Our aspirations are pessimistic for the fast diffusion of complex and deman
ding information technology (IT) aids and decision support systems (DSSs) a
mong farmers. This view arises from some results of the new institutional e
conomics, recent results of empirical decision research, data from farmers
applications of decision models, as well as experiences introducing farm-le
vel DSSs by our own working group at Giessen. For some areas of decision ma
king, one can only hope that the use of models heightens problem awareness
by the user, thus providing additional insights into the usually complex de
cision space. If, however, farmers want to increase the economic efficiency
of their production and marketing processes by decreasing waste and fricti
on, then we must continue to develop and refine knowledge-based DSSs. Probl
em selection, however, should be driven by new results of empirical decisio
n research. Because imperfect knowledge exists regarding the input-output r
elations of agricultural production systems, as well as about the time- and
space-variant uncontrollable input variables, close co-operation with the
biological disciplines of agricultural science is needed. Multi-disciplinar
y research and end user orientation seem to be the key factors for further
progress. (C) 2001 Elsevier Science B.V. All rights reserved.