ICSU World Data System Implementation Plan Trusted Data Services for Global Science

Transcription

1 ICSU World Data System Implementation Plan Trusted Data Services for Global Science WDS decided to adopt five Strategic Targets defined in the Strategic Plan to fulfil its remit and build worldwide multidisciplinary communities of excellence for scientific data services. The overarching objective is to promote data stewardship and support the provision of trustworthy data services for global science. The Scientific Committee is advancing these targets during the period by guiding the implementation of the WDS programme according to the plan laid out below. Working with its members and partners, WDS will endeavour to provide an operational Knowledge Network for use in disciplinary and multidisciplinary research programmes building on a Multidisciplinary Metadata Catalogue for discovery of WDS members datasets. A Global Registry of Trustworthy Digital Repositories and Services building on the WDS Certification will also form an important contribution to the research data infrastructure. WDS will continue to engage with ICSU Flagship programmes and build bridges between data communities across disciplines. Successful approaches and practices of WDS Members will be disseminated and adoption of standards promoted. WDS will work also with funders to ensure that the essential components of the research data infrastructure are sustainable. For each Strategic Target, the plan develops objectives and defines the required resources and risks where appropriate. The deliverables and milestones are also presented with target dates for delivery.

2 Table of Contents Table of Contents Strategic Target 1: Make trusted data services an integral part of international collaborative scientific research Objectives for ST1 Background for ST1 Goal a) Develop an operational Knowledge Network for use in disciplinary and multidisciplinary research programmes. Required Resources and Risks for ST1 Goal a) Background for ST1 Goal b) Engage with the flagship ICSU-sponsored Future Earth programme in collaboration with the Committee on Data for Science and Technology (CODATA). Required Resources and Risks ST1 Goal b) Deliverables/Milestones for ST1 Actions and Tasks for ST1 Strategic Target 2: Nurture active disciplinary and multidisciplinary scientific data services communities Objectives for ST2 Background for ST2 Required Resources and Risks for ST2 Deliverables/Milestones for ST2 Actions and Tasks for ST2 Strategic Target 3: Improve the funding environment for data services Objectives for ST3 Background for ST3 Deliverables/Milestones for ST3 Required Resources and Risks for ST3 Actions and Tasks for ST3 Strategic Target 4: Improve the trust in, and quality of, open scientific data services Required Resources and Risks for ST4 Deliverables/Milestones for ST4 Actions and Tasks for ST4 Strategic Target 5: Position ICSU-WDS as the premium global multidisciplinary network for quality assessed data Annexes Background Objectives for ST5 Required resources and Risks for ST5 Deliverables/Milestones for ST5 Actions and Tasks for ST5 Annex A: WDS Infrastructure Concept Design Considerations More About Relationships Leveraging Existing Infrastructure Annex B: World Data System/ Data Seal of Approval Maturity Model Concept 2

3 Strategic Target 1: Make trusted data services an integral part of international collaborative scientific research Objectives for ST1 a) Develop an operational Knowledge Network for use in disciplinary and multidisciplinary research programmes. This Knowledge Network is the umbrella of several services comprising WDS integrated infrastructure: i. A Metadata Catalogue. This will be built with WDS resources, and is recognized as a tool for discovery of datasets and that can be used in the context of research and exposes search facets that improve discovery across disparate repositories. ii. A Brokering Service that harvests metadata from external data centres and harmonizes metadata according to common documentation concepts (e.g., or ). iii. A Global Registry of Trustworthy Digital Repositories and Services. ICSU-WDS will take a leadership role in establishing this registry in collaboration with DataCite and re3data, and will involve the Research Data Alliance (RDA). The Global Registry will serve several different communities and users: Data producers needing to find an appropriate repository or service Science publishers needing to point authors to an appropriate repository or service Funding agencies needing to inform the data management planning process Data discovery both as direct pointers to data facilities for data seekers and as a directory to drive the integrated metadata catalogue process b) Engage with the flagship ICSU-sponsored Future Earth programme in collaboration with the Committee on Data for Science and Technology (CODATA). W DS and CODATA must establish a strong working relationship with the Future Earth Secretariat; in particular the Data and Observation Task Force which will take on a proactive role in Future Earth s data management. This Action is linked to WDS work with the Belmont Forum e-infrastructure and Data Management activity in Strategic Target 3 (see ST#3a.1). Background for ST1 Goal a) Develop an operational Knowledge Network for use in disciplinary and multidisciplinary research programmes. ICSU-WDS has to establish functioning infrastructure and services to support its objectives, and its mandate. ICSU-WDS has positioned itself as a critical component of emerging Research Data Infrastructure, specifically aiming to provide services in respect of 1. Aggregated metadata services based on the collections offered by Members. 2. Brokering services that harvest and harmonize metadata from disparate data sets 3. Knowledge Network services that align information contained in the aggregated metadata, as well as provide links to authoritative registries. 4. A Registry of Trusted Digital Repositories and Services, which includes the WDS membership as a subset. I CSU-WDS needs to establish funding for these services, for the following reasons: 1. Similar and synergistic efforts have been supported in the past as projects within funding programmes. But these programmes come to an end typically after 3 5 years after which the 3

4 services are often lost. ICSU-WDS must be in a position to take over and operate these services as and when required. 2. More critically, at present there are few systems that are built on the services and components that ICSU-WDS and similar organizations offer. However, this is likely to change radically in years to come. The risk to systems developers is greatly increased by uncertainty about the future availability of foundation services. This creates the need for a sustainable baseline operation within ICSU-WDS in essence, guaranteeing the availability of registry services that will allow third parties to develop value-added services and applications. In summary, ICSU-WDS will be establishing an infrastructure that is based one this: Note the following: WDS will commence the process by establishing prototypes for the following 4 components: a. A metadata harvesting and/ or brokering arrangement, which assists with the synchronization and maintenance of a WDS aggregated Metadata Catalogue. In future, this may be more than one physical implementation, hosted in all probability by WDS Members on our behalf. b. The Metadata Catalogue, which spans a variety of protocols and metadata standards, across multiple scientific disciplines. c. A Registry of Trusted Repositories and Services, established in collaboration with (DataCite/re3data/Databib). This Registry, if it also offers a Digital Object Identifier (DOI) for each repository, begins to form a critical part of the Linked Open Data services used not only by the WDS Knowledge Network but also other end-users. d. The WDS Knowledge Network. This can be done in two phases: A working implementation that is not directly based on the WDS aggregated Metadata Catalogue and the Registry of Trusted Digital Repositories and Services, but aims to provide a proof of concept and integration test. An operationalization phase, utilizing those two services. 4

5 Required Resources and Risks for ST1 Goal a) Estimates of development time required to establish each component, together with potential collaborations and status of funding options: Infrastructure Component Metadata Catalogue Resource Estimate 1 6 personmonths from WDS Contributors Proposals/ Resources Risks NICT PANGAEA 2 ESSI-Lab 3 SAEON 4 NICT (in kind) PANGAEA to host the service on behalf of WDS. SAEON to contribute brokering components and a brokering framework Resources required to manage in-kind contributions across time zones and cultural differences Registry of Trusted Repositories and Services 4 personmonths from WDS PANGAEA 5 WDS-IPO SAEON DataCite re3data In-kind contributions from participants to develop a specification for re3data integration with WDS membership data. Knowledge Network 6 personmonths from WDS Data Brokering 6 personmonths SAEON/ DIRISA NICT/ISDS SAEON/ DIRISA ESSI-Lab PANGAEA EarthCube NICT (in kind) SAEON (in kind) Prototype development by DIRISA and SAEON NICT to consider operational implementation based on the prototype. In-kind contributions from participants in RDA Prototype development by DIRISA and SAEON These components will require a hosting organization each, or need to be hosted by the WDS International Programme Office (WDS-IPO). Depending on the decision in this respect, the development and delivery processes will be different. Irrespective of the decision, the maintenance and support requirements have to be included in the long-term costs. For third parties to become reliant on the infrastructure, this horizon should be of the order of 3 5 years. 1 To establish a working prototype and reference implementation 2 Earth Science and Space Informatics Laboratory 3 South African Environmental Observation Network 4 National Institute of Information and Communications Technology 5 WDS International Programme Office 5

6 Estimate of maintenance and support requirements: 1.5 resources per annum, available full time and part-time. Background for ST1 Goal b) Engage with the flagship ICSU-sponsored Future Earth programme in collaboration with the Committee on Data for Science and Technology (CODATA). ICSU-WDS needs to engage with the flagship ICSU-sponsored Future Earth programme in collaboration with the Committee on Data for Science and Technology (CODATA). The goal of such engagement is twofold: 1. To avoid negative outcomes, particularly illustrated by the International Polar Year , where a large number of especially smaller datasets derived from an international programme are thought to be inaccessible or poorly curated. 2. To maximise the benefit to researchers in particular and society in general by making datasets available in a way that will maximise interoperability options and stimulate re-use and value addition. In particular, the following objectives should be pursued: 1. WDS and CODATA must establish a strong working relationship with the Future Earth Secretariat; in particular the Data and Observations Task Force which will take on a proactive role in Future Earth s data management. 2. A consensus on data management guidance should be sought within this partnership, culminating in firm advice in relation to data management plans, deposit and publication via trusted digital repositories, licensing, and subsequent application. 3. The partnership should endorse, and maximise the re-use of the revised GEO Data Management Principles, currently in preparation, and widen the scope to include domains outside Earth and Environmental Observation, with particular focus on health and social science aspects. 4. It should also maximise the re-use of appropriate Research Data Alliance outputs. Synthesis of coherent guidance is not a simple task, since there are a wide variety of choices and considerations presented to individual researchers. Hence the outputs of this action are likely to include at least the following: 1. A high-level, largely non-technical overview for researchers and data centre managers that places available resources (open source software, international standards and conventions, services for vocabularies and authoritative registries, guidance from bodies such as GEO and RDA) into context, and maps out how these resources are relevant and should be used when compiling a data management plan. 2. Compiling a registry of appropriate guidance, and a blueprint of what appropriate guidance should cover. The only scalable option is to align and stimulate community contributions and to make such contributions easily accessible. An example can be found in work undertaken in 2014 by GEO-BON. Required Resources and Risks ST1 Goal b) Deliverable Resource Estimate Contributors Proposals/ Resources Risks Agreement with Future Earth, CODATA on collaboration Not significant WDS-IPO WDS-SC Chair CODATA WDS-IPO relying on In-kind contributions 6

7 Consensus on Data Management Less than 1 person-month each WDS-IPO WDS-SC CODATA GEO In-kind contributions f rom participants relying on In-kind contributions Lack of Consensus Develop a Review on Data Management 1 person-month each WDS-IPO WDS-SC Future Earth CODATA WDS Members In-kind contributions from participants relying on In-kind contributions Compile a Registry of Guidance 6 person-months WDS-IPO WDS-SC Future Earth CODATA Belmont Forum e-infrastructure and Data Management Availability of funding delayed Deliverables/Milestones for ST1 ST#1a Developing WDS Knowledge Network ST#1a.1) Demonstrate the prototype WDS Knowledge Network to the Belmont Forum in an attempt to obtain funding for the full operational version ( by June 2016 ) ST#1a.2) Take up the in kind funding offer from the National Institute of Information and Communications Technology, and approach RDA and the South African National Research Foundation for additional funding ( by September 2016 ) ST#1a.3) Start development of an operational Knowledge Network built upon the in-progress incarnations of the other component services. ( Prototype by June 2016, Operational KN by September 2016, Full system by December 2016 ) ST#1a(i) Expanding/enriching WDS Metadata Catalogue ST#1a.4) Get all feasible metadata into the Metadata Catalogue and contact those that still do not have a catalogue. Progress has been slow and transition to Issue Tracker System to reinitiate work. Collaboration with OpenAIRE should help (see ST#1a.6) ( by July 2016 ) ST#1a.5) Release an as is version of the Metadata Catalogue and demonstrate at IDW/SciDataCon 2016 ( by September 2016 ) ST#1a(ii) Developing Brokering Service ST#1a.6) Strengthen links with OpenAIRE to become WDS member and extend the DLI framework to WDS members. In parallel, continue to explore a contribution from the National Research Council of Italy, to extend GI-cat to suit WDS Metadata Catalogue requirements ( by June 2016 ) ST#1a.7) Collaborate with the RDA Brokering Working Group in which ICSU-WDS is listed as a use case ( by June 2016 ) and establish a new RDA working group to discuss and agree a framework for brokering components ( by September 2017 ) ST#1a(iii) Building Global Registry for Trusted Digital Repositories and Services ST#1a.8) Continue efforts through WDS membership within DataCite and IPO to work with DataCite/re3data, and use their service as a first stage in realizing an operational Registry in which certification is shown prominently in each profile entry ( by October 2016 ) ST#1a.9) Incorporate all WDS Member information into the Registry such that it can then be used by the Future Earth programme and other stakeholders ( by September 2016 ) ST#1b Engaging with Future Earth 7

8 ST#1b.1) Engagement with ICSU-WDS by Future Earth projects is voluntary. Send out a clear message on why the projects should interact with ICSU-WDS and how they do that ( ongoing) ST#1b.2) Recognize opportunities for collaboration between WDS Members/partners and the Future Earth programme, and initiate and develop connections between them such that these opportunities can be seized (ongoing) ST#1b.3) Develop a guidance framework for Data Management Plans in collaboration with Future Earth, RDA, CODATA, and Belmont Forum e-infrastructure and Data Management (ongoing) ST#1b.4) Establish a registry of applicable guidance documents, stratified by domain and possibly international programme (by October 2016) Strategic Target 2: Nurture active disciplinary and multidisciplinary scientific data services communities Objectives for ST2 a) Build bridges between data communities through workshops and fora, including the SciDataCon Conference co-organized with CODATA. b) Promote better interaction within and across data communities to increase awareness of other approaches and practices. i. Do this within disciplinary communities to promote best practices ii. Do this across disciplinary communities to encourage cross-fertilization c) Document and disseminate approaches and practices employed by successful communities/wds Members to: i. Promote adherence to standards and conventions. ii. Accelerate the process of identifying commonalities. d) Provide models of best practice for data management and data sharing to communities as a tangible knowledge base for capacity building. This Action is directly linked with Strategic Target 4 (ST4). Background for ST2 Modern scientific challenges facing humankind in particular those dealing with environmental and sustainability research are multidisciplinary by nature. The transdisciplinary scientific research needed to tackle these challenges must be supported by internationally coordinated and integrated multidisciplinary scientific data services, which cannot exist and succeed without effective underlying disciplinary scientific data services. WDS membership includes disciplinary and multidisciplinary communities of varying levels of maturity: some communities are already well developed, and have achieved sustainable mechanisms for international coordination, while others are still in the early stages of coordination. By providing a forum for these communities to interact and share good practices, and by promoting collaborations in the context of multidisciplinary scientific programmes, ICSU-WDS can accelerate both maturation of less advanced communities and the emergence of a global community supporting scientific data services. The WDS Scientific Committee is uniquely positioned to not only observe similarities and differences between communities but also open up the communication channels between them. The WDS membership accreditation process provides a clear insight on the different approaches and practices, thus enabling a proper mapping and understanding of their individual characteristics. Communication channels with the different data communities include, but are not limited to, the WDS Members Forum, SciDataCon, special international workshops and conferences arranged by WDS, webinars, other international conferences and meetings, as well as journal publications. The polar research communities are multidisciplinary by definition, and hence a good exemplar. 8

9 WDS membership applications serve as a valuable source of best practice and will be used as a basis for development of a Maturity Model in respect of approaches and best practice. This not only allows existing members to evaluate themselves in respect of performance and potential strategies for improvement, but also informs prospective members and data communities at large about best practice. Current work to align DSA and WDS accreditation within RDA can be assimilated into the Maturity Model, and the same process can be applied in time to other accreditation schema. See Annex B for a conceptual overview of the Maturity Model. Required Resources and Risks for ST2 Component Resource estimate Contributors Proposals / resources Risks Documenting and disseminating approaches and practices Dedicated time DSA WDS WG WDS-IPO TBC Inability to find non WDS-SC willing collaborators in the WG Providing models of best practice Funding or in-kind contribution for a small development project WDS-IPO WDS-SC WDS Members TBC Funding and resources may not be available Deliverables/Milestones for ST2 ST#2a Building bridges between data communities ST#2a.1) Realize SciDataCon 2016 and International Data Week (IDW) in September 2016 ( by September 2016 ) ST#2a.2) Realize the WDS Members Forum in conjunction with IDW ( by September 2016) ST#2a.3) Plan and organize WDS sponsored workshops in Proposed targeted audiences are health and climate change data communities, and in particular the carbon cycle and water cycle ( by December 2016) ST#2b Promoting better interaction ST#2b.1) Cross-fertilize and promote best practices in data management, including practical aspects (where the data will be stored, documentation) ( ongoing ) ST#2b.2) Seek opportunities to collaborate with CODATA in their Workshops as and when applicable ( ongoing ) ST#2b.3) Survey members for data management educational resources and tools that they host on their websites, adding an item to the Biennial Reports ( by July 2016 ) ST#2b.4) Seek opportunities to collaborate with RDA Interest Groups and Working Groups as and when applicable ( ongoing ) ST#2c Documenting and disseminating approaches and practices ST#2c.1) Compile a review document of the current initiatives within a well-understood data community and their scope, the standards employed, and where gaps are seen for improvements and collaboration ( by September 2016 ) ST#2d Providing models of best practice ST#2d.1) Develop the Maturity Model and use it to assess the level of Maturity of current WDS Members. This will provide a benchmarking tool to Members ( by April 2017 ) 9

10 ST#2d.2) Develop an online tool for Maturity Model implementation integrated with DSA WDS membership applications ( by April 2017 ) ST#2d.3) Publish the Maturity Model in an online tool for WDS Members to use for incremental improvement and self-evaluation of maturity. The tool serves as an interface to best practice documentation ( by April 2017 ) ST#2d.4) Produce a journal paper on the Maturity Modelling Tool ( by April 2017 ) Strategic Target 3: Improve the funding environment for data services Objectives for ST3 a) Influence funders regarding the importance to fund research data infrastructure and data activities. b) Explore funding opportunities available for WDS Members through various different public and private sources from national, regional, and international sources. c) Develop strategic partnerships with developing countries to seek funding to underpin the development of data infrastructure and activities in those countries. Background for ST3 WDS should be the international voice to advocate for a sustainable global research data infrastructure (GRDI) and propose mechanisms to better link infrastructures across domains and boundaries. Some parts of the GRDI are available and more or less sustainable (DataCite, re3data) whereas other parts needs to be established and or sustained (WDS Metadata catalogue, Knowledge Network, Data Publication.). Required Resources and Risks for ST3 Component Resource estimate Contributors Proposals / resources Risks 3a Travel and support funding for stakeholders at the workshop needed. Covered by the OECD-GSF and other partners OECD Global Science Forum Risk is that funding agencies attention might be difficult to get. 3b Resources of the IPO can cover these actions. IPO IPO 3c (1 FTE 2-3 weeks) Workshop documents. + (1 FTE 4-6 weeks) Publicizing by the IPO IPO in-kind by SAEON Deliverables/Milestones for ST3 ST#3a) Influencing funders to fund research data infrastructure and activities ST#3a.1) Scoping document and virtual workshop demonstrating to funders the saving that supporting data infrastructure would ultimately bring to them. This will be one of the outputs of the OECD-GSF/CODATA/WDS activity on Sustainable Business Models for Data Repositories, which builds on the initial RDA/WDS Cost Recovery WG report that documents existing business models (by September 2017 ) 10

11 ST#3a.2) Roadmapping exercise and workshops to define the elements of a sustainable and coordinated global data infrastructure. The Data Ecology document should feed into this activity, which will draw heavily on the OECD-GSF/WDS/CODATA study on International Coordination of Data Infrastructures (by September 2017) ST#3a.3) Support the Belmont Forum e-infrastructure and Data Management activity (ongoing) ST#3b) Exploring funding opportunities available through various different public and private sources ST#3b.1) Work with WDS Members to identify available regional and international funding opportunities (by September 2016 ) ST#3b.2) Inform WDS Members about the funding opportunities in arising from the Belmont Forum e-infrastructure and Data Management activity, and suggest ideas to apply for funding (by September 2016) ST#3c) Developing strategic partnerships with developing countries to seek funding ST#3c.1) Explore ways of limiting the cost for obtaining DOIs in developing countries through close collaboration with DataCite and CODATA TG on developing countries (by December 2016) Strategic Target 4: Improve the trust in, and quality of, open scientific data services Objectives for ST4 a) Continue to support and improve a trustworthy data repositories and services certification framework for WDS Regular and Network Members. b) Facilitate access to, and use or reuse of datasets including through publication in particular for multidisciplinary research. Background for ST4 Scientific data services with certified technical and scientific capacities are essential components of the research environment. They play an essential role in ensuring the integrity and availability of datasets, and thus promote trust in science. ICSU-WDS is committed to increasing the quality of, and trust in, the services provided by its Members, and will concentrate on the above targets. Required Resources and Risks for ST4 The resources for ST4 should be primarily covered by the WDS-IPO through its NICT and ICSU funding (WDS Certification). Additional external funding should be sought to support, improve and extend the certification framework. Deliverables/Milestones for ST4 ST#4a Expanding WDS certification ST#4a.1) Provide backing to the DSA WDS Partnership Working Group endorsed by RDA, which has the ultimate goal of creating a global accreditation framework (details here ) ( ongoing ) ST#4a.2) Continue to accredit new WDS Members, who must show evidence of employing international standards for metadata content and catalogue service protocols within the certification process (ongoing) 11

12 ST#4a.3) Work with existing WDS Members to ensure that such standards are being employed. This will happen in reality when aggregating their metadata through the WDS Metadata Catalogue. (ongoing) ST#4a.4) Revise the WDS Data Sharing Principles to account for the broadening of WDS to reflect the demands of new domains, as well as the changing landscape of open access to data. Draft detailed guidelines (completed) ST#4c Establishing data publication as trusted service ST#4c.1) Maintain support for the WDS Data Publishing Working Groups and promote adoption of outputs by WDS Members ( details here ) (ongoing) Strategic Target 5: Position ICSU-WDS as the premium global multidisciplinary network for quality assessed data Objectives for ST5 a) Expansion of membership, focusing on broadening the disciplinary and geographical scope of ICSU-WDS. b) Demonstrate utility of ICSU-WDS by showcasing innovative science and societal benefits of members activities c) Establish SciDataCon as a high-profile international conference on scientific research data, in collaboration with ICSU-CODATA. Background for ST5 ICSU-WDS seeks to create and support a community of excellence between repositories of quality-assessed data with the goal of supporting the science underpinning ICSU programmes and the broader needs of society. There is a need to ensure that this community includes all relevant disciplines and constituencies. In order to ensure continued support for member activities, ICSU-WDS needs to be an advocate for open data, including advertising and promoting activities to the wider science and policy communities. ICSU-WDS also needs to facilitate the sharing of best practices and experience amongst its membership, including through providing high-profile opportunities for communication. Required resources and Risks for ST5 The resources required for ST5 should be covered by the WDS-IPO through its NICT and ICSU funding and specific external funding where needed. Deliverables/Milestones for ST5 ST#5a Broadening scope of ICSU-WDS membership ST#5a.1) Establish Ad hoc Subcommittee to identify gaps in WDS membership (completed) ST#5a.2) Develop membership package, stressing benefits of membership of ICSU-WDS, and including examples from active WDS members (completed) ST#5a.3) Targeted recruitment drive (by December 2016 [ongoing]) ST#5b Increasing active communication and advocacy ST#5b.1) Appoint a dedicated communication officer, in order to increase rate of internal and external communication (by December 2016) 12

13 ST#5b.2) Highlight Members activities through the WDS website (ongoing) ST#5b.3) Use the WDS Blog to communicate to wider community a. Establish rota for SC members for Blog (completed) b. Contact members about contributions to the Blog (completed) ST#5b.4) Seek appropriate channels for ICSU-WDS to make known its concepts (completed) ST#5c Establishing SciDataCon as high-profile conference ST#5c.1) Create high-level committee to coordinate science foci for SciDataCon that are of interest to the scientific and end-user communities (completed) ST#5c.2) Use WDS Members and CODATA National Members as conduits to advertise SciDataCon. ( By September 2016 ) ST#5c.3) Outreach to communities/fields beyond those traditionally associated with the two organizations, in particular to industry and early career scientists, and pinpoint suitable individuals within those communities/fields to engage in the conference as keynotes and otherwise (by September 2016) 13

14 14

15 15

16 Annexes Annex A: WDS Infrastructure Concept WDS is supporting a conceptual design, or vision, of interlinked foundational Global Research Infrastructure that is sustainable, scalable, and distributed amongst leading organisations, data centers, and initiatives. To achieve this, one has to identify the major actors (people, institutions, research outputs, and systems) that contribute to the conduct of science today, and then implement a minimum Knowledge Network that will maximize the return on investment for all (or most) of these actors. A central characteristic of the Knowledge Network is that it should allow contributions from many sources, including, at least conceptually, from non-traditional sources such as project websites, social media, and citizen science - in addition to the obvious source of metadata. WDS foresees that such a Knowledge Network will draw on the example of Linked Open Data, and re-use as many services, components, standards, and existing capacity as is possible. WDS has a role in cementing the missing relationships in the network, and in providing components (nodes) that are not yet available. Design Considerations The Knowledge Network, in its implementation, must satisfy a number of considerations, some of which were mentioned already: Scalability: This is required in two ways: 16

17 Limiting the extent to which institutions, projects, or other initiatives need to process and commit content in bulk by rather relying on many, near-real time contributions from all participants; Allowing for a distributed resource, while avoiding fragmentation: At present, this distribution is a reality already, but it is not possible for humans or machines to access and exploit the network in a predictable, uniform way. Centralisation or a monolithic solution will not work, for a number of practical, political, and technical reasons. Sustainability: This applies to technical, governance, and financial elements of implementation. Technical sustainability pivots on standards, scalability, and open, extensible software. Sustainable governance should include community support and buy-in, quality assurance, and elements of trust. Financial sustainability is a prerequisite, given the extent to which other investment and infrastructure will rely in future on the Knowledge Network. There is no apparent long-term funder for the entire network; hence, the practical solution must be a community consensus on the loose couplings between sustainably funded components. Diversity: A portfolio of initiatives, projects, and established interests all contribute to and/ or own parts of the network. There is no incentive to recreate these components, hence the only sensible approach is the maximum re-use within the Network. Ease of Use: A significant hurdle to current use of research infrastructure and services is the weight of technical knowledge required: it should be easy and simple to contribute to and exploit the network. See box for some ideas. Registries: Any Knowledge Network will rely on essentially a Registry of Registries a sub-set of the Linked Open Data universe with roles assigned to the contents of these registries. Relationships: While many institutions or initiatives contribute the elements of the Knowledge Network (the nodes ), very few, if any, store the relationships between them explicitly. More About Relationships A central aspect of the Knowledge Network, as depicted in Figure 1, is the relationships between the nodes or elements of the network. As an example, People can have the following standard relationships: To other People, by virtue of Collaboration, Association, etc,; To Trusted Digital Repositories, by Depositing Research Outputs; To their own Research Outputs, through Authorship; Indirectly to Coverages, by virtue of Interest in Topics, Locations, and Time Periods; By being Hosted or Employed by Institutions; and so on. Many of these relationships are implicit in meta-data, but cannot easily be extracted by others. For example, the fact that two people are co-authors of a data set implies collaboration, but this cannot easily be determined from meta-data. The Knowledge Network mines meta-data, amongst other sources, for these relationships, but it can also be contributed from many other sources. At a minimum, simple interfaces should be available to commit meta-data, originating in a number of domain standards, to the Knowledge Network as and when it is created/ updated. Collaboration with DataCite in this regard will be key. Leveraging Existing Infrastructure 17

18 Several elements of the Knowledge Network exists already, for example DataCite assisting with the allocation of DOIs to some research outputs (data specifically), ORCID assigning permanent identifiers to researchers, and so on. Figure 2 identifies these existing components provisionally, at this point. Specific actions that can form part of an initial phase of implementation include 1. Collaboration, with DataCite, on leveraging the merger between re3data, DataBib, and DataCite to establish a Registry of Trusted Digital Repositories. This is a sub-set of the registry envisaged by DataCite, or an independent extension of the registry that aggregates quality and certification properties (WDS, Data Seal of Approval, ISO 16363, etc.), managed by WDS. 2. Inclusion of ORCID as a major registry of permanent identifiers for researchers. 3. DOIs for citable data sets, as provided by DataCite. 4. Relationships, mined from WDS and possibly DataCite meta-data, supplemented by information from the WDS membership. The Knowledge Network can grow in subsequent phases in three ways: Extending to other elements (topic coverages, institutions, publishers, funders). Inclusion of repositories that do not necessarily form part of the WDS membership. Contribution of relationships that do not originate from meta-data. All of these are desirable extensions, and improve the utility of the Knowledge Network for users. Annex B: World Data System/ Data Seal of Approval Maturity Model Concept The availability of Trusted Digital Repositories for Data and Trusted Data Services is an integral part of the emergent global Research Data Infrastructure, and is especially important in two contexts: 18

19 1. The growing emphasis on free and open access to tax-funded research outputs assumes that repositories will be available to accept such outputs, and be capable of supporting their long-term availability; 2. Trusted Repositories need to be reliably available as a primary building block for any derived infrastructure or service, for example the provision of indicator-based decision support. The community has long recognised that accreditation and certification of repositories is necessary - and in response, a number of such accreditation and certification have become established, of which the ICSU World Data System and Data Seal of Approval (DSA) are but two. Other include NESTOR (based on DIN), OAIS/ TRAC, and the recently approved ISO 16363/16919 standards. The availability of multiple certification pathways has some advantages, chiefly that there are varying depths of assessment involved (ranging from self-assessment, with our without review, to formal auditing), and that some accreditation schema are more suitable to specific domains than others. The accreditation schema also vary in respect of the scope of criteria that are evaluated - while all schema are focused on technical aspects of process (ingest, publication, quality assurance, preservation) and infrastructure, not all of them cover sustainability and governance. WDS and DSA both follow a process of self-reporting against a schema of criteria, and responses by prospective members are then assessed - generally by more than one reviewer. This process has precedents in peer-reviewed publication and is strongly entrenched in the culture of science. The process thus has many advantages, amongst others that qualitative assessment is done reasonably well by humans - but there are two concerns that are not addressed by such a process: 1. There is likely to be some subjectivity; 2. More critically, the knowledge, best practice, and value of the member responses remain largely with the reviewers, and are not disseminated or well utilised. Publication of member responses will partly address this problem, but the assessment made by the reviewer can be leveraged to provide much more insight. FInally, one also has to consider that it is, at present, quite difficult to determine the correspondence, if any, between the different accreditation schema and processes, and that it is not always easy for repositories or services that seek accreditation to determine what needs to be done to obtain such accreditation, or what a logical progression is to improved performance, based on best practice. These considerations leads to the concept of a Maturity Model for Trusted DIgital Repositories and Data Services, in which a level of performance can be established reasonably objectively by comparing a statement of current practice with a series of best practice statements, each corresponding to the next level of maturity. In a review process, it then becomes simpler to Assess current member performance against a set of objective statements; Flag the member statement as corresponding to a given level of maturity; Flag the objective statements as requiring revision in the light of improved practice reported by the member. The diagram summarises the data elements and relationships in data to support such a review process. 19

20 In the specific case of WDS, the following practical steps are required to realise the concept as a working resource: 1. In the short term, review of new applications need to be conducted with the Maturity Model in mind - asking each reviewer to assign a level of maturity to the information provided by the 6 members, based on a rubric. The current guidance provided to applicants and reviewers presumably applies to a level of maturity that is acceptable to WDS, and should be incorporated into the rubric. 2. If funding can be found, the applications submitted historically by current members, and also those of applicants that were rejected, need to be subjected to the same process, with benefit from the original reviewers comments and suggestions for improvement. 3. From this information, it should be possible to synthesize a matrix of criteria with practice/ guidance statements for each level of maturity. 4. A web-based tool to support the process can be developed once the information in 3 is available, and publicised to members and prospective members. 5. A publication on the concept, approach, and implementation can follow. 6 Updated and improved by the DSA WDS Partnership WG. 20