The paradox of getting better

When doing something, we all strive to do it better, to improve our knowledge on that specific task so that we can be valued for those new fancy skills. This is applicable to everything, we want to improve in work, we want to improve in our hobbies, in sport, in beliefs, in relationships, in everything.

This seems a fairly innocent premise, because at the end, why should becoming better be something bad?

The problem comes when we’ve reached a somewhat high level and want to keep improving. The cost of getting better is much higher, we’re in the flat section of the learning curve, and we might even wonder if it’s worth pursuing that costly level. Knowing that will differentiate us from others, push us to the forward of the distribution of people who are acquainted with the subject, but cost us time, effort and probably most people won’t even appreciate what it is that you are pursuing.

On the other hand, when we know a lot about a subject learning about it becomes much easier, so our learning rate around the topic also increases. I consider that a better metric to assess how much we can learn about something is given by the product of the two, where the area under the curve is an indicative of how much we know about the topic.

The profile of those curves must vary wildly from one person to another, but I believe that the average curve has the following regions:

1. It starts slowly as we need to get familiar with the very basics of the topic (definitions, jargon, rules, etc.)
2. It speeds up to a (probably global) maximum, as we are comfortable with the initial definitions but there are still plenty of things to be learnt.
3. Learning things becomes harder but we learn fast, so the curve is somewhat constant (perhaps increasing a bit at first and decreasing later). Little insights or help from mentors of others have an intertial impact on the curves that manifests as bumps on the constant section
4. The curve slowly decreases as learning something becomes very hard but teaches us little. Eventually, the curve goes to zero, either because we know all there is to know or learning becomes prohibitely expensive.

How hard it is for me to learn something about the topic based on what I know?

How difficult are the new things that I learn?

Product of the two above. How much can I learn right now. The area under the curve relates to how much I know

Another way to look at this paradox is observing that when we get better at something there is also an associate risk that’s growing. For example, when we get better at climbing, we attempt to climb harder things, where a fall would cause more severe injuries, but at the same time we know better how not to fall. In the workplace we might become better at coding so we are assigned to more complicated and serious projects, where an error might compromise the whole project, the system or raise security concerns; but (hopefully) we also know how not to do this. In those examples, the chances of getting seriously injured or fucking up the system are almost constant over time.

I believe that it is important to be aware of the point in the curve we are, as otherwise we are either about to die (fex, if I were to try an Alex Honnold free solo today) or learning far from our optimal rate (fex, if I attended programming 101).

Note: In here I’ve considered learning as useful learning, that is learning that directly drives an advancement in the subject. That is why at the end the curve goes to zero.