Many conversations about open data for agriculture and nutrition promote the win-win scenario of improved livelihoods for farmers, as well as more nutritious, environmentally conscious food. However, examples of open data benefiting farmers often only span one growing season, or include small groups of farmers. This begs the question, does open data truly have the capacity to trigger transformative change in agriculture?
Data exists on a spectrum, which ranges from closed, to shared, to open. Shared data can only legally be shared with certain individuals or groups, due to data ethics recommendations. Just as the food system is comprised of several actors, such as input providers, farmers, retailers and policymakers, who make decisions that affect both others and themselves, the data ecosystem comprises of data collectors, data re-users, data subjects and others. Most actors in the food system fulfil multiple roles within the data ecosystem. For example, a farmer may have data collected about them (data subject), re-use government data (data re-user), and collect data about their farm (data collector). Similarly, a start-up who creates an app for healthy food choices may use the data the farmer has collected (data re-user), and collect their own data on the use of their app (data collector). All of these actors exchange data, as well as information and knowledge, to help everyone make better decisions, whether it is choosing healthier foods or knowing when to plant seeds and harvest crops.
Farmers benefit from data sharing, both directly and indirectly. Many initiatives seek to put data directly in the hands of the farmer, through mobile phone services or various apps, for example, mFarmer and Esoko. While these projects may provide benefits on a small scale, there is not much empirical evidence for long-term impact, especially disaggregated by user type (gender, etc.), and many do not consider how they as institutions can collaborate with other organisations to improve indirect benefits to farmers.
A recent consultation completed by the Centre for Agriculture and Bioscience International, Global Open Data for Agriculture and Nutrition (GODAN), and the Open Data Institute for the Gates Foundation, showed that a lack of coordination and data sharing among institutions (including donors, governments and the private sector) has reduced the speed, efficiency and quality of extension services to farmers. Although there were several small initiatives that worked closely with farmers throughout the duration of the project, there was a general consensus that country-wide support for farmers could greatly improve if data was better shared between large agricultural institutions.
Ecosystem maps show the data and knowledge flows that can help farmers to receive the most sustainable and long-lasting benefits. Farmers are an important source of data, but extra attention should be paid to how other actors in the sector must cooperate in order to provide maximum benefit to farmers and their families.
Farmers will derive benefits from sharing data with a local project, as long as their data is managed ethically and responsibly. Maximum benefit can be achieved when data management and sharing best practices are adhered to throughout the data ecosystem. Additionally, the benefits must always be considered over time. Improvements in harvests or incomes over one growing season are admirable and valuable. However, interventions that are continuously monitored over a longer time period, allow for flexibility and scalability over time and this helps to build stronger, more resilient communities.
This opinon piece by Ruthie Musker, GODAN, is cross-posted from CTA’s Spore Online Magazine. Technical Centre for Agricultural and Rural Cooperation (CTA) and Global Open Data for Agriculture and Nutrition (GODAN) are Partners in GFAR.
Image at top: An agronomy data ecosystem map created by FAIR data in BMGF Agronomy Programmes.
1 thought on “Collaboration is key to achieving the long-term benefits of data sharing”
Reblogged this on Dr. B. A. USMAN's Blog.