Dear Hugo, I think the mentioned article raises an important issue, and yes I think that we need to broaden the scope beyond open data while discussing these issues because it is not always clear on what part of the data spectrum (open-closed and everything in between) organisations will be able to operate.

Dear Suryadi and others,

We welcome everyone to join the GODAN initiative, you can fill in a short webform and you will be added as a partners soon. It is a free membership and we encourage our members to publish their data a any suitable data platform.

We do not have a data repository ourselves but we encourage our partners to publish on their own platform or on the platform of choice. For research data we can mention the Open Data Journal for Agricultural Research (www.ODJAR.org) and our own GODAN Gateway on F1000Reseach. The first is a data focused journal and the second is focused on more general research that specifically uses open data or generates open data. On the F1000 Gateway we also encourage our partners to publish material that is discussion various topic on open data in agriculture and research. 

I you are publishing your data on a specific subject and there is a good use case we would love to hear more on it so we can spread the word.

Warm regards,

Ben Schaap

I agree with your analysis. There are indeed few initiatives that actually try to move towards data coops. One that I know of is the Cool Farm Alliance, they add value to products accross the value chain by being transparent about the GHG emissions at farm level. The data sharing is demand driven, so buyers ask producers to submit the data. 

However, there other existing coops that are working on concepts of data sharing for the benefit of their members. And yes, your three questions are valid and need to take into consideration. Re: 1) I think that this could be the role of existing coops, they need to explore how sharing data and using other open data sources can enhance their position as a coop. 2) This is an interesting idea. 3) This is a crucial question for science and society. If we want to enhance the models and improve the quality of the tools we need to pool the data as smart as possible. The Findable, Accessible, Interoperable and Reusable (FAIR) principles provide a good framework for sustainable data sharing and will help to grow an existing pool of data. However, this requires that the collection of the data is done in a structured way (eg.: by using unique identifiers at farm level?). Could ICT's play a role here to structure the data collection more?