Enabling better science: Results and vision of the OpenAIRE infrastructure and RDA Data Publishing Working Group (6th RDA Plenary)
Download as PPTX, PDF4 likes1,953 views
The document discusses the OpenAIRE infrastructure and the RDA Data Publishing Working Group, focusing on enhancing scientific communication and publication processes in research. It highlights the need for interoperability between publication repositories, datasets, and research outputs, emphasizing the importance of open access and effective dissemination of scientific data. The document also outlines various initiatives, guidelines, and the role of funders in promoting open access and monitoring research impact.
![Modern Scholarly Communication
Publishing beyond literature
Comprehensive
scientific reward by
citation of any research
outcome
Improved
understanding of
research outcome
Better research review-
process [repeatability,
replicability, and
reproducibility of
experiments - Goble,
2009]
Effective dissemination
and re-use of valuable
research assets
Lower costs of science
Publication
Data
Scientific
process
Methodological processes,
executable workflows, piece of
software
Input/output to scientific
process](https://image.slidesharecdn.com/paolomanghirostockopenairerdaseptember2015-150916100447-lva1-app6892-151113111751-lva1-app6891/85/Enabling-better-science-Results-and-vision-of-the-OpenAIRE-infrastructure-and-RDA-Data-Publishing-Working-Group-6th-RDA-Plenary-5-320.jpg)
Enabling better science: Results and vision of the OpenAIRE infrastructure and RDA Data Publishing Working Group (6th RDA Plenary)
- 1. Enabling better science Results and vision of the OpenAIRE infrastructure and RDA Data Publishing Working Group Paolo Manghi paolo.manghi@isti.cnr.it Institute of Information Science & Technologies “A. Faedo” National Research Council, Pisa, Italy
- 2. The research group • http://nemis.isti.cnr.it/groups/infrascience • 27 among senior and junior researchers, sys admins and PhD student (computer science & information engineering) Data interoperability Digital Library Foundations & Management Systems Enabling middleware for service-oriented infrastructures Enhanced publication & compound object models Virtual Research Environments De-duplication of information objects ByDonatellaCastelliandAlessiaBardi,September2015,bardi@isti.cnr.it
- 4. Modern Scholarly Communication ResearchInfrastructuresgo beyondliterature Data-driven: Jim Gray’s fourth paradigm software experiment experiment service Dataset publishing Data repositories Scientific process publishing Web-driven: immediate sharing and access to digital knowledge Literature publishing Institutional, thematic repositories Publisher Journal repositories Research Infra Research Infra Market Place
- 5. Modern Scholarly Communication Publishing beyond literature Comprehensive scientific reward by citation of any research outcome Improved understanding of research outcome Better research review- process [repeatability, replicability, and reproducibility of experiments - Goble, 2009] Effective dissemination and re-use of valuable research assets Lower costs of science Publication Data Scientific process Methodological processes, executable workflows, piece of software Input/output to scientific process
- 6. Funders and projects • Funders are crucial to development of science • If public funders, they push for Open Access mandates • Funders require methodologies to monitor impact (ROI) and adherence to mandates of projects they fund • Projects require the same tools to show off their production Publication DataProject Funding Scientific process
- 7. Scientific communication workflows for research products • Publications: well established (PIDs, metadata, deposition, peer-review, citation, dissemination) • Research data: available for given communities (PIDs, metadata, deposition?, peer-review?, citation? dissemination?) • Scientific process: almost inexistent and not supported by scientific reward mechanisms Deposition Peer-review Dissemination
- 8. Sharing/publishing research data: current solutions • Part of article: subset of research data embedded as figures or tables • Additional material possibly submitted to the journal along with the publication full-text • Independent from article: Data deposited and described at dedicated locations • Data centres, Discipline databases, Thematic data repositories • Linked to article: Data deposited and described at dedicated locations with link from/to article full-texts • Discipline-specific, data papers, deposition guidelines • e.g. DRYAD, PANGAEA, GigaDB • Enhanced publications, research objects
- 9. Sharing/publishing scientific process: current solutions • Part of article: piece of code is included in the paper or as additional material possibly submitted to the journal along with the publication full-text • Independent from article: software, VMs, workflows, Services, e-notebooks are deposited and described at dedicated locations • Software repositories, e.g. Github • Workflow repositories, e.g. myexperiment.org • Linked to article: software deposited and described at dedicated locations with link from/to article full-texts • Software papers, but no real deposition guidelines • Enhanced publications, research objects
- 10. Enabling Better Science Market-placeservicesin OpenAIREand RDA
- 12. • Establishment of an interoperable network of publication repositories • Deposition, discovery, linking and monitoring of research products (articles, datasets, software) produced under National and EC funding • Monitoring the compliance to EC OA mandates for publications • Support decision makers with statistics OPEN ACCESS INFRASTRUCTURE FOR RESEARCH IN EUROPE The point of reference for Open Access in Europe ByDonatellaCastelliandAlessiaBardi,September2015,bardi@isti.cnr.it
- 13. The OpenAIRE infrastructure Human Network e-infrastructure • NOADS: National Open Access Desks • Monitor and foster the adoption of Open Access policies at the local level • Support researchers at the implementation of the Data Pilot • e-infrastructure for monitoring impact of OA mandates and research projects • OpenAIRE guidelines for metadata exchange • Zenodo Repository for the deposition of research products ByDonatellaCastelliandAlessiaBardi,September2015,bardi@isti.cnr.it
- 14. Get support (NOADs) Linked Content Statistics Search & Browse Feedback Claim/deposi t Publications & data Research impact Citations, usage statistics +++ Data repositories/aggrega tors Data Journals Metadata on data Publication repositories/aggregators Institutional & Thematic Open Access Journals/Publishers Usage data Metadata And pdfs National funding EC funding Guidelines for use services Institutional CRIS Systems CERN/OpenAIRE “catch-all” Guidelines for data interoperability OpenDOAR re3data Validation Cleaning & Transformation De-duplication Enrichment by metadata and text mining APIs
- 15. OpenAIRE data model: view from the moon
- 16. The OpenAIRE e-infrastructure: view from the moon www.d-net.research-infrastructures.eu
- 17. OpenAIRE e-infrastructure: hardware numbers Production • 44 CPU cores • 84 GB of RAM • 3,998 GB allocated disk Mining Cluster • 14 servers • 98 CPU cores • 514 GB ram • 18,458 GB allocated disk Data provision cluster • 15 servers • 90 CPU cores • 236 GB ram • 12,300 GB allocated disk
- 18. OpenAIRE information space numbers (September2015) http://www.openaire.eu • 12M publications (de-duplicated) • 200,000 links publication-project from 5 funders • 9,000 datasets linked to publications or projects • 34,000 organizations (de-duplicated) • Collected from: • 600+ “direct” data providers • 5,000+ “indirect” data providers (inherited from aggregators) • End-users…
- 19. OpenAIRE information space numbers (September2015) ByDonatellaCastelliandAlessiaBardi,September2015,bardi@isti.cnr.it
- 20. OpenAIRE’s http://www.zenodo.org • Zenodo repository (production) • Deposition of publications, datasets, software • DOI minting and metadata curation • Community support • Much more… • FREE • Numbers • Publications 16,240 • Datasets 1,477 • Software 4,456 • Other products 1,400+
- 21. OpenAIRE partners and liaisons SHARE ByDonatellaCastelliandAlessiaBardi,September2015,bardi@isti.cnr.it
- 22. Other OA initiatives: international collaborations RDA-ANDS (Australia) SHARE (United States) La Referencia (South America) CAS (China) ByDonatellaCastelliandAlessiaBardi,September2015,bardi@isti.cnr.it
- 23. Sharing research products and context to enable better science in OpenAIRE Publicatio ns EC funding National funding Research Data Research Initiatives Scientific Process OpenAIRE 2009 - 2012 OpenAIRE Plus 2011 - 2014 OpenAIRE2020 2015 - 2018 EC Open Access mandate monitoring European Grid Infra (EGI) Links among publications, data and process National funders Links between publications and data ByDonatellaCastelliandAlessiaBardi,September2015,bardi@isti.cnr.it
- 24. Publications in OpenAIRE • Publications acquisition policy • Open Access publications • Publications linked to a project whose funder is supported by OpenAIRE • Publications are collected from literature repositories and “claimed” by registered end-users • Metadata and full text (when Open Access or agreed with publishers) Publicatio ns EC funding National funding Research Data Research Initiatives Scientific Process
- 25. Funders and projects in OpenAIRE • Collects projects from the following funder sources • European Commission: FP7 and H2020 • Wellcome Trust • FCT (Portugal) • NHRMC (Australia) • ARC (Australia) • Science Foundation Ireland (Ireland) • On the way: Croatian, Dutch, and American (NSF) Publicatio ns EC funding National funding Research Data Research Initiatives Scientific Process
- 26. Enabling better science: publications and funders • OpenAIRE guidelines for literature repositories • “How to describe” publications • “How to describe” projects • “How to put publications in context” with projects • Cooperation with SHARE (US), JISC (UK), La Referencia (South America) • OpenAIRE Services • Offering access to project information by funder • Inferring links between articles and projects of any funders • Monitoring ROI/Open Access of any funders by project (and more) Publicatio ns EC funding National funding Research Data Research Initiatives Scientific Process
- 27. Enabling better science: publications and funders • Literature Broker Service for Institutional Repositories (deliver 2016) • Serving repository managers • Subscriptions based on configurable criteria of publication-repository “closeness” 27 Publicatio ns EC funding National funding Research Data Research Initiatives Scientific Process
- 28. Research Data in OpenAIRE Publicatio ns EC funding National funding Research Data Research Initiatives Scientific Process • OpenAIRE research data acquisition Policy • Must be linked to OpenAIRE publications or to projects • No datasets identified by accession numbers • Dealt with as “external links” • Datasets are collected from data archives and “claimed” by registered end-users
- 29. Enabling better science: research data Publicatio ns EC funding National funding Research Data Research Initiatives Scientific Process • OpenAIRE guidelines for data archives • “How to describe” datasets (inspired by DataCite) • “How to put datasets in context” with projects • OpenAIRE services • Inference of links to datasets from article full-text • Extraction of dataset-publication links from data archives (e.g. PANGAEA, DataCite)
- 30. Research initiatives • OpenAIRE opens to “research initiatives” willing to • Monitor the productivity of the community in terms of publications and datasets • Support the discovery of research made by peers in the same community Publication DataProject Funding Research Initiative Publicatio ns EC funding National funding Research Data Research Initiatives Scientific Process
- 31. Enabling better science Research initiatives • OpenAIRE research initiatives • European Grid Infrastructure (EGI), concepts: EGI Virtual Organizations and EGI disciplines • OpenAIRE services • Inference of links to research initiatives from article full-texts • Monitoring ROI and Open Access w.r.t. relevant “concepts” of a research activity Publicatio ns EC funding National funding Research Data Research Initiatives Scientific Process
- 32. Enabling better science: Scientific processes To be defined • Scientific process acquisition policies • e.g. software, process (e.g. Taverna workflows), methods (e.g. e- notebooks) • Collection strategies • From “process repositories”? E.g. myexperiments.org, GitHub • Guidelines for “process repository managers” • OpenAIRE services • Monitoring ROI of projects in terms of processes! • Inference/extraction of article/process links? Publicatio ns EC funding National funding Research Data Research Initiatives Scientific Process
- 34. Research Data Alliance (RDA) PublishingData ServicesWorkingGroup • Forum funded by the Commission to propel the discussion among researchers and practitioners in the ambit of research data management • Identification of common, cross-discipline problems and yield best practices, recommendations • Organized in Interest Groups and Working Groups • Focus: • Publishing Data Interest Group: umbrella of WGs focusing on enabling a stronger research data publishing infrastructure • Publishing Data Services Working Group: focusing on article- datasets interlinking
- 35. Publishing Data Services Working Group Data-article links • Benefits of creating context by establishing data-article links • Increasing visibility and discoverability • Stimulating reuse and repeatability • Key to make it worth it: • Infrastructural approach: linking needs to be done collectively, at community (and cross-community) level, sharing procedures, policies and technologies • Issues • No common framework for interlinking datasets and published articles • Initiatives live in isolation and cannot be combined
- 36. Enabling better science: giving open access to article-dataset links • Creating “an open, freely accessible, web based service that enables its users to identify datasets that are associated with a given article, and vice versa” • The Service will serve as a flexible sandbox • Major scholarly communication stakeholders involved at different levels • Feed authoritative links to the Service • Access links from the service • Feedback requirements, preferences, recommendations, obstacles to refine/enhance the service
- 37. Harkan Grudd Siddeswara Guru Laure Haak (ORCID) John Helly Francisco Hernandez Simon Hodson Richard Kidd (RSc) Hylke Koers (Elsevier) – co-chair Paolo Manghi (OpenAire) Haralambos Marmanis Caroline Martin Jo McEntyre (EMBL - EBI) Yolanda Meleco Sheila Morrissey Lyubomir Penev Mohan Ramamurthy Howard Ratner Nigel Robinson (Thomson Reuters) Sergio Ruiz (DataCite) Uwe Schindler (PANGAEA) Johanna Schwarz (Springer) Martina Stockhause Carly Strasser Eefke Smit (STM) Jonathan Tedds Joachim Wackerow Juanle Wang Hua Xu Eva Zanzerkia Claire Austin David Arctur Amir Aryani (ANDS) Geoff Bilder (CrossRef) Timea Biro Adrian Burton (ANDS) - co-chair Ian Bruno (CCDC) Sarah Callaghan David Carlson Jamus Collier (PANGAEA) Suenje Dallmeier-Thiessen Tim DiLauro Ingrid Dillo Rorie Edmunds Janine Felden Carol Goble Jeffrey Grethe PDS-WG Stakeholders
- 38. The Data-Literature Service • A one-for-all service model infrastructure for the research data publishing • Increase interoperability • Decrease systemic inefficiencies • Power new tools and functionalities to the benefit of researchers
- 39. Benefits • For data repositories and journal publishers • Linking becomes more scalable and cheaper, ensuring more visibility for data sources (and their “customers”) • For research institutes, bibliographic providers, and funding bodies • Enables bibliographic services and productivity assessment tools that track datasets and journal publications within a common framework • For researchers • Sharing and accessing relevant articles and data easier, more efficient and accurate, thereby increasing scientific reward and enhancing its practices.
- 40. System development and operation: OpenAIRE and PANGAEA Links collection … Harmonizing PID resolution De- duplicating Information Space Web Portal Core Data Model Data Sources OAI-PMHSearch APIs Examples: • Pairs of DOIs • DataCite records • PANGAEA records OAI-PMH intersection
- 41. Information Space Core Data Model Schema
- 42. The Service (BETA) http://dliservice.research-infrastructures.eu Powered by: • OpenAIRE D-NET software • PANGAEA search engine
- 43. Some numbers • Close to 1 Million links and 2 Millions objects
- 44. Providers
- 46. Research products publishing workflow Digital research Products (articles, data, scientific process) Repositories for literature, research data, scientific process Research e-infrastructure Market-place services ByDonatellaCastelliandAlessiaBardi,MassimilianoAssante,September2015,bardi@isti.cnr.it
- 47. Research products publishing workflow Reuse of products Lack of context: no replication Deposition De- contextualisation Staticity Extra Cost Quality Assessment Inefficient peer-review: No repeatability Dissemination Fragmentation in thematic or typology silos Lack of semantic linking
- 48. Research activities • A research activity can be intended as the course of actions, following a scientific method that leads to prove an initial thesis in order to bring novelty to a research field • Every research activity builds upon and produces a wide array of research products Publication DataProject Funding Research Initiative Scientfiic process Research Initiative
- 49. Time for a change in scholarly communication Publishing Research Activities • De facto the literature publishing workflow has been adapted and adopted for other research products • “Elsewhere” and “on date” philosophy • On the contrary, modern research conducted with the support of Research e-Infrastructures is • Strongly contextualized, intrinsically dynamic • Research products should be published “in place” and “during” and together with research activities • Research e-Infrastructures should evolve to support marketplace-like functionality
- 50. Science 2.0 Repositories • Creation of research products in the Re-I is intercepted and the new products are published • Notify peers about research activities and published products via research social networks • Foster continuous open peer review ByDonatellaCastelliandAlessiaBardi,MassimilianoAssante,September2015,bardi@isti.cnr.it
- 51. SciRepo publishing model benefits Deposition In context Products remain “alive” No Extra Cost Alternative products Quality Assessment Continuous and in context Self- assessment Dissemination Unified Automatic and Complete Deposition De- contextualis ation Staticity Extra Cost Quality Assessment Ineffective peer-review Dissemination Fragmentat ion in thematic or typology silos Lack of semantic linking Current Model Cons SciRepo Model Pros ByDonatellaCastelliandAlessiaBardi,MassimilianoAssante,September2015,bardi@isti.cnr.it
- 52. An example of SciRepo Research Activity web page ByDonatellaCastelliandAlessiaBardi,MassimilianoAssante,September2015,bardi@isti.cnr.it
- 53. SciRepo in the real world http://www.i-marine.eu • The iMarine Research Infrastructure features part of the SciRepo social functionalities: • Applications running in the RI generate products that can be shared • Notifications of new research products via News Feed • Research products accessible in the context of the application that generated them • More will be realized for the new BlueBridge project ByDonatellaCastelliandAlessiaBardi,MassimilianoAssante,September2015,bardi@isti.cnr.it
- 54. Thank you Suggested reading: • Bardi A., Manghi P. A Framework Supporting the Shift from Traditional Digital Publications to Enhanced Publications (2015). doi: 10.1045/january2015-bardi • Manghi P., Bolikowski L., Manola N., Schirrwagen J., Smith T. OpenAIREplus: the European Scholarly Communication Data Infrastructure (2012). doi:10.1045/september2012-manghi • Assante M., Candela L., Castelli D., Manghi P., Pagano P. Science 2.0 Repositories: Time for a Change in Scholarly Communication (2015). doi:10.1045/january2015-assante Contacts: donatella.castelli@isti.cnr.it paolo.manghi@isti.cnr.it alessia.bardi@isti.cnr.it