdmt/controlled-vocabulary) Criteria, Assessment, Improvement: Activities/Functions that manage the processes by which internal compliance with criteria (including standards, policies and standard operating procedures) is assessed, and the measures undertaken to ensure continuous, targeted improvement. (Adapted from L’Hours & Bell 2023) Curation, Quality & Compliance: Activities/Functions that cover steps taken by your repository to ensure deposited digital objects reach a defined level of quality and standards compliance before they are made available for reuse. (Adapted from L’Hours & Bell 2023) Curation: Managing and promoting the use of assets from their point of creation to ensure that they are fit for contemporary purpose and available for discovery and reuse. For dynamic datasets this may mean continuous enrichment or updating to keep them fit for purpose. Higher levels of curation will also involve links with annotation and with other published materials. (Source: CODATA Research Data Management Terminology. Curation. https://terms.codata.org/rdmt/curation) Data Compliance: Ongoing processes to ensure adherence of data to both enterprise business rules (government department, university, industry, or agency), and to legal, regulatory and accreditation requirements. Includes five areas: controls, audit, legal compliance, regulatory compliance, and accreditation conformance. (Source: CODATA Research Data Management Terminology. Data compliance. https://terms.codata.org/rdmt/data-compliance) Data Quality: Reliability and application efficiency of data. Perception or assessment of a dataset's fitness to serve its purpose in a given context. Aspects of data quality include: Accuracy, Completeness, Update status, Relevance, Consistency across data sources, Reliability, Appropriate presentation, Accessibility. Data quality is affected by the way data are entered, stored and managed. Maintaining data quality requires going through the data periodically and scrubbing it. Typically this involves updating, standardising, and de-duplicating records to create a single view of the data, even FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 85 | Page if it is stored in multiple disparate systems. (Source: CODATA Research Data Management Terminology. Data quality. https://terms.codata.org/rdmt/data-quality) Deposit & Appraisal: Activities/Functions that cover the phase where the custody of digital objects is transferred from depositor to repository. Digital objects that are offered or requested for deposit may be appraised to ensure they meet defined criteria. (Adapted from L’Hours & Bell 2023) Discovery & Identification: Activities/Functions that cover the application of persistent identifiers and associated resource discovery metadata to digital objects, the provision of resource discovery systems, and providing harvestable metadata to other resource discovery systems. (Adapted from L’Hours & Bell 2023) External Engagement: Activities/Functions that include the means by which the organisational entity engages with external actors, including individuals, organisations, funders, partners and other bodies. Specialist engagement is required for Conceive, Create Collect, Deposit & Appraisal, Access, Support, and Training activities. (Adapted from L’Hours & Bell 2023) Governance: Activities/Functions that cover the organisational hierarchies and processes by which the entity’s mission is managed and executed. (Adapted from L’Hours & Bell 2023) Interoperability: Activities/Functions that manage the protocols and processes by which people, processes, technologies and digital objects effectively interact with others, both within and across organisational entities. (Adapted from L’Hours & Bell 2023) Legal & Ethical: Activities/Functions that cover the framework of legal and ethical requirements that the organisational entity must be aware of and comply with. (Adapted from L’Hours & Bell 2023) Mission & Scope: Activities/Functions that are about defining and communicating the purpose (mission) of the organisational entity and the boundary (scope) of the activities it is responsible for. These activities may involve partners or outsourcing but your repository or data service is the entity that takes ultimate responsibility for the actions and outcomes. (Adapted from L’Hours & Bell 2023) Network: The planned FIDELIS Network of trustworthy digital repositories (TDRs). Ontology: Shared and standardised list of words, terms and phrases to describe components of a particular discipline or domain, along with a taxonomy of their relations. Compare this to a controlled vocabularies, which tend not to include a structure of relations between their terms. Ontologies are typically developed by domain-specific institutions or communities to aid in the precise referencing of elements. (Source: CODATA Research Data Management Terminology. Ontology. https://terms.codata.org/rdmt/ontology) FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 86 | Page People & Expertise: Activities/Functions that cover the management of human resources to fulfil the mission. Includes ensuring that sufficient skills are available, internally or externally (see External Engagement). (Adapted from L’Hours & Bell 2023) Policy & Standards Management: Activities/Functions that cover the adoption and creation of policies and standards to guide practice (see Governance and Workflows). Complies with the Legal and Ethical framework in place. (Adapted from L’Hours & Bell 2023) Preservation: Activities/Functions that include monitoring the technology landscape (see Technical Infrastructure below) and the user community (see Reuse above and External Engagement below) for changes that affect the use or understanding of digital objects. If necessary, preservation actions are taken on the data (e.g., format migration), or on the metadata (e.g., updated ontologies), or on the whole object (e.g., emulation of the environment in which the digital object is rendered and used). These preservation actions then ensure preservation outcomes i.e. the continued viability of the digital object. (Adapted from L’Hours & Bell 2023) Preservation: An activity within archiving in which specific items of data are maintained over time so that they can still be accessed and understood through changes in technology. (Source: CODATA Research Data Management Terminology. Preservation. https://terms.codata.org/rdmt/preservation) Provenance & Authenticity: Activities/Functions that ensure that information is provided about the digital objects that describes their history (provenance) and provides evidence that the digital objects are what they claim to be (authenticity). Provenance and authenticity information can be sourced at the point of deposit, generated during curation and preservation actions, and shared with (meta)data users. (Adapted from L’Hours & Bell 2023) Provenance: A type of historical information or metadata about the origin, location or the source of something, or the history of the ownership or location of an object or resource including digital objects. For example, information about the Principal Investigator who recorded the data, and the information concerning its storage, handling, and migration. (Source: CODATA Research Data Management Terminology. Provenance. https://terms.codata.org/rdmt/provenance) Registry: Database containing information about trusted repositories that are provided by repository managers and are useful for human and machine users. These registries do not contain information about all metadata descriptions of digital objects, nor do they offer a list of PIDs of all stored digital objects. They do offer information based on standardised types on how to retrieve such information (e.g., the port under which OAI-PMH can be accessed to offer metadata). A registry requires the assignment of a permanent, unique and unambiguous identifier to each item. (Source: CODATA Research Data Management Terminology. https://terms.codata.org/rdmt/registry) Release & Publishing: Activities/Functions that manage the processes by which information - including papers, policies and other content - is released from your repository or data service entity. This is a ‘push’ activity by the entity, in contrast to the ‘pull’ process by which users request Access to FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 87 | Page digital objects. It includes release and publishing of the information necessary for External Engagement, and to provide evidence during external assessment (see Criteria, Assessment, Improvement). (Adapted from L’Hours & Bell 2023) Repository: In this survey used to include research repositories and archives as well as repository and archive services. Repository: Physical or digital storage location that can house, preserve, manage, and provide access to many types of digital and physical materials in a variety of formats. Materials in online repositories are curated to enable search, discovery, and reuse. There must be sufficient control for the physical and digital material to be authentic, reliable, accessible and usable on a continuing basis. (Source: CODATA Research Data Management Terminology. https://terms.codata.org/rdmt/repository) Research & Development: Activities/Functions that include any work, internal or external, often delivered through projects, that falls outside ‘business as usual’. Includes - but is not limited to - work that identifies and develops new and novel approaches to data, metadata, business processes, technology, security and research infrastructure. Where assessment indicates a need for improvement that falls outside of normal maintenance upgrades, this may be delivered through Research and Development (R&D). (Adapted from L’Hours & Bell 2023) Resources: Activities/Functions that cover the organisational hierarchies and processes by which the organisational entity’s human and financial resources are managed. This includes a knowledge of and an effective deployment of human, financial and energy assets. (Adapted from L’Hours & Bell 2023) Reuse: Activities/Functions that ensure that the outcome of Deposit & Appraisal, Curation, Quality & Compliance, Access, and Preservation activities/functions result in digital objects being able to be used and understood by the (meta)data users for as long as your repository or service has promised. This includes ensuring that digital objects are furnished with sufficient information to support understanding and use over time. In cases where (a copy of) the digital object is not handed entirely over to the user (e.g., direct download to a researcher’s computer), your repository or data service mediates reuse. This mediation includes the provision of remote secure access systems, safe rooms or other tools where the (meta)data remains fully or partially under the control of the service provider. Supporting reuse depends on an understanding of the community of users including, but not limited to, targeted ‘designated communities’. This understanding is gained through External Engagement (see below). (Adapted from L’Hours & Bell 2023) Rights Management: Activities/Functions that cover the management of the permissions, prohibitions and obligations related to the interactions between digital objects and actors (such as individuals and organisations), both inside and outside your repository/data service. (Adapted from L’Hours & Bell 2023) FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 88 | Page Security: Activities/Functions that include measures that ensure the appropriate physical and virtual protection of the digital objects held by the entity and of other related information including staff and user information. (Adapted from L’Hours & Bell 2023) Semantic Artefact: A machine-actionable and -readable formalisation of a conceptualisation, enabling sharing and reuse by humans and machines. These artefacts may have a broad range of formalisation, from loose sets of terms, taxonomies, thesauri to higher-order logics. Moreover, semantic artefacts are serialised using a variety of digital representation formats, e.g., RDF Turtle, and OWL, using XML (RDF) and JSON-LD. (Source: Franc et al. 2022) Here also used as a broader term to include controlled vocabularies, ontologies, terminologies, taxonomies, thesauri, metadata schemas and standards. Semantic Artefact Catalogue: A broader term to include libraries, registries, listing or repositories of semantic artefacts and also platforms often named terminology/vocabulary service/server. Storage & Integrity: Activities/Functions that cover the means by which digital objects data and metadata are stored, persisted and replicated, and by which replicated copies are validated as identical, or restored from a backup in the event of errors. (Adapted from L’Hours & Bell 2023) Taxonomy: A controlled vocabulary with a hierarchical structure used to classify things or concepts. Terms within a taxonomy have relations to other terms (parent/broader term, child/narrower term). (Source: Franc et al. 2022) Tech Watch: Activities/Functions that cover the process by which the wider landscape of hardware, software and information technology services are monitored. Drives changes to the provision of technology for delivering services internally and externally. Aligns with Community Watch and informs Curation and Preservation. (Adapted from L’Hours & Bell 2023) Technical Infrastructure: Activities/Functions that include the overall provision of hardware, software and information technology service management to support the entity. (Adapted from L’Hours & Bell 2023) Thesaurus: A controlled vocabulary following a standard structure, where all terms have relationships of three kinds to each other: hierarchical (broader term/narrower term), associative (related term), and equivalent (use/used for or see/ seen from). Some terms in thesauri might have additional explanatory notes, such as scope notes (brief explanations about the coverage of the term or of how it should be used in indexing) or history notes. Thesauri are defined in the ISO 25964. (Source: Franc et al. 2022) FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 89 | Page Purpose mentioned Mentions Services cited (number of mentions) Registries: licences 4 Creative Commons Licences Registry (3) Registries: organizations 4 EDMO (European Directory of Marine Organizations)(1), Research Information System of the institution (1), ROR (1), ROR, ORCID, European Directory of Marine Organisations (EDMO) (1) Registries: persons 4 ORCID (4), SURFconext (1) Dataset Assessment 3 BIDS Validator (1), F-UJI (1), Neurobagel (1) Semantic: geographical metadata 3 geonames (2) Open Street Map (1) Alignment of workflows 2 Common policies and guidelines for repository consortium (1), Common policies and guidelines with DataverseNO (1) Deposit 2 ORCID (1) Preservation service 2 APTrust (1), National Digital Preservation Service (PAS) (1) Project information 2 https://gepris.dfg.de/gepris/OCTOPUS (1), OpenAIRE (1) Registries: stations 2 https://gnss-metadata.eu/ (1), “IGS site log (standard to store GNSS station information - text based)” (1) Access content through a federation 1 AgroPortal/OntoPortal federation (1) Data archive 1 Aila Data Service (1) Data models 1 Neurobagel data model (1) Data repository 1 CrossAsia Search (based on PICA format, part of the GBV framework database) (1) Dataset identification 1 OpenAIRE (1) FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 96 | Page Purpose mentioned Mentions Services cited (number of mentions) Discovery 1 VerbundFDB (German Network of Educational Research Data) Search (1) DOI, diffusion and preservation 1 Nakala (1) Enable submission to users 1 ProteomeXchange/PRIDE standalone submission tool (1) Federated search for data across multiple data repositories of photon data 1 PaNOSC federated search portal (1) Metadata editor 1 COMEDI (1) Multiple purpose 1 Reposis Service of the GBV (1) None 1 None (every service is embedded) (1) Publication catalogues 1 EuropePubMed Central (1) Registries: persistent identifiers 1 Persistent Identifier (PID) Registries (1) REST API standardization 1 Optimade (1) Standardize descriptions in metadata 1 CESSDA vocabulary service (1) Storage 1 Fedora software (1) FIDELIS has received funding from the EU’s Horizon Europe research and innovation programme under Grant Agreement no. 101188078 97 | Page