Increasingly, multi-agent systems are being designed for a variety of compl
ex, dynamic domains. Effective agent interactions in such domains raise som
e of the most fundamental research challenges for agent-based systems, in t
eamwork, multi-agent learning and agent modelling. The RoboCup research ini
tiative, particularly the simulation league, has been proposed to pursue su
ch multi-agent research challenges, using the common testbed of simulation
soccer. Despite the significant popularity of RoboCup within the research c
ommunity, general lessons have not often been extracted from participation
in RoboCup. This is what we attempt to do here. We have fielded two teams,
ISIS97 and ISIS98, in RoboCup competitions. These teams have been in the to
p four teams in these competitions. We compare the teams, and attempt to an
alyze and generalize the lessons learned. This analysis reveals several sur
prises, pointing out lessons for teamwork and for multi-agent learning.