The Initiative for Open Ag Funding supported efforts to end hunger and food insecurity. It ran from February 2016 to December 2017.
The project aimed to improve the availability of timely, comprehensive and comparable information that donors and implementing partners needed to make more effective agricultural investments.
At the start of the project, in-depth interviews were held with a wide range of potential agricultural aid data users, in order to understand their needs.
The simplest questions were often the hardest to answer, such as “who is doing what, where, with whom, and to what effect?”
To help answer those questions more efficiently, the project partners provided donors, implementing partners and other data publishers with technical assistance and tools. The aim was to produce better quality data using the International Aid Transparency Initiative (IATI) common data standard.
The team worked in-depth with two donors – USAID and DFID – to use IATI data while planning an agricultural development project. This developed some important lessons on how open data can improve coordination and planning.
The project concluded that to meet the needs of potential data users, the availability of transparent information must improve across four areas of agricultural work:
The Initiative for Open Ag Funding invested in a range of technical solutions to help donors enhance the quality of their data. They are all open-source and the ‘tool chain’, which links our geocoder, classifier and validator is publicly available. The headline tools are.
Auto-Geocoder: This searches for location names contained within an uploaded project document. Once it has identified potential locations it will ask the user to verify. Once confirmed, the tool will generate and export the relevant geocodes so that the specific location of the project is easy to identify.
Auto-Classifier: This identifies the relevant value-chains contained within an uploaded project document. Once identified, it will return the results to the user alongside the relevant machine-readable AGROVOC codes. As with geocoding, this is a critical piece of information that data users require.
Data Validator: This assesses a donor’s data and report on whether activities and files meet a certain quality. It will provide feedback on where to enhance data, as well as why this information is useful.