If you’re using Safari browser today, with every page load, you see the progress bar that I somewhat accidentally developed.

Let me explain.

Back in 2012, we wanted to create a better, more fluid version of the progress bar. Our current progress bar was discrete and somewhat outdated, not aligning with a more modern world where fluid animations were the norm.

So, we did some research and came across a paper where researchers looked into how different progress bar styles affect what people think and how they feel about them.

One cool thing we found out is that people like progress bars that start slow and then speed up. This type of progress bar creates an impression of increasing speed and reduces the anxiety associated with waiting.

Taking this psychological trick into consideration, I started hacking on a demo.

The problem was that loading a website is non-deterministic, meaning you don’t really know exactly when the page has fully loaded. This made it challenging to implement the speed-up effect at the right time for the progress bar, which we jokingly called the rocket effect.

My colleagues from WebKit helped me with the trigger for the rocket effect, but it was still up to me to animate the progress bar continuously while the page was loading.

I literally wrote a simplistic, even dumb, algorithm as a temporary solution just to have something working.

We showed it to Don Melton, the father of Safari. He liked the idea and told us to continue working on it.

Now, I needed to rewrite my simplistic code and make it more aligned with the realities of progress loading.

I worked on it for two more weeks, doing some fancy calculus, logarithmic functions, exponentials, and generally covering my whiteboard with equations. It was more “factual,” but somehow it didn’t feel better than the first version we did.

At this point, it didn’t even cross my mind to use the first version of the progress bar we developed. But after showing Don our best “factual” version, he hit the table and said, “That’s not how it fucking used to look!” I wasn’t sure what to do at that point. Then he added, “I just want it to look the same way you showed me the first time.”

I was puzzled, but we complied. I returned to the most simplistic version of it, and we shipped it. Users loved it, and I still can’t believe after all these years that it works as well as it does.

The most successful solutions often involve psychological considerations. It’s not just the reality that matters; it’s how people perceive it.

Rory Sutherland talks about this in one of the best TED talks ever made, “Perspective is everything.”

At our new company, Topics, we apply the same principles to our language learning software. It might not be objectively different, but we’ve designed it so people don’t feel like they’re learning a language.