On Meta-Plans: 1

14 July 2020
0146

Planning is hard. There are lots of things to consider.
You have a set of resources, a set of constraints, new information is coming in all the time, short term targets aren’t met, people lose enthusiasm, social dynamics come into play.

Making bad plans is costly. There’s:
– lost time, and material resources
– opportunity cost
– burnt social capital

I’d like to think more about meta-planning, and maybe you’d like to do that with me.
By this I mean: I’d like to learn how to make good plans. In fact, I’d like a plan for how to do it.

Now, when drawing up plans, there are multiple dimensions along which you might try to optimise. For instance:
– the level of pressure that you need to feel in order to perform optimally
– the degree to which you should pre-allocate your expected future resources to a plan of action
– the detail with which you should conceptualise that plan of action
– the frequency with which you should stop to update your plan
– the time you should spend attempting to evaluating all of the above.

A priori, it is not entirely obvious to me what the correct ways to go about determining these are, even though there exist heuristics out there to facilitate this kind of strategic decision-making. (think: OODA Loops or the 70% Rule or even Optimal Stopping theory)

But in time, I hope I’ll write about each of these in-turn.
Nevertheless, it is clear to me that any attempts to grapple with the above will be a function of:
– your temperament (think: risk-aversion, stress-tolerance, ability relative to peers, emotional regulation, adaptability, and even your moral values)
– your objectives (think: timeframe, likelihood of success, externalities, opportunity costs)
– your values (think: what you find meaningful in life now, and how you expect this to evolve)
– your environment (think: your family, friends, intellectual influences, the media you consume)

tldr; I want a plan for drawing up plans.

Post-script
Now, in order to manifest the irony inherent in thinking about thinking, I recognise that it is probably wise for me to commit to thinking up good models of decision-making, but not overcommitting to the ones I already have right now (think: OODA loops or Regret Minimisation or even the dissection of the problem I undertook above) for precisely the same reason that a lot of these self-same models encourage you not to overcommit to plans of action, i.e. to maintain optionality and openness to fresh information.

That is: like all good teachers, these ways of thinking encourage you to question them.

Additional Reading
An interesting essay on the Value of Information.
Commitment vs Flexibility: an influential paper from MIT Economics that outlines a principal-agent formalisation of this problem and comes to the conclusion that minimal-savings policies nearly always characterise the optimal solution set.

i.e. it’s nearly always a good idea to commit a minimal amount of future resources to a course of action and not deviate in the face of temptation until given new information.

It is one of life’s many circularities that we are meant to learn about how to live only by living. And so, I suppose time will tell.

“Life can only be understood backwards, but it must be lived forwards.”
— Soren Kierkegaard