Knowledgable Nudges
Earlier this week I was tackling some Advent of Code [1] puzzles from 2023. After being stuck for a few hours with day 12’s problem I took a peek online and found I was 90% on the right track, just missing one small optimisation. Also this week, a game called Kerbal Space Program (KSP) had its sequel all but cancelled and commentators are pointing at poor project management. Specifically, not utilising the old KSP team at all. Both of these are examples where a knowledgeable nudge can go a long way.
KSP is a game about building rockets and going to space and has realistic orbital mechanics (XKCD [2]). Orbital mechanics can be quite tricky to solve if you don’t know which methods to use (e.g. [3]). Add to this the game also simulates the solid mechanics of the rockets themselves which is different difficult problem in itself. It’s a game that’s built on difficult, but not insurmountable, problems. The old team had been through all this and figured out ways to overcome these problems, it is confusing as to why the publisher forbid them from even knowing about the sequel. Months of experimentation and testing could be skipped if a little bit of that hard won knowledge could be accessed. It might not have saved the project, but that lack of knowledge certainly contributed to the massive schedule overrun and eventual (apparent) shelving of the project.
In the Advent of Code problem, I spent hours pondering how to tackle the problem in a new way. When I looked at the discourse online, people repeatedly mentioned one small optimisation; I was sceptical that it would be enough but when added to my solution it ran near instantly. The optimisation was memoization [4] where you remember parts of a problem so you don’t have to resolve those parts if encountered again. For one of the puzzle problems there were 552 parts, and 355 could be used twice, the computational savings were literally astronomical in size. If no one had told me this simple optimisation would save that much effort I don’t know how many days it would take me to find it, if ever.
Sometimes, you can’t just push through a problem, you need a tip or a bit of advice from outside to turn an impossibly difficult problem into one that you can solve. If you’ve ever heard of the 10x worker, this is the sort of thing that could give someone that sort of impact. The above anecdotes are technical, but this applies equally well for interpersonal and strategic problems as well. We should be ready to ask for advice from mentors or colleagues when we get stuck as they may have just the bit of knowledge we need.
If you’re working on a thorny analytical problem, say how to schedule work, or how to figure out bottlenecks in your operations, send me an email (Hello@NorthCardinal.com.au) and I might be able to give you the nudge needed to get moving again.
[1] https://adventofcode.com/2023
[2] https://xkcd.com/1356/
[3] https://en.wikipedia.org/wiki/Leapfrog_integration
[4] https://en.wikipedia.org/wiki/Memoization