CGIAR (ex- Consultative Group on International Agricultural Research) is a large network of 15 agricultural research centers located all over the world, which generate and disseminate knowledge, technologies, and policies for agricultural development through the CGIAR Research Programs. The amount of research outcomes produced by the researchers of these centers is huge; still, there was no concrete plan for facilitating access to these outcomes nor any hint of opening up access to them until 2012 when the CGIAR Consortium approved the CGIAR Principles on the Management of Intellectual Assets. It was obvious that the CGIAR Consortium had the intention to make all of its data and research outputs open and harvestable and in this context, these principles were approved and adopted in March 2012.
Some time ago, back in April 2014, CGIAR (and more specifically Piers Bocock, Director of Knowledge Management and Communication at the CGIAR Consortium) presented the status, the progress and the plan regarding the implementation of an open access policy in the CGIAR Consortium. The presentation of a solid plan for the next months as well as the promise for substantial support to the research centers (referring not only to funding for the implementation of the Open Access Policy but also in terms of support packages with material related to the implementation of the policy. Most of the related news and documentation are available through the CGIAR Open Access page; the deadline for all CGIAR Centers to develop and implement an Open Access Policy in their document repositories (that means to be able to show that they have a fully functional infrastructure and workflows) is currently December 15th, 2014.
It is obvious that the CGIAR Consortium has all the good will (along with a time plan including specific actions to be implemented in specific time frames) to push things forward with the implementation of Open Access in all CGIAR Centers but is this enough? It is obvious that since then there has been some progress towards the implementation of the Open Access policy, based on the time frame presented by Piers Bocock: Related guidelines and documentation were prepared and shared either publicly or directly with the CGIAR Centers while the support packages and funds seem to be a work in (ongoing) progress. This task may seem straightforward but it has some challenges; not only the CGIAR Consortium needs to ensure the funding required by each CGIAR Center for implementing the Open Access policy (it is the funding that will allow the allocation of resources and effort towards the implementation of the Open Access mandate) and prepare/provide the basic guidelines and support packages that will allow the design and implementation of OA plans, but these Centers exhibit a wide diversity in terms of existing infrastructure, workflows and staff allocation needed for this purpose. Different CG Centers seem to exhibit a wide variety regarding their current status, capacity and efficiency as well as the status of the preparations for the implementation of the OA Policy proposed by the CGIAR Consortium.
About the CGIAR Open Access & Data Management Policy
The CGIAR Open Access and Data Management Policy clearly defines several aspects of the OA policy to be implemented in all CGIAR Centers. It is based (and therefore is fully compliant) with the CGIAR Principles on the Management of Intellectual Assets (PDF), a document published back in March 2012 and refers to the open access to different types of Intellectual Assets. The new document now clarifies several aspects of the CGIAR Open Access Policy including:
- The period covered by the Policy: The transitional period will last for 5 years, starting from the publication of the Policy (October 2013).
- The information products covered by the Policy: peer-reviewed journal articles, reports and other papers, books and book chapters, data and databases, data collection and analysis tools (e.g. models and survey tools), video, audio and images, computer software, web services (e.g. data portals,modeling on-line platforms), and metadata associated with the information products above.
- Requirements that should be taken into consideration by the OA Implementation Plans, such as:
- Openness, respecting intellectual property rights, confidentiality, sensitivity (including price and politically sensitive information), farmers’ rights and privacy. In the case of publications, options such as self-depositing a post-print version (self-archiving) or the use of Open Access journals should be explored, respecting the embargo periods as defined by each commercial publisher.
- Technical infrastructure (e.g. metadata repositories); the plan should make use of existing infrastructure and repositories, where possible, in order to avoid duplication of effort;
- Syntactic and semantic interoperability, referring to the use of standard metadata schemas and controlled vocabularies, as well as the use of standard protocols for exchanging information with other repositories and existing infrastructure (OAI-PMH, APIs etc.);
- Data storage and preservation for future use, focusing on the management, maintainance and curation of the information products;
- Copyright and open licenses, regarding the use of suitable open licenses;
- Incentives and professional expertise, referring to the devision, adoption and promotion of means to promote and establish Open Access in the most appropriate way;
- Translation of documents and other media, where the translations should also be made Open Access;
- Provision for limited internet connectivity, so that users with limited connectivity will still have access to versions of the information products;
- Open Access and Data Management Plans, which should include a strategy for maximizing opportunities to make information products Open Access.
In addition, the document titled “CGIAR Open Access and Data Management Implementation Guidelines” (PDF) provides more specific information on the implementation of the OA Policy by the CGIAR Centers.
Transforming requirements into an implementation plan: Practical considerations
It is clear that an Open Access implementation plan should be aligned with the requirements defined in these two documents produced by the CGIAR Consortium and take into consideration practical aspects such as:
- The use of standard open licenses (such as the appropriate Creative Commons licenses) for the description of all information products;
- The support of both Green and Gold Open Access paths for ensuring Open Access to publications and other information products produced by the CG Centers;
- The use of the CG Core metadata schema as the standard for the description of the information products with metadata; this can take place either at the repository level (if integrating CG Core is an option) or through a metadata transformation from the metadata schema currently used by the CG Center to the CG Core at the interoperability level;
- The use of interoperability options: In all cases, standard, widely used controlled vocabularies should be used (where applicable) and the software solutions applied (such as digital repositories) should follow interoperability standards and protocols, such as OAI-PMH, LOD, RSS, Rest-based APIs etc, even SPARQL endpoints, where applicable;
- Discarding problematic infrastructure & using existing standards: It is obvious that a number of CG Centers are currently using custom/home-made, non open source or obsolete (not currently supported) database tools for managing their metadata and information products. Now it will be a good time for an early “spring cleaning”; metadata available through these tools should be exported from these tools in a way that will ensure that not a single piece of valuable information will be lost in the process and then imported in a modern, open source and supported tool that will be used from now on – for instance, AgriDrupal is recommended for this purpose, as it is based on the widely used and opn source Drupal, featuring components specific for the agricultural information management and interoperability (e.g. the use of the AGROVOC thesaurus for the classification of resources). It should be noted that this process should be applied in a way that respects the current metadata management workflows (in case they are functional) but still it should discard all time-consuming and incompatible processes;
- The (re)use of the existing infrastructure: Since CGIAR has already established a centralized repository (CG Space) which serves a number of CG Centers, this should remain at the heart of any new planned infrastructure, harvesting metadata from the metadata repositories of the CGIAR Centers. It can also be used as a CG Center’s repository, in case the Center does not have the capacity or resource to host an instance using its own infrastructure;
- Staffing: Our (short) analysis revealed that not all CG Centers have the same capacity in terms of information and knowledge management; some of them have a really well-staffed Information and Knowledge Management division, while others could use additional support from metadata experts, knowledge managers and related specialized and experienced staff. Based on the new metadata management workflows, key roles should be knowledgeable staff that will ensure the proper function of these workflows.
How can Agro-Know support the CGIAR Open Access policy?
The Agro-Know team, consisting of people with different but related expertise, such as agriculture domain experts, information specialists, data integration experts and developers (among others), has the expertise and previous experience to design and implement such Open Access implementation plans. Agro-Know is an active contributor in European and international standardisation initiatives in relation to the agricultural data representation and storage, with involvement in fora such as the Agricultural Information Management Standards (AIMS) of the Food and Agricultural Organization of United Nations (FAO) and the Coherence in Information for Agricultural Research for Development (CIARD). It is supporting the open data and models movement, participating and contributing in initiatives such as the Research Data Alliance and its dedicated Agricultural Data Interoperability Interest Group, the World Bank’s Global Food Safety Partnership where it co-chairs the Knowledge & Learning Systems Working Group and leads the Database Subgroup, as well as the Global Open Data for Agriculture and Nutrition initiative (GODAN).
By applying components of its Data Framework, Agro-Know can further support the CG Centers in the development of additional data-powered tools and solutions, building on their existing infrastructure and linking them to existing communities and infrastructures. The following list highlights a number of expected contributions but should not be considered as a full list of them:
- Opening up access to information products and research data: Agro-Know has the experience to design and implement Open Access and Open Data Management Plans and is currently working on several instances, such the development of agri-food domain-specific requirements for the Open Access and Data Implementation Plans for the Horizon 2020 projects;
- Propose solutions for enhancing the existing workflows related to data authoring, management, transformation, discoverability, as well as the improvement of their quality reusing existing infrastructure and software tools where possible;
- Enhance discoverability: Design and implement user-friendly interfaces such as finders and even thematic portals (aggregating related sources from other quality data sources) powered by a metadata aggregators and appropriate metadata management workflows that will further improve the retrieval of the information products of each CG Center and their interoperability with existing infrastructure and data sources;
- Enhance the dissemination of research outcomes: Opening access could be facilitated by establishing ways through which the open access repositories and products could be discovered and retrieved. In this direction, the established Open Access repositories of the CG Centers (along with individual services of the CG Centers such as blogs, RSS feeds, datasets etc.) should be registered in CIARD RING, a global directory of web-based information services and datasets for agricultural research for development (ARD). This will allow other providers of integrated services in several networks to discover, access and re-use the open data of the specific CG Centers. As a next steps, we need to ensure that the Open Access repositories of the CG Centers are harvested by the existing agricultural bibliographic aggregators such as FAO AGRIS. Through a strategic partnership with the Food & Agriculture Organization (FAO) of the United Nations, the Chinese Academy of Agricultural Sciences, and the ARIADNE Foundation, Agro-Know is hosting the Data Processing Unit of the traditional AGRIS service – a global information system collecting and making accessible bibliography on agricultural science and technology – currently referring to almost 8 million bibliographic records from more than 150 institutions from 65 countries.
The Open Access mandate by the CGIAR Consortium provides a lot of opportunities for updating the current infrastructures and workflows of the CG Centers, leading to the enhanced accessibility of their wealth of agrifood research outcomes. At the same time, it is a big responsibility and a challenge, due to the amount of work that needs to be done; however, Agro-Know has the capacity to undertake this task and provide high-quality outcomes that will fully meet the requirements of the CG Centers, regarding their Open Access Implementation plans; in fact, Agro-Know has already started working on the design of the Open Access Plan in the case of ICARDA, one of the 15 CG Centers 🙂