The Perils of Local Maxima
Stuck on a Mountain
If you’ve practiced something for a long time, you may find yourself in a rut. Technically proficient though you may be, you have not reached your full potential.
You have been here for a while now, but do not worry: there are ways to escape the island you are on.
This is “local maxima” problem, and it’s an issue with AI, LLM’s and philosophy. Neat
This problem has been around for time immemorial, so the antique punching bag I’ll choose today is Aristotle.
That nerd thought he had everything figured out, but his so-called “Principleof theMean” has a glaring weakness.
Once you find a “best practice,” or you win locals, or you become the village hero, you peak. There’s a character trope of the “peaked in High School” type for this.
It’s a common thing, but how can we push beyond our limits?
How do Limits get Made
If you plan on growing or improving something, be it a skill, resource, or other capability then you eventually run into limits.
Limits are sometimes created from the environment. As an example, if you’re trying to get stronger and improve your capacity to deadlift, then you can’t realistically expect to lift much past 500 kg: even getting there is probably out of range. However, humans were not always able to lift anywhere close to that - so what happened?
In this example, there’s a confluence of circumstances. First, the current world record holders are huge guys. Eddie Hall and Hafþór Júlíus Björnsson are both over 6 foot tall men, who are absolutely massive. I think it’s safe to say that a large number of people this large and fit haven’t really existed until relatively recently. Second, the world relatively recently changed to be fairly open, international, and highly diverse.
It probably wouldn’t have been possible to set the current world record for deadlifting until after globalization, and industrialization. I argue these two things, because there’s 2 simple sources for getting stuck at local maxima:
- Physical Constraints (true maxima*)
- Lack of competition / diversity
Stickiness is due to the tendency to ingrain, and settle in.
Grace Hooper, one of the creators of my field (Coding / Computer Science etc.) said it well:
Humans are allergic to change. They love to say “We’ve always done it this way.” I try to fight that.
So fight you must: to break the local maxima you must break the landscape.
Shaking it Up: leaving locals
One of the most obvious ways to mix it up, and shake the landscape is to go outside.
If you’re the best at what you do where you do it, then change where you are. Chances are, you’re not in the premier place for your community. Even if you are, there’s probably somewhere at least comprable. E.g. if you’re a coder in San Fransisco, you could still go to Seattle, or even Boston and visit to get some idea of other groups and areas.
Another great way to expand is to go to conferences and meetups. You don’t need to flex. You don’t need to present, you don’t even need to really have a goal. You just have to go, be friendly, and meet a few people, and see new things. The convention doesn’t even have to be directly associated with your niche: If you’re not into security, check out a DEFCON (or a local equivalent); if you’re a web programmer, check out a Cpp con. If you’re bored at one of the meetups or conventions, take a left turn and go to a convention you haven’t dabbled in, or a field you haven’t explored - you don’t need to convert
just go
Find Mentors and Heroes - Emulate them
If you’re bottoming out in “tutorial hell,” sometimes it takes an example of someone who already made it.
If you’re doing games, maybe don’t immediately try to imitate John Carmack, but it wouldn’t hurt to read up on how he got his chops, try some things he’s tried and aspire.
Finding aspirational figures is one way to realize if you’ve hit a local maxima. If someone else has gotten there before you, then it means it’s possible.
If you can, find a mentor that will push you off of your local maxima. Usually this means that you’ll find someone who’s better than you are, and ask for their coach, team lead, teacher etc. or them.
Don’t ask John Carmack to teach you how to code.
Something that’s cool about mentorship is that your teacher or mentor doesn’t have to be better than you to be helpful. Top athletes have trainers, who paradoxically cannot be top athletes: you can find a mentor / coach that can push you.
When to Stay Put
There’s an important prerequisite to breaking the mold: I said at the top of this post that you need to want to grow.
One cannot grow forever, and growth isn’t a virtue, but I’ve found myself to enjoy the grind of improvement, challenge, and reaching for more difficult projects and more strenuous tasks. I really like coding, and I really like making things - so I challenge myself at those things.
If you do not like to code, or if you don’t find the idea of challenge interesting, then you don’t have to, and you’re not worse for it.
I sincerely mean it. On the other hand, if you’ve found yourself in a rut, sometimes it just takes the thought of “I could do that,” to get out.