The Engineering of Error - or what I learned with Charles Watson

What is The Engineering of Error?

I recently watched some online lectures by Charles Watson, whom Victor Pontes had always urged me to check out. Victor suggested Watson because he approaches creativity from a very practical perspective, often demystifying the concept. It’s interesting because most people tend to think of creativity as either artistic expression or a search for something. However, Watson also celebrates creativity as a form of work. Out of all the lectures I watched, one that particularly caught my attention was titled “The Engineering of Error.”

Watson is an Englishman who has lived in Brazil for many years, and I found myself wondering how best to translate that title — whether “engineering” should be paired with “failure” or “error,” and what “error” truly means in this context. The first lecture began with a Leonard Cohen song, which includes the refrain,

[!quote] “Ring the bells that still can ring. Forget your perfect offering, there is a crack in everything, that’s how the light gets in.”

This idea of a “crack” intrigued me, some idea of disruption maybe… The second lecture ended with a discussion on restriction, featuring insights from renowned choreographers as Lia Rodrigues and Deborah Kolker, who view restriction as something that aids the creative process.

What struck me is how Watson begins with a song about imperfection and ends with a discussion on restriction? It made me curious about the true meaning of “the engineering of error”, what the stages of such a process might be?

From Mistakes to Events: Miles Davis and the Art of Improvisation

Watson also introduced an interesting concept with his story about Miles Davis. In the story, one of Davis’s musicians plays a note that was, let’s say, a mistake — something unplanned. However, Davis continued improvising on that note, making it work in the context of the performance.

Instead of calling it a mistake or a failure, he refered to it as an “event.” This shifts the perspective from seeing it as an error to viewing it as an event, perhaps even as a form of chance or serendipity. But I don’t think this fully captures what Watson means by “the engineering of error,” so I found myself questioning what an error truly is?

An error, after all, is often defined as a deviation from what is correct, expected, or intended. This implies that there’s a path leading to what is considered correct or intended.

I began to think about measurement tools, especially in physics, where there’s often a defined margin of error in the experiments. For instance, anyone who has had to cut wood and adjust their technique along the way knows that the measurements don’t always turn out exactly the same, because each measuring tool has its own precision of error.

Precision vs. Accuracy: The Lessons of the Dartboard

Reflecting on my experiences, I recalled a lesson from a computer engineering class where a dartboard was used to illustrate precision and accuracy. The first dart might hit the center of the target, while the second goes outside, and the third lands somewhere on one of the rings. In a second attempt, all the darts might land in roughly the same spot, but not exactly at the center.

This scenario illustrated how something can be precise but not accurate. This idea connects back to the definitions of error in statistics, where an error is the difference between a measured or observed value and the true value.

By weaving all these ideas together, I’m exploring how error can be seen not just as a deviation but as a vital part of the creative process, something that can be engineered, worked with, and even celebrated. Perhaps “the engineering of error” is about understanding and utilizing these deviations — these cracks — as opportunities for creativity and growth.

What initially caught my attention was the title of this lecture series, “The Engineering of Error.” Engineering is the practice of using natural sciences, mathematics, and the engineering design process to solve technical problems, increase efficiency and productivity, and improve systems. This gives me the sense that engineering is a field where knowledge is applied, in a way, to solve problems.

These are usually technical and practical problems, involving mechanisms or mechanics. Typically, we associate engineering with structure, machinery, apparatus, or manufacturing processes.

So, when I think of engineering as a process, a machinery, a development of structures, or a technical solution, it makes me wonder how one could mechanize error? The title intrigued me because it suggests that there could be a process — a workflow, so to speak — for error.

Watson hints at this idea of empirical mathematics by bringing up examples like how many rockets SpaceX had to launch before they successfully recoveries the rockets. According to him, they had 20 failed attempts before they achieved their goal. And all of these attempts were documented.

Here you have a company, and a figure closely tied to it — Elon Musk— narrating the story of these 20 failed attempts to reach their objective. There’s an intention, and they failed in that intention, which could be considered an error, a deviation from the goal.

Then there’s the element of risk; you think, “Wow, 20 failed attempts—what makes us think the 21st will succeed?” Watson mentions something about creativity being a significant interplay between the familiar and the unknown. That’s the space where you operate.

Embracing Restriction: The Boundary Between the Known and the Unknown

It made me question whether there’s a method to error in this context. Is there an engineering, a design process, for error? If we think of it this way, what does “The Engineering of Error” imply? If error is a deviation, then is engineering a process to help you when you deviate, or perhaps even to help you deviate intentionally? It’s fascinating to think about whether there’s a structured method or design to managing errors, especially in creative or innovative endeavors.

When I think about error as a deviation and engineering as a process to help when that deviation occurs, it makes sense to understand this deviation as an “event.” The first step is to accept, embrace, and move on from it. Now, if we’re talking about intentionally creating error, perhaps it involves rejecting the notion of an expectation or a “correct” outcome. Though I’m not entirely sure, it seems that in either case — whether dealing with error or entering a process without a clear reference point, where anything could be an error or nothing is an error because there’s no deviation — you need something crucial: repertoire.

Watson doesn’t focus heavily on this, but it’s something that stayed with me. Documenting those 20 failed rocket landings at SpaceX isn’t just about hope that the 21st attempt will succeed — it’s about study and learning. Repertoire is key; it’s something invaluable. It’s about connecting all the things I’ve learned and applying my own experience. Trusting my repertoire, trusting my experience — whether it’s physical experience, prototyping experience, or life experience — helps in making decisions.

This also makes me think of statistics, where there’s something called systematic error, which is often caused by bias. It could be that I consistently err when I’m attached to my biases. But beyond that, it seems that the framework of “The Engineering of Error” is about respecting experience and repertoire. This trust in experience and repertoire is perhaps why Watson ends by talking about restriction. In a way, he’s presenting restriction as an opportunity to use one’s experience and repertoire.

It’s almost like an embrace because restriction creates a boundary, it makes the limits clear. I think that’s it — restriction clarifies the boundary between what is familiar and what is unknown. You trust in your experience and repertoire, and then you begin to try. Maybe this is less about the engineering of error and more about the engineering of trying or the engineering of experiments. In the end, summarizing all this for myself, I think I embrace error as a try-out.