Hobby as infrastructure

Reflection on why I invest time and attention in projects that don't generate direct financial return, written partly as justification to myself, and partly as honest consideration of the opportunity calculation involved in this choice, which is not always trivial and is rarely articulated with clarity by whoever is in the same dilemma.

At some point in the last years I stopped feeling embarrassed about dedicating a good part of my extra time to projects that don't bring in any money. This happened after noticing, in retrospect, that practically every interesting opportunity that appeared in my professional life came out of something I did without direct payment, on the loose end of the week, in a project that seemed, at the moment of decision, completely accessory. The formulation I arrived at, and that I try to apply with some criterion since then, is that hobby of the right nature is infrastructure, and treating it as such changes what makes sense to invest in it. Infrastructure doesn't have to pay immediate return, it has to exist for the rest of the work to be possible. The road that connects the farm to the market doesn't sell corn, but without it there is no corn sold. The well-chosen technical hobby fulfills analogous function.

The initial thesis: hobby is rest, work pays off

The dominant premise in contemporary professional culture, even if rarely declared explicitly, operates under the idea that free time should be dedicated to rest, leisure, or relationship, and that work time is the only dimension in which career capital accumulates. That premise is not entirely wrong, given that burnout is real, that attentional fatigue has cumulative costs, and that the social expectation is that healthy professional separates productive time from restorative time. Whoever dedicates a good part of extra time to unpaid project, within that logic, is classified as workaholic or idealist, and that classification is not entirely unfair.

The counter-thesis, however, is that the opposition between productive time and free time supposes that all unpaid activity equals rest, and that supposition is false. There is a subset of unpaid activities that operates as patient capital, accumulates value over time, and produces return on a horizon much longer than the monthly salary cycle. Recognizing that subset, and treating it differently from pure rest, is what opens space for conscious investment in hobby-infrastructure without it being confused with workaholism.

When hobby is infrastructure and when it is distraction

Not every hobby provides this service, and many don't, and I should note that there is nothing wrong with that. The point is not to rationalize every pastime as investment. The point is to recognize that there is a subset of activities that deserves distinct treatment.

The practical criterion I try to apply has three conditions that, ideally, converge.

The first is that the activity develops technical skill with application beyond the hobby. Learning carpentry develops skill that probably stays restricted to carpentry, even if the metaphor can be extended to other contexts. Learning data analysis develops skill that appears in any future role. The difference is not that one is better than the other, it is that the second composes with the rest of the work, and the first stays isolated.

The second is that the activity generates citable public artifact. A repository on GitHub, a blog post, an open study, are all artifacts that can be linked in resume, in professional conversation, in work proposal. Activity that doesn't generate artifact is pure consumption of time, even if pleasurable and legitimate. The generation of artifact is what allows today's investment to be redeemable tomorrow, given that it makes the work visible to the market.

The third is that the activity has long horizon of compounding. Skill that pays a lot in one year and little in five is different from skill that pays little in one year and a lot in five. Compounding is what distinguishes career built from momentary opportunity, and hobby-infrastructure operates under the second regime.

When these three conditions converge, what looks like distraction is, in practice, building of long-term leverage. When they don't converge, it is rest, and rest is also necessary, but not by the same argument.

What counts as return

The most common confusion, in light of the possibility of treating hobby as infrastructure, is measuring return by direct revenue generated. This fails because real return operates in three parallel channels, all delayed relative to investment.

The first channel is the network of qualified contact. Public research on technical topic attracts people interested in the same topic, and those people are exactly who will have useful information about future opportunity. You can't force this network via conventional networking, it forms on its own around the public work that signals genuine interest in the theme. Return appears three years later, in conversation that was born from a shared link.

Next comes the set of opportunities that become visible after crossing certain technical threshold. There is a type of work that is only offered to whoever has demonstrated prior capacity to do it, and the only way to demonstrate that capacity outside of paid work is, eventually, through the hobby. Public research on Brazilian sectoral data doesn't pay, but makes visible to the market the capacity to research Brazilian sectoral data, and that visibility is prerequisite for entire set of opportunities that otherwise simply don't appear.

Last, the shifting of the learning curve. Technical skill developed in hobby keeps accumulating as time passes, and the compound advantage of having started three years before most peers is large enough to invert relative career position. This is hard to visualize prospectively, given that the gain seems small in any individual quarter. Seen retrospectively, in five-year window, it is in the long run the most relevant difference.

The pivot question

It is worth noting that accepting the thesis of hobby as infrastructure does not authorize dedicating time indiscriminately. At what point does the pastime stop being distraction and become investment? How much free time is reasonable to dedicate to something that doesn't produce immediate money, without compromising rest, personal relationship, or more predictable second source of income? Is it worth the opportunity lost in sleep, in family time, in hobbies that serve as genuine rest? Is there real risk of investment in hobby becoming elegant escape from professional problems that would deserve direct attention? And, perhaps the hardest, to what extent is the justification "I am building infrastructure" real argument or comfortable rationalization for a compulsion that serves other emotional needs?

In my view, the reasonable answer recognizes that the cost is real, operates in two dimensions, and deserves explicit consideration before adoption of the practice.

The temporal dimension is obvious, there is finitude of hours per week. The attentional dimension is less obvious but perhaps more relevant, given that the attention dedicated to the hobby is attention subtracted from other things, including other things that could be more important in that period of life. Person with small child, with sick parent, with career in moment of critical transition, should think twice before adding another serious project to the week.

The third way: priority conditioned on stability

In broad terms, the rule I try to apply is that hobby-infrastructure deserves high priority in periods of external stability, and low priority in periods of instability. There has to be surplus attention capital before starting, otherwise, it becomes burden instead of leverage. And I should note that this rule is personal, not universal, given that each life context has variables that exceed any generic recommendation.

There are several projects of mine that fit this logic. Panorama SUSEP is the most visible given that it has public editorial ambition. This notebook is less visible but operates with the same intention, to produce citable textual corpus on themes that interest me technically and professionally, in long horizon, without expectation of immediate return. I write this explicitly given that the justification for allocating time this way is rarely articulated, and tends to be confused with hobby-distraction or with failed attempt to start a business. It is neither one thing nor the other. It is conscious choice to treat technical skill and public visibility as long-term infrastructure, with temporal horizon proportional to what this type of capital demands. Perhaps it would be worth that more professionals articulated this difference for themselves, not to justify every pastime as productive, but to distinguish the pastime that pays in three years from the pastime that just consumes the afternoon, given that the confusion between the two has real cost over the career, even if invisible in any individual quarter.

— Natan, Apr/2026