3System Farming
Bruce Edmonds
3.1 Introduction
Consider farmers. They may know their animals, crops and land in some detail, but are under no illusion that designing a farm in advance would contribute more than a small part to its success. Rather, they understand that they have to be acting upon the elements of their farm constantly to try to get acceptable results – less an exercise in careful planning than disaster avoidance using constant monitoring and continual maintenance. In such a situation, new ideas cannot be assessed on the grounds of reason and plausibility alone (even those suggested by scientific theory), but have to be tried out in situ. Solutions are almost never permanent and universal, but rather a series of tactics that work for different periods of time in particular circumstances. Techniques that are seen to provide real benefit (even if that benefit is marginal) are adopted by others nearby so that, in the longer run, a community of specific and local practices evolves. This chapter suggests that in order to effectively produce complex socio‐technical systems (CSS) that work, we have to learn to be more like farmers and less like mathematicians or traditional engineers.1
This is a difficult lesson to adsorb. As students, we are taught the value of good design – constructing the system sufficiently carefully so that we know in advance that it will work well once built. We know that bugs are bad and we can limit them by careful specification and implementation. ...