a framework for data and information sharing in global agricultural research
DESCRIPTION
A Framework for Data and Information Sharing in Global Agricultural Research. Ajit Maru GFAR Secretariat Rome. Outline. Development of CIARD framework What data and information are and could be shared? Reusability of shared information Interoperability - PowerPoint PPT PresentationTRANSCRIPT
Ajit Maru
GFAR Secretariat
Rome
Development of CIARD framework What data and information are and could be shared? Reusability of shared information Interoperability Action areas for A Framework for Data and
Information Sharing – Technical, Institutional and for Community participation
Recognition that the most important challenges to agricultural development such as from◦ Climate change, ◦ Need for sustainable use of natural resources and
energy ◦ Preventing spread of trans-boundary disease and
pests◦ Loss of Agro-biodiversity cannot be tackled without improved and enhanced sharing of data, information and knowledge globally
Benefits of improved information sharing:
Available information and knowledge can be more easily discovered and put into effective use;
New information and knowledge can be generated, Market failure” or inability to make effective use of
research outputs is reduced; Efficiency and effectiveness of research and
innovation is increased, and duplication is reduced; Greater inclusiveness and participation in research
and innovation is fostered.
Benefits of improved information sharing:
Sharing of data and knowledge ushers greater equity in access and use of agricultural knowledge across and among communities and can lead to greater equity in the benefits of development efforts.
CIARD partners did basic groundwork and organized two International Consultations :
1.Electronic Consultations on E-Agriculture in April 20112.Face to Face Consultation at Beijing, China in June 2011
Data already shared:◦ Bibliographical descriptions of research outputs (e.g.
www.fao.org/agris); ◦ Information about standards, tools, services, datasets,
and events (e.g. www.fao.org/aims, www.ciard.net/ring, and www.agrifeeds.org);
◦ Data on plant genetic resources (e.g. www.sgrp.cgiar.org, www.genesys-pgr.org); agricultural science and technology indicators (e.g. www.asti.cgiar.org);
◦ Agricultural factsheets and e-books (e.g. www.cabi.org/cabebooks); locally produced research re-packaged for wide dissemination (e.g. www.gains.org.gh);
◦ Soil and land-use maps (e.g. INRA Morocco); and remote sensing data (e.g. AREA Yemen).
Data that can be shared:◦ Research Data: Raw data, Processed data,
Information Objects◦ Research Data with Farmers◦ “Hidden”Knowledge such as through social
media; blogs, presentations, minutes etc◦ Sharing with machines
Sharing with machines through existing protocols and formats such as: ◦ Open Archives Initiative Protocol for Metadata
Harvesting (OAI-PMH) for sharing metadata records;
◦ Linked Open Data (LOD) for integrating information between many sites;
◦ Rich Site Summary (RSS) for distributing news items on the Web
◦ Resource Description Framework (RDF) for describing information in a form that is easy to integrate.
The potential value of any given information to someone else is not always obvious, so some people recommend erring on the side of sharing “everything”.
Intelligent aggregators can sift through well-tagged chunks of information, extracting and re-packaging the information for new audiences or purposes (example: the extensionist who merely needs one photo from a long report) or re-packaged for different target audiences.
Downside of sharing too much information◦ Spreading resources that are already too thin◦ Doubtful quality
Information is needed about what is being shared, preferably ranked for completeness and quality for the benefit both of information consumers and of creators of value-added services.
Interoperability is a feature of datasets— and of information services that give access to datasets— whereby data can easily be retrieved, processed, re-used, and re-packaged (“operated”) by other systems.
The less pre-coordination required to achieve this, the more “interoperable” the source.
Interoperability ensures that distributed data can be exchanged and re-used among partners without needing to centralize data storage or adopt common software.
“Mashing up” data from multiple sources can lead to new insights about relationships between factors such as weather, markets, crops, and geographic location.
Interoperable data can more easily be pulled together into specialized services, such as crop portals and virtual research environments.
Through Closed Systems – Google, Facebook
Through “Open” Systems – Web 3.0 or Semantic Web standards◦ Globally unique names (identifiers) for things or
Unique Resource Identifiers (URIs) ◦ Common “grammar” for data – RDF, Linked Open
Data◦ Use of shared vocabularies - Agrovoc
Five steps to open data and information 1 star - *- Your content is available on the Web, in
whatever format, under open licenses 2 star -**- Your content is available as machine-
readable structured data [i.e. MS Excel table is better than an image of the same]
3 star-***- Your content is available in non-proprietary formats [i.e. Comma-Separated-Values (CSV) format in preference to MS Excel]
4 star-****- You use RDF standards and URLs (URIs) to identify your content so that people can point to it.
5 star -*****-Your content is linked through RDF to other people’s content to provide context and add value.
An RDF Graph illustrating URIs
For details: http://www.w3.org/TR/rdf-primer/
RDF ClassSchema and Relationships
A framework for sharing of agriculture-related data and information has three dimensions around ◦ Technical aspects and technologies, ◦ Institutional and organizational aspects, and ◦ Participation of the community of actors and
users of the data and information shared.
Action Areas – Technical◦ Services, Tools and Infrastructure – CIARD.RING,
Tools Wiki◦ Standards & Systems Architecture – FAO-AIMS,
open standards for protocols, ontologies, vocabularies that are used across domains, and engage the wider community with the process for setting open standards.
◦ CIARD partners working on Agrotagger, Agrifeeds, AgriVIVO, CIARD.RING Interface improvement, support for Ontology and Thesaurii development
Action Areas – Organization/Institutional◦ Fill capacity gaps in technical aspects of data
interoperability and in practical methods for managing and migrating data
◦ Advocate change in Institutions, Organizational culture and processes
Action Areas – Organizational/Institutional◦ Appropriate Policies, Strategies, Rules,
Regulations, Norms◦ Change in Organizational Structures especially
reward and accountability and research funding
◦ Embedding data and information sharing in work processes
◦ New Institutions for Governance of Information Flows
Action Areas – Community Participation Involve all actors and stakeholders
◦ Policy makers, Research managers, Scientists, Users, Producers, Market Intermediaries, Consumers etc.,
◦ Involve Community to “pull” information such as done by Social Media, Google and Bing Maps, Services Reviews etc.,
Action Areas – Community Participation Generate evidence for advocacy
We need to create an initial mass of services – a web of data related to agriculture
Greater participation in CIARD and use of CIARD Pathways
Strengthen CIARD Community for Advocacy and Action