Digital Ecosystems as Levers for Agroecological Redesign

Making agriculture more sustainable asks for more than new agronomic techniques: it asks for a rethinking of the organisational and relational architectures that hold agri-food territories together. This contribution examines how integrated digital ecosystems act on that architecture, and takes bio-districts as its setting — multi-actor arrangements that already gather producers, consumers, local authorities and researchers around a shared territorial vision.

Four structural barriers keep that vision from becoming widespread practice: information fragmentation, which leaves cropping choices to imitation and past prices; high transaction costs along chains concentrated at their intermediate stages; the absence of a shared monitoring framework, which leaves territorial governance without a verifiable evidence base; and digital dependency, where proprietary platforms move data and decision-making power outside the territory. Hence the research question: to what extent can integrated digital ecosystems lower those barriers and, at the same time, reshape producer–consumer relationships toward co-created agroecological value?

The thesis is that designing digital infrastructure is a governance matter before a technical one, and it is pursued along three axes: shared farm notebooks in a common format, aggregated at district scale into a hub of collective knowledge; direct producer–consumer channels that return margin and information to the production stage; and digital product passports carrying agroecological performance indicators drawn from records the farm already keeps.

The tools are built in a territorial living lab, refined round after round with researchers, farmers and local communities, against six principles of digital inclusion. The first tangible outcome is doliFARM, an open-source digital ecosystem whose architecture was derived inductively from coordination problems documented in the field; releasing it under free-software licences is itself a governance provision, since it sets the conditions under which a territory can inspect, modify and re-deploy its own infrastructure.

Future work will define outcome indicators together with bio-district actors and collect data in pilot bio-districts against a pre-adoption baseline. The design is formative: with a limited number of cases and no control group, no causal inference is intended. Whether digital infrastructure supports or constrains the agroecological project depends on its governance — open licensing, collective control of data, adherence to principles of digital inclusion — rather than on its functional performance.