We had the pleasure of participating in the 3rd Joint National Conference on Information Technology in Agriculture at the University of Ruhuna, Mapalana Campus, on the 29th of December 2011. Papers were presented by scholars in ICT (primarily from U of Moratuwa) addressing agriculture problems and by scholars in agriculture (primarily from U of Ruhuna) that had ICT either as the instrument (e.g., sensing technologies) or as object of study (e.g., how do farmers obtain ag information disseminated through various ICT and other modes?). The interdisciplinary cooperation is leading to a joint MSc program in agricultural informatics, we were told.

While senior scholars were listed as co-authors, all the presentations were made by young scholars, including some based on undergrad thesis work. The encouragement given to young people to present their research in a peer-review setting was praiseworthy. It would however not be a bad idea to put a little more effort into preparing young scholars to make effective research presentations, something that we at LIRNEasia do through the tutorials and coaching we provide in the context of CPRsouth.

Interdisciplinary work is great, but requires more effort either in the form of the researcher really taking the trouble to speak multiple disciplinary “languages” or, the more logical route of working in teams, which seemed to be the favored mode among those who gave papers on the 29th. The trouble is that agriculture is not really a single discipline, containing within it as many disciplines as a small-sized university. So, for example, I felt that economics could have been better represented in the mix and that, in fact, some papers suffered from the lack of economics.

This was the point I chose to emphasize when I was invited to make comments at the closing panel. Those looking at the agricultural supply chain would be well served by using concepts from economics such as transaction costs, without engaging in the fashionable but counter-productive bashing of “middlemen.” Of course what is intriguing is the possibility of using “agents” (software entities that operate with preset rules) that seemed to reflect a lot of research going on at Moratuwa. Now that could reduce transaction costs in a way simple economics cannot.

Research on farmers’ information needs and the means they use to meet those needs is interesting to us, especially because we have been studying the subject for some time, the most recent effort being by Kapugama, Lokanathan and Perera. The research reported on at the conference had been done in one village in Sri Lanka. The paper says simple random sampling was used, while the abstract says it was a purposive sample. I tried to get it clarified at the conference but failed. Another paper read at CPRsouth7 on a related set of research questions had serious problems with sampling as well.

Nevertheless, the findings were congruent with our findings about farmers relying mostly on their peers, preferring voice to text and so on. It went further, to show that the government’s Nenasalas were of no relevance to farmers. But the recommendation was that more resources be put into Nenasalas.

This is a problem we experience at CPRsouth too. Research shows something has failed. Most often, the recommendation is to put more money into it. Rarely does anyone recommend that the plug be pulled. Why?