To improve in a craft, extensive practice over an extended duration is typically imperative. Malcolm Gladwell claims that we need at least 10,000 hours of practice to become masters of our work. Although there are objections on the relationship between practice hours and the degree of excellence, it is undeniable that repeated practice allows us to trust our body and appreciate our unbelievable ability to adapt and learn unconsciously.

With the investment in time and persistence, we can correct ourselves bit by bit. It's a mini marvel of evolution at work when we engage in trial and error, practicing continuously and embracing mistakes that bring improvements that spring up from time to time.

To be more specific, I think we can discern three categories of the trial-and-error process, or different ways to improve, apart from the hard work just mentioned.

01 / Low-Cost Sandbox loops

The first category is defined by a virtual or low-stakes framework where mistakes carry zero physical friction or material consequences. In digital environments, code, or sandbox simulation spaces, the cost of an error approaches zero, and feedback loops are near-instantaneous. Here, rapid, uninhibited trial and error is a highly efficient strategy, allowing the practitioner to stumble forward safely and uncover creative solutions through sheer iteration speed.

02 / Moderate Friction loops (Tough Joy)

The second category involves individual physical mastery with a high safety margin but clear physical friction. This is beautifully illustrated by activities like skateboarding. When practicing a new trick, mistakes hurt—they result in falls, bruises, and occasional cuts—but they rarely carry permanent or catastrophic risk. This friction generates a form of "tough joy." The immediate physical feedback forces the body and mind to calibrate, learning unconsciously from mistakes until the movement patterns gradually align and the trick is conquered.

03 / Zero-Tolerance Environments

The third category shifts dramatically into high-stakes, unforgiving environments where mistakes must be avoided at all costs. Safety margins and measures, such as a diving partner, an additional regulator, and constant monitoring of oxygen levels, are indispensable in mitigating inherent risks.

In these realms, we actively minimize the trial-and-error process and learn predominantly by following the correct procedures slowly and repeatedly. Often, the only instance that an error proves useful is in the aftermath of an accident, when we have no means but to learn painstakingly at an irrevocable cost. It is for this reason that pilots and engineers are extremely risk-averse, as they operate in an unforgiving environment.

// Strategic Alignment

In conclusion, a smart person should grasp the essence of their craft and choose the appropriate way to improve. Reed Hastings echoes a similar sentiment in his book No Rules Rule. He reckons that we should apply specific strategies for different industries. For industries aiming at the prevention of error, stringent rules and processes are required to eliminate mistakes. Conversely, for the innovation industry, mistakes are imperative as they invite rapid improvements and new ideas.

From now on, start to think and choose wisely between stumbling and staying untouched. Choose the correct way to improve.