“We Are the Force of FOSS4G!”

Survey found that about 50% of geospatial research in CGIAR uses FOSS4G tools, but researchers' engagement with the developers is still low.

At the 2018 FOSS4G (Free and Open Source Software for Geospatial) Conference, our CGIAR-CSI community co-organized a track on the use of FOSS4G in agricultural research and development, together with the Geospatial and Farming Systems Research Consortium at UC Davis. The track included 26 presentations covering a variety of topics, from the mapping of crops and livestock, inequality in access to resources, adoption of technologies, to the analytical studies on the economics of land degradation and fertilizer prices. For the opening of the track, Jawoo Koo from IFPRI, who coordinates the activities of CGIAR-CSI Community, presented “Use of FOSS4G for spurring innovation in agricultural research for development”.

Based on an email survey within the CGIAR-CSI Community, Jawoo reported that about 50% of CGIAR’s geospatial research uses FOSS4G tools. Recognizing that FOSS4G is not always the best option for everything, and the researchers have the freedom of choices (most of the time), Jawoo stated that this is an impressive result — especially compared to a previous round of similar (informal) survey done six years ago. At that time, only about 30% of respondents said to use FOSS4G, especially when the work involved a partner who did not have the licenses to use the same software. Now, however, there is an almost equal level of responses on the use of FOSS4G for both internally in their own work and externally in a collaborative work. Jawoo noted that, by seamlessly using the same research tool and analytical environment with whoever you work with, this is a welcoming trend that helps researchers to work more efficiently.

As a part of the talk, Jawoo invited two colleagues from CGIAR, Francis Muthoni from IITA and Mansoor Leh from IWMI, to share their FOSS4G stories from the fields. Francis develops his own geospatial data analysis tool for targeting agricultural technologies using R and train his local collaborators in Tanzania. Mansoor also uses R to develop his own tool to help farmers identify the best water management strategy in Laos. They both reported that this workflow, where the researcher is the developer, user, and trainer of the tool, helps to produce research outputs more quickly and effectively.

Even with the increased use of FOSS4G, however, the survey found the agricultural research community’s involvement in the development of FOSS4G wasn’t much to report, yet. Only a few respondents said they customized/modified the codes when it was absolutely necessary, to a minor extent (e.g., translation into a local language), and they were not shared back with others. Researchers also rarely provided useful feedback to the developers, except there was a real inconvenience (e.g., reporting critical bugs). Based on this, Jawoo concluded the talk with a call to action: Let’s engage and give back (more) to the FOSS4G Community — to further contribute to and improve the tools we love and use every day!