The weight of our decisions #

Designing isn't simple #

Individual agency exists even if we don’t program for it; but as we document data and feed more of it through our information technology, where we don’t support agency becomes an ersatz morality police that most of us don’t even intend. Most of it is simply the quick regurgitation of how one or a handful of people have figured out their path through a problem set, and then spent their time trying to get it programmed.

The reality is that the fewer people who were involved in the testing and figuring out of the process, the more likely it’s a finite slice of potential rather than a composite. The problems are twofold: the complexity of the process, and the willingness of people to run with outputs when they don’t understand the logic of the process. Small, finite processes mostly don’t do to badly — or fail so blatantly that they are hot-fixed. Complex processes are more prone to automation bias, in turn leading to all kinds of pervasive algorithmic problems.  

People are variable: in what they see as the solution, in how they design for and around it, and in how they consume provided solutions.

In some spaces, people just want to get shit done, and the simple option is all they want: screw agency. In other spaces, people feel the oppression of having their who-ness diminished or outright removed.

Those spaces often overlap. They can nest. They can be exactly the same spaces in every respect, but that one person feels it like a dirty job done quickly and reasonably painlessly, and one person feels it like the weight of an authoritarian boot. It can even be the same person at different points in time. Wherever two extremes like that exist, it almost always is a spectrum node: extremes defining the edges, with a potentially infinite number of variations between.

So how do we decide, we people who are building, what to include? While acknowledging that we can’t put several lifetime’s worth of effort into a week’s work? During a time when we are building optimal solution frameworks from scratch, because we figured out a societal breath ago that we were doing some fuckuperry? 

To date, we’ve mostly been working towards sharing information and building tools. We’ve butt up against the network and started to feel the overwhelm of transitioning information constructs. Where once the ubiquity of hierarchical information dispersal was accepted, now the limits of that construct are being felt, even if people can’t point to it and slap their foreheads in suprised acknowledgement.

Some people see their favorite stability points being moved and get angry. Some people see new connections and insights that are a joy. Both are interpretations and reactions to the same, unchanged, data.

Those ‘some people’ are often the same person, leveraging different aspects of their mind on tools. It’s growing pains. The pain sucks, but it’s part of growing. Ignoring it won’t make it go away; we have to work our way through it.

And in today's (2025) political landscape, I feel it necessary to point out: growing pains don't starve people, don't make them sicker than a generation ago, and don't keep pruning information until it only works their way.

That’s insight into the pain point, but it still doesn’t answer the question of how we decide what to build.

From one point of view it’s a culture question. In a prioritized capitalist culture, money is the driving force. If it doesn’t make a person some money, there’s no point in doing a thing. If it makes a person some money, that’s all the reason necessary for doing it.  But that highlights the underlying reason for this book: humanity lives beyond the confines of a single culture; we live beyond the exigencies of a singular priority. There is no singular, defining right with everything else as wrong. People are complex. Systems are involved that we don’t even recognize yet, let alone the systems that we do recognize but barely understand. 

We’re designing for people, as people, in all our people-y ways. The labels and who-ness that are attracted to certain roles change through time, space, and cultures. I have been a content manager, UX designer, UX engineer, and information architect, predicated on company culture, all while doing many of the same tasks, and definitely using the same skillsets and working towards these emerging models. 

We will always be people, though. In our strengths and weaknesses, emotions and trust, and in all our complicated cognitions: always people. 

If we design for what we want (a common decision-making tool), we’re forgetting the infinite possibility of people. 

If we design for a mean (like “best practice”), we’re forgetting that we cannot assume the nuances of one are the reflection of everyone, and are confabulating that the mean of everyone is the only available truth for one.

If we design to bend people to ‘higher’ will — for money, for power, for sensibility — people will eventually be in so much pain that can’t be gaslit away, they’ll get rid of it. Thwarting agency doesn’t stop the exodus, it only slows it down. 

If we design shortsightedly, it will eventually be thrown out for the longer vision – or we accept dying as a species, like with climate change. 

If we design only for the cool factor of showing what is possible, the unintended ramifications will bite us in the ass, like generative AI will unless it stops prioritizing “sounding human" above "valid, non-hallucinatory information, gaslighting as a solution to being questioned.”

In the gestalt, people generally want to get along. In the gestalt, people won’t actually put up with having their pain ignored, especially if they see others who can successfully neutralize the same pain. 

Intentional design is not easy. It doesn’t always look the same, follow the same process, and have the cadence of a precision timepiece. It’s rarely one and done — and the ones that managed to be close to one and done are sweated over. They don’t follow an arbitrary clock constructed to meet the wishes of profit. They go down rabbit holes, check gnarly math, pull in and digest consilience data.

Our most innovative ideas come from minds that don’t stick to their lane. They stew on and balance data through time and permutations, question everything, and then construct hypotheticals around what actually works. Innovation is the output of lived scientific methodology; often without a paper trail, but with a super-high reality adherence. 

This is almost the opposite of capitalism as it currently functions. 

The people and businesses who follow — spreading newly understood information, sometimes building on it, sometimes conflating their own biases, sometimes misunderstanding to an egregious fuckuppery — are part of our whole. Every emulation is a statement of “oooh, I think you’re on to something.” This is where capitalism is right now. We're focusing on shortcutting innovation to opportunistically develop what someone else was too stupid to see as being valuable.

As information and process is understood, it can be implemented closer and closer to clockwork. But the mimic will rarely have true innovation. It’s not looking forward, it’s not exploring. The closer something can stick to a predefined schedule, the less innovation is involved. The less innovation is involved, the more it’s mimicking — sometimes with some jazz riffs. The faster the schedule is, the less thought goes into the output. 

We can research, test, and design towards the group we intend to service. It will always leave out edge cases as a starting point, but it’s still a working model that at least is definable and can be iterated upon to expand and eventually include edge cases.  Sometimes, just changing “edge case” to being about something other than populations, we can design something more deliberately inclusive. Edge-cases previously dismissed can become the expanded features, or the garbage data, or the elasticity of the solution. All we’re doing is edge-casing a difficult solution that we haven’t thought of yet. 

We can find our gaps, iterate again, and keep at it until we finally achieve a holistic tool. It will develop an information structure problem over time, but that’s part of building. The trick is to recognize it and fix the infrastructure before it collapses. 

Think of it like a growth and quality. We, in our short-sighted goals for quarterly profits, focus on the likes of radishes (20-30 days), maybe some corn (80-90 days). They feed us, absolutely. The calories and nutrients exist and are useful. It's useful, but an overproduced crop will still struggle to find outlets. And that's only for as long as a blight doesn't spread through acres of radish/corn monoculture.

Vintners understand that many of the best wines come out of old-growth vines that were planted with the intention of longevity. In vintners circles, 3 years is still a young vine — an age-to-profit that can’t be reconciled with the likes of radishes and corn. And while it’s possible that there will be a complex wine that comes out of that for a single, stellar season, it’s not good bet. The good wine bets are from vines that have been thriving for decades, and are growing in a complex environment. The environment affects the undertone scents and flavors of the grapes.

Adhering to a singular goal, a singular answer, or a singular environment is simple. It's not the answer to get to long-term viability, or nuanced expression.

There are roads to travel, mistakes to find and fix, more mistakes to make, countless misunderstandings to correct, and understandings and systems to recognize. We’re not really getting the difference between culture and society, so ‘society’ will likely continue to be conflated.

In other words, there’s not a new paradigm in here. It’s just a reminder: people are complex. What we design will ultimately be, in some far-off future, just as complex. If there is a want for a thing to be one of the functional parts of that future, remember to build for people as people. Adhere to reality, not wish states, and the work might be built upon.