Research and case studies dep

Data is the cornerstone of many AgDev projects

For both AgDev projects that are focussed specifically on data use, reuse, sharing and generation, and for those with other primary goals. High-quality data is a necessity for long-term, sustainable, replicable and visible impact. When the power of data is unlocked it enables research-led and informed decision making, enables and fosters innovation. It can help funders, governments, project leads, national partners and actors across the agricultural ecosystem from politicians to smallholder farmers make lasting and effective changes to solve complex global and local problems.

FAIR data principles are a set of tools and ideas designed specifically to improve the way data is used within scientific research and development projects across all sectors, and can bring enormous benefits to help everyone working in the Agricultural Development (AgDev) sector deliver long-term impact.

FAIR data fundamentally aims to improve data access, increasing the data available for AI to learn from. Any government looking to tap into the AI potential for its agricultural ecosystem cannot do so without prioritizing access to high-quality, AI-ready datasets.

At CABI we have looked at how artificial intelligence can help solve global crop loss, and you can read the article by Gaby Oliver here.

On the technical side, FAIR implementation often requires a digital infrastructure that does not yet exist, or standard protocols for data sharing, and agreements in place which ensure interoperability and reusability. On the cultural and political side, there can be a general resistance towards making data openly or more openly available. Data is a very lucrative resource, so entities or individuals or sometimes even governments that own or control or store large amounts of data may be hesitant to make it accessible ‘across the board’ to different actors. They may fear a loss of competitive advantage, or losing influence on how data is used. These attitudes are understandable but can breed a culture of data hoarding, perpetuating data silos and breakdown or fragmentation of existing datasets. This often stems from a misunderstanding on what FAIR implies and what implementing FAIR data practices might entail. For example, FAIR data does not mean data is open to everyone. Open data by definition means that there are no barriers in terms of access or checks or balances around who can access it, and where they can access it. FAIR does make data easier to access and more shareable, but you retain autonomy over who can access it, how and when. This resource explores how to implement these decisions, including through technical controls or specific data-sharing agreements, to ensure that data generated from an investment, for example, can only be used by others in ways you want them to use it, and for specific or defined purposes.