Search

This Science Project (SP) contributes to the estimation of the impacts of the invasive species on the European Biodiversity and Ecosystems. This topic is important for European Green Deal and the new European Biodiversity Strategy. The SP is also linked with the socio-economic issues because of the NIS implications to the local ecosystems and their services, and their societal goods and services. Since many of the above impacts may be of local scale, they may alter common practices in circular economies. The SP is implemented by: (a) Combining different sources of data and information; (b) Using a dual workflow to analyse the data; (c) Integrating its resources with core EOSC services and potentially horizontal services available; (d) Engaging the relevant scientific communities. The users will be able to: (a) Analyse distribution patterns of invasive species from different sources of data; (b) Compare the above patterns; (c) Provide managerial suggestions to relevant authorities; (d) Build on the existing infrastructure to address more complex questions (e.g. future scenarios).

Virtual Research Environments, Science Gateways and Virtual Laboratories are systems aiming at serving the needs of their designated communities of practice by providing them with a working environment for performing their tasks. These systems have been proposed and exploited in diverse application domains and scopes ranging from education to simulation, collaboration, and open science. This paper analyses the literature published from 2010 to start characterising this manifold family of systems. In particular, the study identified and analysed a corpus of 1167 research papers to highlight their distribution over time, the most frequent publication venues and the characterising topics.

This Presentation looks at the question of how and why to bring about greater reusability of Research Platforms (variously called Virtual Laboratories, Virtual Research Environments, or Science Gateways). It begins with some context for the Australian Research Data Commons, where the authors are based. It then examines the infrastructure concerns that are driving the need for platforms to both exist and remain sustainable, and the connection from this to reusability. The presentation then proceeds to illustrate the ways in which FAIR is being extended to a range of research objects and infrastructure elements, before reviewing the work of the FAIR4VREs WG. The core of the presentation is an examination, with examples or case studies, of four different paradigms for platform reusability: accessing, adopting, adapting, and abstracting. The presentation concludes by examining actions undertaken by the ARDC to increase the likelihood of reusability., Paper for this presentation available at https://zenodo.org/record/6621962

Maps have always stimulated people's imagination and spatiotemporally supported storytelling; however, they cannot alone represent the life and emotions associated with the territories they describe. Story maps are an IT solution to enrich maps with such information. They are online applications enriched with multimedia and textual information that tell map-based stories. Current software for story map building is either commercial or requires advanced IT skills that make it hardly used by environmental experts. This deliverable describes the Story Map Building and Visualizing Tool (SBVMT), an open-source and free-to-use tool to build and publish story maps, which overcomes common drawbacks of other software by operating within the open-science e-Infrastructure (D4Science) used in the MOVING Project. SBVMT includes new features such as an ontology to represent story maps, Semantic Web technologies for data representation, automatic connection to Wikidata, secure multi-user collaboration in story building, and visualisation of the narrative either as a story map or a timeline. This deliverable shows how SBVMT can overcome the perceptual gap between territory and map. We evaluated its usability and effectiveness from both the point of view of experts building the story map and users interacting with it. Using SMBVT we created the story maps related to the MOVING selected regions. Furthermore, exploiting the semantic web technologies, we implemented several SPARQL queries that allow linking different stories and discovering new knowledge.

In multizentrischen, interdisziplinären Forschungsvorhaben ermöglichen sogenannte virtuelle Forschungsumgebungen gleichzeitige und standortunabhängige Zusammenarbeit, beispielsweise bei der Erfassung, Verwaltung und Auswertung von Forschungsdaten. In Ermangelung einer passenden Lösung für die Schlafforschung wurde im Rahmen dieser Arbeit basierend auf Anforderungen von ForscherInnen, KlinikerInnen und VerfahrensentwicklerInnen eine cloudbasierte virtuelle Forschungsumgebung für Biosignaldaten entwickelt. Den Kern bildet ein auf XNAT basierendes klinisches Datenmanagementsystem. Es ist mehrbenutzerfähig, erlaubt feingranulare Rechtevergabe und speichert sowohl Formular- als auch Signal- und deren Metadaten strukturiert ab. Über ein eigens entwickeltes Plugin werden Signaldaten im European Data Format (EDF) unterstützt und herstellerspezifische Inkonsistenzen in EDF-Kanalbezeichnungen anhand einer umfassenden Liste normalisiert. Analysen werden automatisch bei Datenänderungen in XNAT angestoßen und über einen Job Management Service in einer auf OpenStack basierenden Cloud-Umgebung in Docker-Containern ausgeführt. Die Analyseergebnisse werden mit den Originaldaten verknüpft. Neue Verfahren können kontinuierlich hinzugefügt werden und unterliegen, um Reproduzierbarkeit zu gewährleisten, einer Versionskontrolle. Multidimensionale Signaldaten lassen sich über eine Webanwendung einlesen, visualisieren, scoren, clientseitig pseudonymisieren und live mit einer weiteren Partei audiovisuell besprechen und auswerten. Über eine XNAT-Integration können die Daten in der Cloud-Umgebung analysiert werden und etwaige Ergebnisse werden direkt in die Oberfläche zurückgespielt. Da die Anwendung im Browser läuft, kann sie ohne Installation und spezielle Benutzerrechte ausgeführt werden. Um die Benutzerakzeptanz der Anwendung zu testen, wurde eine Studie mit 16 Personen (14 Fachpersonal, 2 PatientInnen) durchgeführt. Nach einer Einführung in die Software füllten die ProbandInnen einen Fragebogen im dichotomen Antwortformat aus, über den die korrekte Funktionalität des Ein- und Mehrbenutzermodus geprüft und eventuelle Fehler dokumentiert wurden. Anschließend wurden die Zufriedenheit und, anhand des System Usability Scale (SUS), die Gebrauchstauglichkeit auf einer fünffach gestuften Likert-Skala angegebenen. Alle ProbandInnen konnten die Anwendung verwenden, wobei zehn ProbandInnen nicht horizontal scrollen konnten, weil ihre Maus dafür nicht ausgelegt war, zwei ProbandInnen konnten den Mehrbenutzermodus nicht verwenden, was jeweils auf die Krankenhausfirewall zurückzuführen war und bei einem Probanden konnte aufgrund von Mikrofonproblemen der Audiokanal nicht getestet werden. Es zeigt sich eine hohe bis sehr hohe Zufriedenheit mit den Teilaspekten der Anwendung und eine sehr hohe Gesamtzufriedenheit (n = 16; MD = 5; R = 1). Der aus den Ergebnissen des SUS errechnete Score (MD = 86,3; R = 23; M = 87,2; SD = 7,7) deutet auf eine gute bis exzellente Gebrauchstauglichkeit hin. Im Sinne der Open-Science-Bewegung sind alle in dieser Arbeit vorgestellten Komponenten Open Source unter einer freien Lizenz. Dadurch konnten sie bereits in anderen Projekten weiterverwendet werden., In multicentre, interdisciplinary research projects, so-called virtual research environments (VRE) enable simultaneous and location-independent collaboration, for example in the acquisition, management and analysis of research data. For want of a solution for sleep research, a cloud-based VRE for biosignals was developed based on the requirements of researchers, clinicians and developers. The core is an XNAT-based clinical data management system. It is multi-user capable, allows fine-grained permission management and stores electronic case report forms as well as signal and metadata in a structured way. A plug-in supports biosignals in the European Data Format (EDF) and normalises inconsistencies in channel labels using a comprehensive list. Analyses are triggered when data changes in XNAT and executed via a job management service in an OpenStack cloud environment within Docker containers. Analysis results are linked to the original data. New procedures can be added continuously and are version controlled to ensure reproducibility. Via a web application multidimensional signal data can be imported, visualised, scored, pseudonymised, and audiovisually discussed and evaluated live with another party. Via an XNAT integration, the data can be analysed in the cloud environment and results are fed directly back into the interface. As the application runs in a browser, it works without installation and special user permissions. To test the user acceptance of the application, a study was conducted with 16 participants (14 professionals, 2 patients). After an introduction to the software, a quantitative survey in dichotomous response format was used to check the correct functionality of the single and multi-user mode and to document any errors. Subsequently, satisfaction and, using the System Usability Scale, usability were evaluated on a five-point Likert scale. All subjects were able to use the application, although ten subjects couldn't scroll horizontally because their mouse didn't support it, two subjects couldn't use the multi-user mode because of the hospital firewall, and one subject couldn't test the audio channel due to microphone problems. A high to very high level of satisfaction with the sub-aspects of the application and a very high level of overall satisfaction were found (n = 16; MD = 5; R = 1). The score calculated from the SUS results (MD = 86,3; R = 23; M = 87,2; SD = 7,7) indicates good to excellent usability. All components presented in this thesis are open source under a free licence and are already being reused in other projects.

Science gateways, virtual laboratories and virtual research environments are all terms used to refer to community-developed digital environments that are designed to meet a set of needs for a research community. Specifically, they refer to integrated access to research community resources including software, data, collaboration tools, workflows, instrumentation and high-performance computing, usually via Web and mobile applications. Science gateways, virtual laboratories and virtual research environments are enabling significant contributions to many research domains, facilitating more efficient, open, reproducible research in bold new ways. This paper explores the global impact achieved by the sum effects of these programs in increasing research impact, demonstrates their value in the broader digital landscape and discusses future opportunities. This is evidenced through examination of national and international programs in this field.

A personal vision from an ecosystem scientist, experienced in facilitating inter- and trans-disciplinary research for addressing barriers, needs and requirements for the next generation of VREs., supported by the Belmont Forum and informed by the Australian Terrestrial Ecosystem Research Network.

The automatic transcription of text in handwritten and machine-printed historical documents has become an established process in the Digital Humanities, its use ranging from large scale archival or library settings to research groups and individual scholars. While considerable progress on understanding limitations and advancing the state of the art has been made in recent years, this research remains largely limited to documents written in European writing systems, most importantly the Latin script. One of the largest and most diverse literary cultures largely ignored by current document image analysis research is the Arabic one. This thesis contains a comprehensive study on the features of Arabic-script documents and their challenges posed to state of the art optical character systems through both a theoretical analysis of the Arabic script and two case studies of retrodigitization on printed classical and modern material. The principal limitations of common methods identified in these studies were subsequently addressed. Two trainable layout analysis methods following the baseline paradigm achieving comparable results to the state of the art while incorporating additional features necessary for the segmentation of complex document pages, a basic method for processing of multigraphic text lines, and the flexible Kraken OCR engine integrating these methods are presented. We show the usefulness of this OCR software not only for traditional text recognition but also a novel character alignment task. Further, we present the eScriptoriumvirtual research environment (VRE) for annotation and transcription. This VREis specifically designed to be able to treat non-Latin, among them Arabic, script material more effectively than existing alternative systems. In the course of this work we also prepared multiple openly licensed training and evaluation datasets for Arabic text transcription and layout analysis., La transcription automatique de textes dans les documents historiques manuscrits et imprimés est devenue un processus établi dans les humanités numériques, son utilisation allant des archives ou des bibliothèques à grande échelle aux groupes de recherche et aux chercheurs individuels. Bien que des progrès considérables aient été réalisés ces dernières années pour comprendre les limites et faire progresser l'état de l'art, ces recherches restent largement limitées aux documents écrits dans les systèmes d'écriture européens, et plus particulièrement à l'écriture latine. L'une des cultures littéraires les plus vastes et les plus diverses, largement ignorée par les recherches actuelles sur l'analyse d'images de documents, est l'écriture arabe. Cette thèse contient une étude compréhensive sur les caractéristiques des documents en écriture arabe et les défis qu'ils posent aux systèmes de reconnaissance optique de caractères de pointe, à travers une analyse théorique de l'écriture arabe et deux études de cas de rétro-numérisation sur des documents imprimés classiques et modernes. Les principales limites des méthodes courantes identifiées dans ces études ont ensuite été traitées. Deux méthodes entraînables de segmentation des pages suivant le paradigme de la ligne de base, permettant d'obtenir des résultats comparables à l'état de l'art et comprenant des caractéristiques supplémentaires nécessaires à la segmentation de pages de documents complexes, une méthode simple de traitement des lignes de texte multigraphique et le logiciel ROC flexible Kraken intégrant ces méthodes sont présentés. On montre l'utilité de ce logiciel de ROC non seulement pour la reconnaissance de texte traditionnelle mais aussi pour une nouvelle tâche d’alignement des caractères. En outre, on présente l'environnement de recherche virtuel (ERV) eScriptorium pour l'annotation et la transcription. Cet ERV est spécifiquement conçu pour pouvoir traiter des textes non-latins, dont l'arabe, plus efficacement que les systèmes alternatifs existants. Au cours de ce travail, on a également préparé plusieurs ensembles de données d'entraînement et d'évaluation sous licence ouverte pour la transcription de textes arabes et la segmentation de pages.

This poster summarizes the results of the CLARIAH-DE Work Package 2: Tools and Virtual Research Environments. Tools and research environments LR Switchboard Documentation Future tasks and potential enhancements, Funded by the German Federal Ministry of Education and Research. Funding Reference Number 01UG1910 A to I.

This deliverable D2.7 describes the Blue Cloud architecture as it is known at Month 20. It is the second release of the architecture and an update of the earlier 1st release of the Blue-Cloud architecture document D2.6. It is expected that there will be further developments and refinements to the Blue Cloud system. For that purpose, one more release of the architecture document is planned, namely report D2.8 in Month 27. In order to make it easier for readers and reviewers, a table is included as part of Chapter 1, which indicates the elements and sections of this Deliverable 2.7, which have been updated or added in comparison with the earlier Deliverable 2.6. The technical framework of the pilot Blue-Cloud features: 1) the Blue Cloud Data Discovery and Access service component to serve federated discovery and access to blue data infrastructures 2) the Blue Cloud Virtual Research Environment (VRE) component to provide a Blue Cloud VRE as a federation of computing platforms and analytical services. In this report, the current architecture and functionalities of each of these components are described in detail as well as the roles of partners that are developing and hosting modules. The Blue Cloud Data Discovery and Access service architecture is based upon a combination of the DAB metadata broker service of CNR-IIA, and the SeaDataNet CDI service modules as developed by MARIS, IFREMER, and EUDAT in the framework of the EU SeaDataCloud project. For the Blue-Cloud Data Discovery and Access service and its modules, additional developments were needed such as adapting and upgrading of existing services, adding new services, testing modules, integrating modules, and testing the integrated service, in order to achieve the planned functionality. The Blue Cloud VRE is largely based upon the existing D4Science e-infrastructure as developed and managed by CNR-ISTI. This e-infrastructure from the start hosted already multiple Virtual Labs and offered a variety of services. These services have been adopted and adapted for the Blue Cloud and new services have been added, while also new Virtual Labs have been constructed and deployed as part of the Blue-Cloud Demonstrators. The D4Science e-infrastructure also already had proven solutions for connecting to external computing platforms and means for orchestrating distributed services, which are instrumental for smart connections to the other e-infrastructures in the Blue-Cloud system. The Blue Cloud demonstrators are developed as Virtual Labs embedded in the D4Science VRE e- infrastructure and supported by data input from the Blue Cloud Data Discovery and Access service and other data resources, and additional computing services. The demonstrators are worked out in a cooperation between WP3 and WP4, analysing their scientific workflows and technical set-up, and considering the D4Science VRE infrastructure and services that provide the basis platform. As part of their development, the demonstrators have required upgrading of existing functionality and development of additional services. This is described, where relevant, in this document. In addition, consideration is given to integration aspects, such as two-way linking between the 2 Blue Cloud components, and expanding the VRE with additional platforms for computing and algorithms, and where needed for specific demonstrators, direct access to data infrastructures. Moreover, aspects of authentication and monitoring are considered on full Blue Cloud scale. The Blue Cloud architecture as described in this report, is designed to be scalable and sustainable for near-future expansions, such as connecting additional blue data infrastructures, implementing more and advanced blue analytical services, configuring more dedicated Virtual Labs, and targeting more (groups of) users.

This deliverable describes the activities carried out during the first 18 months within Work Package 9 for the SoBigData e-Infrastructure operation activity since its deployment, including a detailed set of usage indicators (i.e. the number of users, accesses to resources, usage of resources from scientists, etc.). It also reports the deployment and procedures governing the operation of the Virtual Research Environments, the catalogue, and the ones for the services devoted to Data Analytics. A total of 15 Virtual Research Environments (VREs) have been created and are operational. In particular, the SoBigData gateway provide its users with: 6 Exploratories VREs paired with the use cases (Demography, Economy & Finance 2.0; Migration Studies; Societal Debates and Misinformation Analysis; Social Impacts of AI and Explainable Machine Learning; Sports Data Science; Sustainable Cities for Citizens); 2 Virtual Lab VREs - SoBigDataLab and OpenScienceGraphLab to exploit and experiment tools and solutions; 3 Applications VREs - TagME, SMAPH, M-Atlas; 2 Project Internal VREs - SoBigData.eu VRE for the communications and collaboration among project and initiative members and SBD-InfraCore VRE for supporting SoBigData-PlusPlus WP9; and 2 Literacy And Training VREs - the SoBigDataLiteracy, supporting Critical Data Literacy of task T.2.4, creating a curated collection of literature of interest for the SoBigData Community, and the e-Learning_Area VRE to host training materials developed within the SoBigData project. As of June '21, the 15 existing VREs served more than 8,000 users by a total of more than 30,000 working sessions, with an average of 1500 working sessions per month with increasing trend. This required to deal with approximately 40 issue tracker tickets (14 requests for support, 6 requests for incidents and bugs, 9 requests for new features, and 9 requests for Virtual Machine or Container creations).

This deliverable D5.2 "Virtual Research Environment Operation Report years 1-2" describes the activities carried out during the first 24 months of the DESIRA project within Work Package 5. Specifically in Task 5.1 "Knowledge Infrastructure: the DESIRA Virtual Research Environment" and Task 5.2 "Integration of Services and Tools, and Use Reporting". It reports the procedures governing the operation of the VREs as well as the status of the aggregated resources at mid-term in the DESIRA infrastructure. Virtual Research Environments (VREs) are "systems" specifically conceived to provide their users with a web-based set of facilities (including services, data and computational facilities) to accomplish a set of tasks by dynamically relying on the underlying infrastructure. VREs are among the key products to be developed and delivered by the DESIRA project to support Project coordination, Living Labs activities and Rural Digitization Forums activities. The development of VREs is based on three main activities: (i) the development of software artefacts that realise a set of functions (including those needed for accessing specific datasets), (ii) the deployment of these artefacts in an operational infrastructure following the release procedures and tools, and (iii) the final deployment and operation of well-defined Virtual Research Environments by exploiting the facilities offered by the underlying D4Science infrastructure and its services [1, 2]. This report documents the last of the above three activities - i.e. the exploitation of the services and technologies offered by the underlying infrastructure to serve the needs of defined scenarios - as implemented in the context of the DESIRA project from June 2019 to May 2021. The DESIRA Infrastructure Gateway actually offers end-user access to 14 VREs. As of June 2019, 1 VREs were created and operated. Specifically, the DESIRA Project VRE (cf. Sec 3.1.1) was created before the project kick-off. As of May 2021, these VREs have served the needs of more than 370 users and more than 7.000 user sessions. This required dealing with 177 tickets (103 related to the project management, 28 requests for support and enhancements, five requests for incidents and bugs, 14 requests for VRE creations).

The Blue-Cloud project is piloting a cyber platform bringing together and providing access to multidisciplinary data from observations and models, analytical tools, and computing facilities essential to support research to understand better and manage the many aspects of ocean sustainability. This goal is realised by developing, deploying and operating the Blue-Cloud platform whose architecture consists of two major families of components: (a) the Blue Cloud Data Discovery and Access System to serve federated discovery and access to 'blue data' infrastructures; and (b) the Blue Cloud Virtual Research Environment (VRE) component to provide a Blue Cloud VRE as a federation of computing platforms and analytical services. This Deliverable D4.4 "Blue Cloud VRE Common Facilities (Release 2)" is the revised version of the D4.2 "Blue Cloud VRE Common Facilities (Release 1)". This revised version of the document covers the second period of the project, from M13 up to M27, including the up-to-date information of the services reported on D4.2 and the new services that have been developed and added to the VRE common facilities in the reporting period to serve the needs of the Blue Cloud community. The major changes and new services this deliverable introduces are: an Orchestrator (cf. Sec. 3.3), i.e. a software that allows for a declarative, technology agnostic definition of workflows to coordinate the execution of tasks across diverse services and systems; enhancements to the Workspace service to support tailored storage persistence and satisfy different application scenarios (cf. Sec. 4.1); enhancements in the Publishing Framework (cf. Sec. 6), namely the catalogue extension to deposit catalogue items to Zenodo and the facility to publish geospatial data from the workspace; the facility to interface with the Data Discovery & Access System (cf. Sec. 7.1) to transfer datasets of interest into the workspace for future uses; the notebook to facilitate the exploitation of the WEkEO Harmonised Data Access (HDA) API (cf. Sec. 7.2). This deliverable also updates the Identity and Access Management (cf. Sec. 3.1) and the Analytics Framework (cf. Sec. 5.1 and Sec 5.2) with minor changes reflecting the activities performed in the reporting period. A description of all the services previously documented in D4.2, not modified in the period, is preserved for this document to be self-contained and provide the reader with an overall description of the whole VRE Common Facilities offering. A total of 15 services and components are described in this deliverable by reporting their design principles, architecture and main features. These services and components contribute functionalities to the Blue Cloud VRE Enabling Framework (Identity and Access Management, VRE Management, Orchestrator), Collaborative framework (Workspace and Social Networking), Analytics Framework (Software and Algorithm Importer, Smart Executor), Publishing Framework (Catalogue Service) and improved support for RStudio, JupyterHub, ShinyProxy, and Docker Applications. Additionally, two new VRE services, aiming at bridging two VRE external systems such as the the WEkEO1 catalogue from Copernicus and the Data Discovery and Access from Blue-Cloud with the VRE tools are described. Services and components discussed in this deliverable have contributed to 14 gCube releases, from gCube 4.26 (November 2020) to gCube 5.6.0 (November 2021). They have been used to develop and operate the Virtual Laboratories of the Blue Cloud gateway https://blue-cloud.d4science.org and its underlying infrastructure. At the time of this deliverable the Blue-Cloud gateway and its services are serving more than 730 users with a total of 19000+ working sessions.

This document serves as a white paper, which describes the AGINFRA PLUS vision of a next-generation community driven web-based research infrastructure, as well as present and future application scenarios that can be envisaged from experiences with the piloted AGINFRA PLUS user communities and provides recommendations that aim to inform the roadmap for developing AGINFRA PLUS infrastructure further.In order to further enhance the positioning of the project in the digital science ecosystem, Agroknow worked with avariety of stakeholders, including all project partners, who were invited to contribute to this white paper “Digital Science Recommendations for Food & Agriculture”. External contributors included strategic digital infrastructure initiatives (such as OpenAIRE, the FNH-RI Research Infrastructure for Food, Nutrition and Health, and the METROFOOD Research Infrastructure for promoting metrology in food and nutrition), as well as international stakeholders (such as the University of Guelph, Canada; and the Chinese Academy of Agricultural Sciences).&nbsp

In the AGINFRA PLUS Sustainability Roadmap, we describe the way in which the technology partners of the consortium are envisioning how the key software infrastructures, services and tools of the project are going to be sustained after the project ends. This deliverable first provides an overview of the key software outcomes and highlights the ones selected for further exploitation. Then, it focuses on the targeted stakeholders and market segments, the business models chosen per targeted segment, as well as the operational and financial resources required. After the project’s Mid-Term Review meeting, an early version of this deliverable was prepared and submitted according to the reviewers’ request, to reflect our thinking and approach in terms of sustainability at that point of time. In this new version of the deliverable, we significantly revise, extend and complete our original thinking. It has become clearer to us that in an era of radical digital transformations in the food and agriculture sector, the way in which project outcomes may be positioned and offered to potential users has changed. This creates numerous opportunities for the partners that are interested to continue this investment after the end of the project. The five (5) key technology offerings that we have decided to continue investing into are the following: D4Science-powered VREs (CNR): existing or new Virtual Research Environment (VREs) that specific communities or organisations may continue using or creating, over the D4Science e-infrastructure. EGI-powered Jupyter Notebooks (EGI): existing or new scientific applications that will be using Jupyter Notebooks for interactive data science and scientific computing, over the EGI e-infrastructure. EGI-powered Galaxy Workflows (EGI): existing or new scientific workflows that will be executed over Galaxy, over the EGI e-infrastructure. Data Harvesting & Data Linking Services (Agroknow): existing or new software applications that will be using the semantic indexing and data lookup services of Agroknow (or request a white-label set up of their own similar service). ARPHA-powered Data Journals (Pensoft): existing or new scientific data journals that will be powered by ARPHA. In terms of large-scale computing facilities and cloud storage resources, the above technology offerings have been powered by the two European e-infrastructures that have been serving as the backbone of the project: D4Science e-infrastructure (CNR & UoA): the Data Infrastructure that connects more that 7.000 scientists in over 50 countries, hosts more than 100 VREs for a variety of communities and that promises 99,8% service availability. EGI e-infrastructure (EGI): the federation of cloud providers and data centres, spread across Europe and worldwide, that delivers advanced computing services to scientists, projects and research infrastructures. The deliverable has modified and extended its initial scope of describing an “AGINFRA Association”, as the consortium has decided not to move forward with such a legal entity. Not all partners were equally interested in a joint commitment about the way to continue to support and evolve the various project outcomes. It was therefore decided that a select number of partners responsible for the above key AGINFRA PLUS technology offerings (namely, Agroknow, CNR, Pensoft, and EGI) develop jointly a plan about the future of these offerings. To this end, the original title of this deliverable “AGINFRA Association Blueprint” has been revised to “AGINFRA PLUS Sustainability Roadmap”. This Roadmap includes the following parts: An overview of the main technology outcomes of AGINFRA PLUS, also comparing their IP, TRL and licensing to the ones originally foreseen in the DoA. An introduction of the need that we see in the market & technology outcomes that we have decided to offer: what is the situation in the current market, with the food and agriculture sector being under a huge transformation; who are the potential users/customers that we may address; which are the specific market segments that we believe have potential for our partners and offerings. An overview of the market segments that we target and the offerings that are relevant: what are the specific audiences and their needs; which are the AGINFRA PLUS offerings that we see as relevant for these audiences; what is the competitive advantage that we have against the competition; what is the business model that we have decided to adopt in each case. A description of the strategic, operational and financial requirements to continue supporting each offering: which are the labour and computing resources needed; what are the corresponding costs; how do the financial forecasts look for each offering.

This document serves as a white paper, which describes the AGINFRA PLUS vision of a next-generation community driven web-based research infrastructure, as well as present and future application scenarios that can be envisaged from experiences with the piloted AGINFRA PLUS user communities and provides recommendations that aim to inform the roadmap for developing AGINFRA PLUS infrastructure further. The initial version of this white paper has been prepared with contributions from all partners and was submitted to the EC on M25. In order to further enhance the positioning of the project in the digital science ecosystem, Agroknow worked on a next version of this deliverable that has been organised as an edited volume. A variety of stakeholders, including all project partners, were invited to contribute to this volume titled “Digital Science Recommendations for Food & Agriculture”. External contributors included strategic digital infrastructure initiatives (such as OpenAIRE, the FNH-RI Research Infrastructure for Food, Nutrition and Health, and the METROFOOD Research Infrastructure for promoting metrology in food and nutrition), as well as international stakeholders (such as the University of Guelph, Canada; and the Chinese Academy of Agricultural Sciences). The full text of the volume can be found in Annex A of the present document.

AGINFRA PLUS aspired to provide a sustainable channel addressing adjacent but not fully connected user communities around Food & Agriculture, by exploiting core European e-infrastructures such as EGI.eu, OpenAIRE and D4Science. To this end, the project developed, extended and provided the necessary specifications and components for allowing the rapid and intuitive development of variegating data analysis workflows, where the functionalities for data storage and indexing, algorithm execution, results visualization and deployment are provided by specialized services utilizing European large-scale, cloudbased infrastructure assets. Furthermore, AGINFRA PLUS aspired to establish a framework facilitating the transparent documentation, exploitation and publication of research assets (datasets, mathematical models, software components result and publications), in order to enable their reuse and repurposing from the wider research community. Thus, the vision of AGINFRA PLUS project was to develop a common technical infrastructure that could initially serve three user communities (namely, Agro-climatic and Economic Modelling, Food Safety & Risk Assessment and Food Security) and it could be evolved to an AGINFRA food cloud demonstrator that could be positioned as the European Open Science Cloud (EOSC)1 agri-food thematic cloud. WP8 concentrates on the dissemination of the project and its results among the identified target groups by using online and offline dissemination channels and activities. More specifically, WP8 focused on creating awareness and engaging further the scientific communities that are related to each one of the three user communities of the project (namely, Agro-climatic and Economic Modelling, Food Safety & Risk Assessment and Food Security) and (b) creating general awareness about AGINFRA PLUS and the types of innovative services that scientists may use, in other scientific communities and networks (Task 8.2). Moreover, linking AGINFRA PLUS with international initiatives and networks that are working on open, big and interoperable data for agriculture and nutrition towards contributing to the corresponding standardisation work falls within the scope of WP8 under Task 8.3. Additionally, alignment of the work held in AGINFRA PLUS with the conception, development and deployment of the European Open Science Cloud (EOSC) and its integration with the existing core e-infrastructures is part of Task 8.4. Furthermore, adopting a more sustainable plan for engaging SMEs in the project falls within the recommendations and instructions that we received after the Mid-Term review regarding the number of engaged to our project SMEs. Finally, WP8 managed to establish a sustainable legal entity form (through a not-for-profit association) for the further operation and evolution of AGINFRA PLUS as a joint venture of involved stakeholders of innovative services that scientists may use, in other scientific communities and network (Task 8.5). WP8 has created an online platform (including project website and social media channels) and a set of printed dissemination materials to promote the project (Annex Ε). By the end of the project, the project website has been visited by 2.257 people (unique visitors). The social media channels count 265 community members (Facebook), 987 followers (Twitter) and 777 readers (Medium). Finally, the project has organized 32 major events with selected group of researchers and practitioners (see Section 4.1 and Annex A, Annex B and Annex C). Moreover, the project has conducted 85 AGINFRA PLUS presentations and panel participations at scientific conferences, workshops and other events and was represented at 8 conferences and fairs with a booth or distributing leaflets (see Section 4 and AnnexD). In total, more than 5.000 stakeholders have been reached and 21 scientific publications were published (See Section 6). This deliverable describes online and offline dissemination channels, as well as activities, which were conducted until the end of the project. Moreover, it provides an outlook of the dissemination activitiesn that were implemented in order to increase project’s engagement with SMEs. The collaboration of AGINFRA project with other EU and international projects by providing tools and services is demonstrated along with a description of the data journals created for the project. Finally, Key Performance Indicators are described and applied to measure the effectiveness of dissemination

The Blue-Cloud project plans to pilot a cyber platform bringing together and providing access to multidisciplinary data from observations and models, analytical tools, and computing facilities essential to support research to understand better and manage the many aspects of ocean sustainability. To achieve this goal, Blue-Cloud is developing, deploying, and operating the Blue-Cloud platform whose architecture consists of two families of components: (a) the Blue Cloud Data Discovery and Access service component to serve federated discovery and access to 'blue data' infrastructures; and (b) the Blue Cloud Virtual Research Environment (VRE) component to provide a Blue Cloud VRE as a federation of computing platforms and analytical services. This deliverable presents the Blue Cloud Virtual Research Environment constituents by focusing on both new services and revised existing services that have been developed in the reporting period to serve the needs of the Blue Cloud community. In particular, this deliverable describes a total of 11 services and components. These services and components contribute functionalities to the Blue Cloud VRE Enabling Framework (Identity and Access Management, VRE Management), Collaborative framework (Workspace and Social Networking), Analytics Framework (Software and Algorithm Importer, Smart Executor), Publishing Framework (Catalogue Service) and improved support for RStudio, JupyterHub, ShinyProxy, and Docker Applications. The services are described below by reporting their design principles, architectures, and main features. The deliverable also describes the procedures and approaches governing services and components released by highlighting how Gitea (as Git hosting service), Jenkins (as automation server), and Maven (as project management and comprehension tool) are used to guarantee continuous integration processes. Services and components discussed in this deliverable contribute to 11 gCube open-source software system releases (from gCube 4.16 up to gCube 4.25.1) and are in the pipeline for the next ones. They have been used to develop and operate the Virtual Laboratories of the Blue Cloud gateway https://blue-cloud.d4science.org and its underlying infrastructure. At the time of this deliverable (November 2020), the gateway hosts a total of 8 VREs and VLabs, including five specifically conceived to support the co-development of some of the Blue-Cloud demonstrators (namely, the Aquaculture Atlas Generation for Demonstrator #5, the Blue-Cloud Lab for several demonstrators, the GRSF pre for Demonstrator #4, the Marine Environmental Indicators for Demonstrator #3, the Zoo-Phytoplankton EOV for Demonstrator #1). This gateway and its tools serve more than 400 users that (since January 2020) performed a total of more than 5000 working sessions, more than 1700 accesses to the Workspace, and more than 750 analytics tasks. These exploitation and uptake indicators are likely to grow in the coming months thanks to data updates and continued use, further development of existing VLabs, and finally, the creation of new ones.