🌿  back

20.87 YRS

thinking practices

In knowledge work, tasks often vary in terms of how reliant they are on external feedback. For me, writing articles like this one is mostly a matter of reflecting on ideas in my mind, and then later putting them together in a more coherent form. When shaping the essence of these articles, I don’t rely on feedback from external systems that much, as I’m mostly running thought experiments and rephrasing ideas in the workshop of the mind. At the other end of the spectrum, programming is way more reliant on external feedback from systems like compilers, linters, and CI/CD pipelines. There is a constant back and forth between developer and the IDE, which would make it impractical to deeply reflect on each and every issue before addressing it. One might argue that the human mind hasn’t evolved to think in code and software architecture, but it’s doing a pretty good job with language. This lack of innate specialized tooling forces us to create and use external aids.

A benefit of using external systems (e.g. thoughtware) is that it’s us who built them, rather than nature, which means that we have a good understanding of how they work. Many would argue that the brain is truly a masterpiece of the blind watchmaker, but Feynman would readily point out that we don’t understand what we can’t create. Having a complete understanding of the systems we are in sync with during knowledge work has the benefit of being able to indirectly quantify and structure our work through them. You can run unit tests on a function you’ve defined in Python, but it’s tricky to determine if a set of ideas you’ve been cultivating in your mind is ripe for being persisted in an article. Besides measurable goals, processes mediated by external systems also benefit from having subtasks which you can point to and sets of components which you can refer to by name. Unfortunately, this lure of systematicity which comes with using external systems makes us overly reluctant to embrace and encourage deep thought, a fundamentally internal practice.

To drive the point home, just think of the contrast between highly external tasks (e.g. programming, modeling, designing) and highly internal tasks (e.g. thinking, reflecting, envisioning). There’s nothing unusual about programming for an hour, but it feels pretty weird to describe your next task as reflecting for an hour. Among the factors which contribute to our reluctance to embrace deep thought are the lack of metrics available to gauge it, and the non-trivial connection it has to practical outcomes. A less immediate but still informative way of understanding this internal-external spectrum of tasks is to try to sketch out some tasks as flowcharts, a process similar to the task analyses conducted in cognitive ergonomics. Tasks structured by external systems tend to be easier to elaborate on at a granular level (e.g. Subtask 2.3.1: Compile code.) compared to the less concrete internal tasks (e.g. Task: Reflect?).

But then what of these practices which we engage in on a largely cognitive level? How can we normalize, promote, and support them? This line of inquiry is especially relevant when considering thinking which is not directly related to measurable outcomes in a trivial way – deep thought conducted as an end in itself, a mental equivalent of blue skies research. Who knows how much conceptual gold we could find if only we expanded the definition of knowledge work to include open-ended reflection?

One way to better align deep thought to the ideology of modern productivity is to engineer some lightweight structure and metrics around it. For instance, what if you had a queue of open-ended questions you wanted to reflect on and your task was to come up with a certain number of answers to each one of them? The queue would have a certain size, and you would enter in specific answers you can point to – already some peace of mind for the analytical among us. Additional constraints might force you to specifically come up with diverse or surprising answers. But such a piece of thoughtware would still be light and general, allowing you to engage in your thinking practice both in the peacefulness of a forest and while dancing to instrumental music at home. Automatically pipe your resulting insights into your conceptarium, analyze them through the lens of the ideoscope, and you’ve already got yourself a minimal IDE for deep thought.