A mathematical model of societal impact

There is power in intellectual diversity.

The notion isn’t well integrated into the human psyche, as far as I’ve observed, perhaps because the language doesn’t exist to describe it in a concise way. People refer vaguely to things as “cross-disciplinary” with a sense of its positive connotation, but why is it a good thing in its essence?

Every concept learned by a person gives them a new supply of metaphors by which to understand the world. I often describe my current approach to life as a “breadth first search” of intellectual territory, drawing the term fro computer science, and I don’t know if I use that phrase to describe to approach or if the phrase enabled me to conceive of the approach.

I’m trying to identify some of the more useful and widely applicable examples of these concepts, to rid them of their domain-specificity and confer their power more widely.

One of these concepts is the idea of building models. It’s a simple idea: simplify a complex system down to something we can understand. What I didn’t realize is the extent to which this simplification can occur while still ending up with a useful abstraction at the end.

In a standard macroeconomics course, people are likely to learn of a model of long-run economic growth that consists of two terms: labor and physical capital. Give these two terms complementary exponents (add to one), multiply them together, and bam! You’ve got long run growth. Not only that but these quantities, which can be defined, conceived of, and measured in an infinite array of ways, are simply represented as letters, L and K.

Forget CEOs, sales and marketing, branding, and individual intelligence and initiative, it all boils down to Ls and Ks. It’s not complicated. I like to think I could have thought of that model.

It’s difficult to stare at a thorny problem, especially ones that involve humans are all their associated degrees of freedom, and try ti distill out the few variables that matter. But sometimes, when living in a world and a society far to complex for any one person to understand in its entirety, models are the only way to stave of paralysis.

As an example, consider the problem I’m currently facing: what classes do I sign up for next semester? There’s a lot of options, and a lot metrics that may be of interest. Am I interested in the material? Do I get credit towards my degree? Will it broaden my thinking significantly? Should I be prioritizing side projects or socialization over academics? Does this class in any way help me achieve self-actualization and lead a productive happy life? Do any of these questions matter?

I’ve decided that my answer to the last question is yes. Nihilism is so lazy, and I see its attraction. Having a clearly stated desire to go forth, make an impact, improve peoples’ lives, and have a legacy is awfully difficult to maintain when you have no clue how to decide on classes, much less the details of your contribution to humanity. Maybe I should go be a ski instructor.

But I decided instead to build a model that can help guide my decisions about my future, and indirectly about my scheduling problem.

I want to accomplish something that improves peoples’ lives. That necessarily must involve a change from the current status quo.

So now we’ve got that impact (I) is somehow related to familiarity with/skepticism of the status quo (Q).

But knowing what the status quo is won’t help anyone actually change it. To enact change, a person needs to have a skill, or preferably multiple skills. They have to be familiar with some subset of the domain of human capability, and the more the better. We’ll call that A.

Let’s look at edge cases. If a person has a lot of A but zero Q (arguably the case for many students at elite universities), their I-value is necessarily zero. They have no idea what exists in the world to be changed. In the same way, having zero A precludes a person from doing anything significant: all they can do is complain about how bad things are.

So I’ll hazard a guess and say they’re multiplicative. That is, I = A * Q.

Like any good model, the ramifications take much more time to write down than the equation. In my case, it dictates that I should take a lot of survey and seminar courses (to rapidly “cover ground” in the realm of human capability) and seize all opportunities to learn about the workings of the “real world”.

The first part of this has been going well, and I trust it will continue to do so. I’ve fully abandoned a “depth first search” further into computer science in favor of a BFS across synthetic biology, optics, microfluidics, artificial intelligence, microcontrollers, cognitive psychology. Whatever it takes to pump up my Q value.

The second part is harder, as there is no formal system like a university for improving A. I seek out conversations (interrogations?) with people whose lives and professions are a mystery to me. What does the minute to minute schedule of a doctor look like? How to architects do their job? How do construction workers delegate tasks?

Separating out the Q-enhancing and A-enhancing activities may be dangerous. After all, it is their cross pollination that gives rise to new and potentially useful ideas. Occasionally I stumble across a new technology that I’m convinced has just begin to percolate into the fabric of human society. One example: short range wireless communication (which is nearly synonymous with Bluetooth), which has just recently become commoditized to the point where $20 and access to Amazon will get you pretty far. In a situation like this, I seize the technology and mentally flail around through the list of major industries. This works pretty well actually.

Of course, the model is fundamentally incorrect. But it’s damn useful.