[ 23/11/20 ]
Learn as much as possible about complex systems, evolution, chemistry, quantum mechanics, Bayesian inference, physics, history, and psychology, and then be prepared to dance with eternal uncertainty; as you constantly make micro adjustments to influences on an eternally uncertain future.
In terms of modeling systems, become familiar with as many different classes of modeling system as possible, with a minimum set being: finite element analysis, agent based modeling, MCMC simulations. Learn the strengths and weaknesses of each, and learn what contexts each is most reliable and useful in, and pay particular attention to changes in context; because many systems are very context dependent.
Accept before you start that the systems are more complex than you can possible deal with in detail, and that all approximations you use will contain the possibility of error; and be prepared to adjust strategies and clean up consequences when errors do occur – because they will.
While I have a certain sympathy for that position, I think we do have slightly better than even odds at a positive outcome – right now I’d give it about 60/40 in favour of a future we would be happy with. And that requires that more people start to recognize the very strong relationship between cooperation and freedom, and accept the fact that competition reduces both complexity and freedom.
And it is a very complex problem space, and not very many people seem to be interested in putting in the effort required to get a reasonable sort of feel for the complexities and uncertainties that seem to actually be present. Most seem to prefer the comfort of overly simplistic model, even if they fail periodically (and there is a certain sense in which I can see and acknowledge the evolutionary utility of that class of strategies in out past, I am just not sure that they are survivable in a future with the technological capabilities that we now have, and are rapidly developing).
I think we have gotten to a position in which we as a species can no longer survive total system failure – so we do actually need to put in the effort to gain a reasonable understanding of the complexities present.