Sharing a screen mid-meeting is a glimpse into someone’s psyche. Exactly how many tabs, spread across how many windows, does this person distribute their thoughts? Is their desktop a chaotic mountain of icons, or devoid of icons entirely? Is the background an image of family, a place they would rather be, or did they even bother to change the default background?

On these observations, I am wholly comfortable being observed. My tab count varies widely, but is typically best described in dozens. I usually have my tabs distributed across 2-5 windows. My background is both a photo of my family and a place I love to be. Whatever this tells you about me…great (and maybe let me know? I’m genuinely curious.)

But I am not an open book, as there remains something I loathe for you to see, and will avoid to the greatest extent possible. I hate typing on screenshare.

I’m a perfectly competent typist, my form is generally good though not perfect (no doubt a byproduct of my mother running the school computer lab during the formative years, which included typing classes). But the moment I am being directly observed, my typo rate skyrockets. I go from being someone who can type at a rate close to my speaking rate to someone who looks like a monkey trying to happenstance their way into a Shakespearean sonnet.

I sometimes wonder if typing on a typewriter would help cure me of this, similar to the thoughtfulness of a fountain pen finally giving me a cursive script worthy of correspondence. The modern keyboard layout was, after all, optimized for typewriters. Perhaps by going backwards, I would find a new way forward.

How QWERTY came to be

Early typewriters had a tendency to jam. Specifically, if adjacent keys were struck in quick succession, the typebars would collide and get stuck. This created a challenge for letters adjacent in the alphabet that are frequently used next to one another. Letter combinations like S-T and D-E were a problem, as well as some that appear further apart in the alphabet but still found themselves too close in the mechanism. The solution was to redesign the keyboard layout so that the typebars of commonly paired letters were placed further apart, reducing the frequency of jams. This is how the QWERTY keyboard came to be.

There’s a common misconception that QWERTY was designed to slow typists down. But in fact, the opposite is true. By reducing jams, the QWERTY keyboard actually increased the functional typing speed on the typewriters it was designed for. It was a genuine solution to a real problem.

But modern computers have no typebars. There is nothing to jam. Other keyboard layouts, like Dvorak and Colemak, are better matched to how our fingers move and how often letters appear (in English). By most measures, they’re faster and less fatiguing. So why do computers still ship with a QWERTY keyboard?

Switching keyboard layouts means weeks or months of typing more slowly before you return to your prior speed, and longer still before you see any gains. In the modern context, the tradeoff would probably be worth it in the long run. But few people make it.

Computers were first produced with QWERTY keyboards because, in the grand scheme of everything involved in switching from a typewriter to a computer, changing how to type was an unnecessary distraction. It would have kept people away from computers. Even in the computer-native workplaces of today, the perceived inconvenience is just too high. Of course, you can understand the resistance. After all, how many other keyboard layouts have you tried?

In Success in Product Markets and Nature, I talked about how evolutionary success, like product success, is ultimately determined by context. What I didn’t say then is that this type of context isn’t passive. It’s a system of interconnected parts that can determine what success can even look like.

The pressure to match that existing context is so high that it explains phenomena far beyond keyboard layouts, from the success of Netflix to the rise of mammals, and can be used to intuitively understand so much more.

The Context Was Already Built When Netflix Arrived

Netflix launched in 1997 as a DVD-by-mail service. At this time, broadband internet was a big step up from the dial-up tones of just a few years before, but still far from sufficient to enable video streaming. Blockbuster was the market leader and had the advantage that customers could walk into the store and walk out with their movies in just a few minutes. But the supply of movies was limited to what a given store could display, and there were only so many copies of each movie. Netflix competed by having a larger library to rent from, and although you had to be willing to wait a couple of days, you didn’t actually have to lose your home.

We look back and see Blockbuster and Netflix as highly differentiated products, but they both fill the same context. People want to watch movies at home, with good video and audio quality, usually on a large screen, and with as little friction as possible.

People often talk about Blockbuster’s decision not to buy Netflix in 2000 as a missed opportunity. But it wasn’t so obvious at the time. There was still no streaming, and perhaps had there been an acquisition, Netflix would never have reached this potential. In 2000, Netflix was a DVD-by-mail service operating at a loss, while Blockbuster still had customers coming to its stores daily. The decision to add a warehouse-based mail service was not a clear-cut value proposition.

But as internet speeds increased through the early 2000s, piracy platforms created a new opportunity. That core desire of watching a movie at home could now be satisfied without leaving the home, a la Netflix, and on the same day, a la Blockbuster. Before streaming was possible, these opportunistic platforms fulfilled a consumer desire with a previously impossible combination of speed and convenience.

There were, of course, the initial moral arguments against streaming platforms, and over the course of a few years, court decisions made it harder and harder to access these platforms without risk to the end user. But over these few years, internet speeds and reliability continued to improve, and a new possibility emerged: streaming.

By 2007, the demand was no longer in question; years of piracy had made it clear that consumers would go to considerable lengths for convenient home viewing. Netflix launched its streaming service as a legal way for consumers to have instant access to movies they wanted to watch. And after years of losing revenue to piracy, the major studios had an incentive to get on board.

While we see the product difference between Blockbuster and Netflix as quite large, it’s important to understand that the core context never changed. The service that Netflix provides today is one that has to match this enduring context. We often speak about the transition to streaming services as though the ecosystem of consumer entertainment fundamentally changed, but actually, the ecosystem changed very little. What changed is the technology that fills the role of content delivery in that ecosystem.

The Ecosystem That Survived the Asteroid

We tend to make a similar fallacy when we talk about the rise of mammals after the extinction of dinosaurs. In soaring language, we describe the downfall of the cold-blooded titans, followed by the rise of the warm-blooded mammals. But like describing Blockbuster and Netflix, that’s not quite what happened.

Estimates suggest roughly 75% of all species went extinct following the Chicxulub impactor. That’s catastrophic. But while the total species loss was huge, a recent study concludes that the ecosystems these species were a part of were largely preserved. The underlying structure of how life organized itself, which niches existed, how energy moved through the system, and the basic architecture of ecological relationships, survived even as the dominant players disappeared.

Think of it like a significant round of layoffs. The number of employees drops dramatically, but the culture of the organization, the way teams relate to one another, the unwritten rules of how work actually gets done, these tend to persist in ways that can surprise even the people still there. When hiring resumes, the roles aren’t identical. The new hires are still entering a context shaped by what came before, and that context shapes what success looks like for them.

Mammals had been present throughout the age of dinosaurs, constrained to small niches that the dominant species hadn’t occupied. When the extinction event removed those dominant players, mammals inherited a world still structured by 165 million years of dinosaur-shaped ecology. The available niches, the prey species, the environmental pressures: all of it bore the imprint of what had been.

Just as cacti wouldn’t have evolved in the rainforest, the mammals that diversified after the extinction were shaped by that context. They filled recognizable versions of the same ecological roles the dinosaurs had occupied, through different biological strategies, fitting into a structure that was already there.

The dinosaurs shaped the mammals that replaced them. Having built the context that persisted after they were gone was enough.

The Same Pattern, Different Scales

This pattern shows up at every scale of biological organization, not just across millions of years of evolutionary history.

Your body is doing a version of this right now. Homeostasis, the process by which the body maintains stable internal conditions like temperature, blood sugar, and hormone levels, is an active regulation organized around reference points established over time. These systems have evolved to maintain stability. When you try to shift something, your weight, for example, you’re moving a configuration that dozens of systems are reinforcing to stay the same. Weight loss is hard-earned because these systems respond with increased intensity of hunger sensations, conservation of energy (you simply don’t want to move), and even a slowdown in your metabolism. It takes more than a few weeks or months of habit change to override a context developed and reinforced for years.

The same applies at the behavioral scale. The standard advice for changing habits is to change the context, not just the behavior. Habit stacking, when you connect a new behavior to an existing one, works because it inserts something new into an established context rather than fighting it. People who successfully maintain sobriety don’t continue to hang out with their dealer. They change the context.

The longer a context has been in place, the more energy it takes to shift. A habit formed last year is more vulnerable than one formed over a decade. A culture two years old is easier to change than one reinforced over twenty. The longer a pattern has persisted, the more context is actively reinforcing it. That isn’t stubbornness. It’s what resilience looks like from the inside.

Change the Context to Change the Outcome

When something won’t change the way you expect, a parent’s health habits formed over fifty years, your own metabolism after months of effort, an organization that keeps reverting to old patterns, remember that what you’re focusing on is the symptom of a context. The context is what’s holding it in place. The question worth asking is “what is the context still reinforcing, and has that context actually changed?”

Change is only possible in the proper context. Change the context to change the outcome.