Agriculture data sharing: Conceptual tools in the technical toolbox and implementation in the Open Ag Data Alliance framework

Abstract

Data privacy has become a critical issue within the agriculture and food industry. The real-time conversion of data to information has been shown to be incredibly valuable to the industry but often requires sharing data with software, platforms, customers, and regulators outside the data owner's control. While much work in this area has focused on legal protections for data privacy, less has been devoted to technical architectures to support different sharing models. This paper defines some "tools in the toolbox" for designing such systems that are accessible to both technical and nontechnical audiences as well as several "sharing design patterns" using the Open Ag Data Alliance (OADA) framework. These tools and patterns are helpful in classifying and understanding both existing and future data flows in agriculture and their privacy implications.

Big data promises and obstacles: Agricultural data ownership and privacy

“…(2022) and Ault et al. (2022) both highlight frameworks that address a broad range of data related issues. Stacey et al.…”

“…The Open Ag Data Alliance framework described in Ault et al. (2022) provides the means to compare various data sharing models and their privacy implications, which also has the potential to address a wide range of data challenges spanning different disciplines and economic sectors. Wilgenbusch et al.…”

“…Ault et al. (2022) take a different, but complementary approach by describing, “the technical tools (e.g., software code and system architectures) that can be used to model data privacy and sharing issues” rather than focusing on the legal frameworks that undergird them. In a similar manner, Stacey et al.…”

“…Stacey et al describe how the Global Burden of Animal Diseases will facilitate research towards interoperability across livestock datasets, but they also provide examples of how the relevance of this framework transcends this specific domain of research. The Open Ag Data Alliance framework described in Ault et al (2022) provides the means to compare various data sharing models and their privacy implications, which also has the potential to address a wide range of data challenges spanning different disciplines and economic sectors. Wilgenbusch et al (2022) apply their standards-based approach to the CGIAR Platform for Big Data in Agriculture and Responsible Data Guidelines (CGIAR, 2020) to demonstrate how mappings, similar to those used to address HIPAA requirements in the healthcare sector, can address the ambiguity that often accompanies simple data use guidelines.…”

“…They contend that the risk-based approach used in healthcare, if applied more broadly in agriculture, would help to establish trust among stakeholders, which in turn will help to accelerate data sharing. Ault et al (2022) take a different, but complementary approach by describing, "the technical tools (e.g., software code and system architectures) that can be used to model data privacy and sharing issues" rather than focusing on the legal frameworks that undergird them. In a similar manner, Stacey et al (2022) emphasize the importance of data security and extend the Findable-Accessible-Interoperable-Reusable (FAIR) framework to include the "Secure by Design" approach and suggest that FAIR be extended to FAIRS data-sharing framework to explicitly account for security needs.…”