In Defense of Minimalism
Three Things #64: April 16, 2023
I’ve been going through old journals, photos, and notes recently. I noticed a theme: my entire adult life has been very hectic. It’s involved so many people, places, and activities, separated by many thousands of miles and enormous cultural differences. I’ve lived, studied, worked, and traveled all around the world. Professionally I’ve never done one thing at a time. I studied both engineering and liberal arts, then business and international studies. I’ve always had multiple side projects, gigs, and passion projects. I’ve worked in finance, healthcare, and blockchain in roles including founder and CTO, finance and accounting, HR, recruiting and operations, marketing and communications, and R&D. I’ve made friends from all sorts of places and every walk of life, studied foreign languages, sampled food from around the world, earned and spent and lost a fortune.
There’s one common theme that ties together all of this exploration: maximization. I’ve tried to maximize the breadth of my experiences. It’s hard to say exactly where this behavior came from. I think the simplest explanation is optimization, an idea that has always appealed to me as a rational thinker. The world was pretty small and life was pretty quiet growing up. We didn’t have much money so we hardly ever traveled, and dining out was considered a special occasion. I barely tried any ethnic food before college. The fledgling Internet helped somewhat, but the opportunity in my small town for cultural exposure or exploration was close to nil.
So when I found myself at college surrounded for the first time by like-minded, curious people, and tons of unfamiliar food and cultural activities thousands of miles from home, I went a bit mad. I still remember how it felt, attending a student event fair and feeling like I wanted to sign up for basically everything. I double majored. I worked multiple part time jobs. I began to travel. I studied abroad not once but three times.
This behavior continued for the better part of two decades. It represents the longest and most important phase in my life so far, but it’s a phase that finally came to an end recently due to a confluence of factors: the pandemic, marriage and, finally and most importantly, having a child.
In the moment, it’s impossible to know which phase of our lives we’re in. We always need some time and some distance to evaluate and understand how the current moment fits into the tapestry of our lives. In reflecting on my life recently, I’ve come to understand that I’ve turned the corner, completed a phase shift, and now find myself in a new chapter.
This is a good opportunity to reflect on the journey and on the good and the bad consequences that this behavior has brought about in my life these last couple of decades.
Thing #1: Why Maximize?
Maximization seems a logical thing to do when the universe of options and opportunities is large. If you don’t know where you want to live, whom you want to marry, or what your favorite food is, the best way to find out is to try a few candidates. The question is where you stop.
As an engineer I think in terms of algorithms like gradient descent and constrained optimization. Given a set of constraints and a universe of possibilities connected in a graph with a certain cost of moving from point to point, given enough time it’s theoretically possible to find the optimal point. Indeed, every time you run a Google search or scroll a Twitter feed, every time you ask a maps application to plot a route for you, under the hood it’s running an algorithm like this. Of course you want the most relevant search results and the fastest, shortest, cheapest, safest, greenest route from point A to point B.
Why wouldn’t you also want the very best in your life? While finding the perfect cup of coffee or the best slice of pizza might not be essential to happiness, finding the right place to live, the right career, and the right partner probably are. These decisions are big ones that we make a very small number of times in our lives and they’re worth taking time and getting right, ideally the first time (as the cost of switching can be quite high).
Simply put, life is too short for mediocre quality, whether that’s the things we spend our time doing, the food we eat, the places we visit, or most importantly the people we spend our time with. You may need to date dozens of people to find the very best candidate, but as long as you’re constantly improving and dating more and more eligible, compatible candidates, it’s worth taking the time to find the right one. The same is true of work and of places. You may need to work a dozen jobs to find the one you’re passionate about or live in a dozen places to find the one you love. But over time and in the process of exploration you’ll learn about yourself and discover which things matter most to you, and you’ll move closer and closer to perfection.
Why settle for anything less than the best when you can keep improving and keep moving towards perfection? It’s a big world and there are many, many options to choose from. You have every right to be picky.
Thing #2: Everything Has a Price
Everything in life is a tradeoff. Maximization may lead to more optimal outcomes, but at what cost?
The first and most obvious cost is the toll that maximizing takes on us. Someone trying to maximize will by definition never settle. They’ll constantly be looking for something better because they’ll never know if they’ve found the very best, and they’ll constantly second guess their decisions. This constant searching takes a lot of time, but perhaps even more importantly it also takes a psychological toll. While in search mode you feel insecure, anxious, and unsettled. Imagine feeling that way your entire life! (I’m sure some people do, and I feel sorry for those people. They quite literally need help.) Research shows that, while maximizers do on average achieve slightly better results, satisficers (the opposite, i.e., people who are satisfied with what they have) are on the whole much happier.
In the AI world there’s a concept known as exploration vs. exploitation. The idea is that an AI agent can function in these two modes, and it needs to make a decision what percentage of the time to spend exploring the world and what percentage of time to spend making use of the resources it’s already found. If it spends too much time in exploration mode, it’ll run out of time and resources to do any exploitation. On the contrary, if it doesn’t spend enough time in exploration mode, it won’t find enough resources to exploit. It’s a fine balancing act, and it’s truly hard to know when it’s time to switch from exploration to exploitation (or back again).
There are many examples of this dilemma in the human world. Anyone taking advantage of natural resources like oil, lumber, or water has to make precisely the same decision. But so does someone looking for a job, a partner, or anything else important. There’s a parable that sums up the dilemma quite nicely: a man is told to go into the forest, pick only a single flower, but try to find the most perfect flower within one hour. If he picks the first flower he sees, he’s guaranteed to find one, but it’s very unlikely to be an especially nice flower. If he optimizes too hard and tries to find a truly perfect flower, it’s likely he’ll come back empty handed.
There’s real math one can employ here if one is so inclined. There is actually a mathematically precise point at which one should switch modes, depending on a few variables. But this is not how humans typically operate, in practice. As with most things in life we’re far from perfectly rational. We use a variety of heuristics that are messy and difficult to describe, and vary from person to person.
The reality is that the description I posted above, about needing to date 100 people or work dozens of jobs to find the perfect one, is not how dating or careers work. If this description sounds incredibly naive and misguided, that’s because it is: it’s precisely where you end up if you try to put a simplistic algorithm above common sense and experience. People end up in relationships and careers for all sorts of reasons, but with notably few exceptions optimization, in the mathematical sense of the word, doesn’t really play a role.
One problem with optimization is that it by definition requires quantification. You must define the criteria and parameters you’re optimizing for, and then you have to quantify the extent to which each candidate satisfies those criteria. To be thorough you must go a step further and also weight each criterion by priority. You might just be able to do this for a job or a potential place of residence—goodness knows I’ve tried, and I have the spreadsheets to prove it—but I’ve largely given up because it’s a fool’s errand. And trying to quantify a potential partner is even more foolish. People are notoriously bad at knowing and understanding their own preferences in the first place; trying to quantify and systematize the criteria we use to make such big, important decisions simply doesn’t work for most people. In my experience, if you go through the exercise and write down and weight all of your criteria, you’ll meet someone who ticks every box but with whom you simply don’t have chemistry. Then you’ll meet someone with whom you have strong chemistry who doesn’t tick any of the boxes. And it’s simply not possible to rank-order candidates or “constantly date more eligible, compatible partners” as I suggested above. That’s the simple reality of how human relationships work.
Another big problem is that once you get into the habit browsing becomes a hobby. You’ll get used to it and you won’t want to stop doing it, whether that means constantly moving around and living a nomadic lifestyle, doing gig work or consulting work for many clients, or constantly changing romantic partners. It’s far too easy to tell yourself that it’s all in the name of optimization. But the friends I know who have taken this approach to career, dating, or life in general are almost universally unhappy. I don’t know which direction the arrow of causation points here but you can guess for yourself.
The simple reality of human life is that luck doesn’t just happen to us; we create our own luck. Or, to be more precise, we work hard to create the circumstances to take advantage of luck when it strikes unpredictably. This means some degree of browsing and exploring, yes, but it equally means investing: in jobs, relationships, and the other things in our lives. In my experience a successful career or relationship is maybe 5% happenstance (having met the right person or found the right job at the right time) and 95% hard work. Incidentally, this is why I think Asian-style arranged marriages work far better than most Westerners realize. The more you invest in a thing, the better that thing will seem to you. As Julia Child famously put it, “Really, the more I cook, the more I like to cook.”
Finally: it’s essential to make lots of room in your life for serendipity. You can’t overly plan or prepare for everything. If you think that your ideal job needs to involve exposed brick and coworking spaces and free lunch and infused water, you might miss an even better, higher potential opportunity right in front of you that doesn’t happen to tick that arbitrary set of boxes. Same if you think your ideal partner has to be tall, wealthy, well educated, gorgeous, and also into pickleball. The most fun, successful jobs I’ve worked and the most successful relationships I’ve had are often those that ticked few of my predefined boxes, because they brought me joy, surprise, and novelty.
Thing #3: A Recipe for Happiness
As I said above, the maximization phase of my life is over. This new chapter is one of focus and minimalism. It’s manifest in several interesting ways. For one thing, I’m less interested in meeting new people. I’ve met a lot of people and I’m blessed to know tons of good people. There are no major social gaps in my life that aren’t filled, personal or professional. I’m not disinterested in meeting new people—I still enjoy meeting people—but I’m less interested. The bar is higher. When I do plan to meet new people, I try to be more focused about the people I meet. Rather than meeting people just for the sake of meeting new people, I’d rather meet people that I’m likely to have chemistry with, people I can create value for, or vice versa, and people I can build something meaningful with.
Similarly, I’m less interested in travel for the sake of going new places. I used to be excited to jump on a plane and go pretty much anywhere, whether I’ve already been there or not. I lived several years as a nomad, hardly ever coming home. I didn’t think twice about working on a laptop on a dodgy wifi connection in a crowded cafe in a foreign city, or about sleeping a different place each night. I’m still fine doing those things once in a while, but I’m a lot less interested in doing it often. Being home feels nicer and nicer, and I’ve realized how much happier, healthier, and more productive I am at home. I want to get back to a place where going somewhere, especially somewhere far from home, is an event, something I plan for and have a chance to feel excited about, rather than a thing I do weekly without thinking about it too much. I’ve been a lot of places, and it’s not that I’m disinterested in visiting new places but, as with people, the bar is higher. I’m much more inclined to visit a small number of places where I know I can meet incredible people and have an impact, and where the people and the place can have an impact on me.
This is doubly true for my career. I’ve found my calling and it’s unlikely I’ll do anything too radically different in this lifetime. (Whether or not that extends as far as, say, AI, and whether I’ll be able to escape disruption, is an interesting question!) Of course I’m still interested in trying new things and learning new skills, but I’m more interested in mastering the things I’m already good at—going vertical, so to speak—than I am in picking up tons of new skills. A good example is programming languages. After learning around a dozen and using a handful professionally, the bar for me to learn another new language is really, really high. I learned Go a few years ago and now use it regularly. I made an exception for Rust. I honestly hope I never have to learn a new programming language again.
Like “old” people the world over, I’ve found the things and people I like and the things I’m relatively good at, I’m pretty happy sticking with them 80% of the time and I’m trying to minimize everything else. In particular I’m trying to minimize distraction and focus way more on the people and the work that matters. That’s why I’m trying to travel less, spend less time in big cities, and spend less time on social media. Time and attention are by far our scarcest resources, and if we’re able to free them up a bit by removing distraction then we’ll have greater reserves available for the things that really matter and for when important new things and new people show up, which does happen from time to time.
While I’m not overly interested in complex math, I do think straightforward heuristics help. Here’s one that has worked well for me. When I first arrive in a new place, I’m by definition in exploration mode since I don’t know my way around at all. One of the things I’ll inevitably look for is the best sandwich in town. I’ll pick three or four cafes and sandwich shops, visit, and pick the things that look the best on the menu. I’ll probably find one that I really like, and I’ll add that to my repertoire since I’ll want to have it again later. I’ll repeat the process, visiting the place I already know I like and eating the sandwich I know I like around 20% of the time, and trying new places and new sandwiches 80% of the time. Once I’ve found more cafes and more sandwiches I know I like, I’ll grow my repertoire and spend, say, 40% of the time in “exploitation” mode, eating my favorite sandwiches, and the other 60% of the time exploring new options. Eventually, when my repertoire is big enough, I’ll max out at 80% exploitation of known tasty sandwiches, and 20% of the time I’ll still try new things. Note that this means both trying new cafes 20% of the time, but also trying new menu items at my favorite cafes at least 20% of the time. It’s essential to always reserve some time for exploration because cafes come and go, sandwiches change in price and quality, and my own tastes change over time.
The same is true of my career: 80% of the time I try to use tools I’m already familiar with, and 20% of the time I try new tools. Professionally, as well, it’s essential to leave some room for novelty and serendipity.
You don’t have to use precisely the same numbers. And knowing when your repertoire is large enough—whether it consists of past dating partners, programming languages, places visited or lived, or tasty sandwiches—is obviously subjective. But 80/20 is a good starting point and I think this framework is likely to work for most people and most categories of choices.
Start in exploration mode. Gradually transition to exploitation mode as you gain knowledge, experience, and confidence. Find your goldilocks spot: don’t transition too quickly nor too slowly. Allow yourself to be satisfied and don’t second guess your decisions. And minimize, but not down to zero. This is how to reconcile maximization and minimization, and is a pretty good overall recipe for happiness and satisfaction.