Profile photo
Sunny WuUofT Computer Engineering

Equilibrium is Optimal

June 19, 2026

The most, the best, the worst, yes, no. You correlate extremes with success but it's never so clear cut. People are bigger than the system they're in and act in response to what the system opens to them, not what it tells them to do.


If you had to slow people down driving their road tanks (cars) that can go way faster than they need to (also caused by the Law), you'd have a hard problem to solve but an interesting application of psychology.

Yeah road engineers want to hard cap speed so people don't drive too fast because surely all drivers will follow it despite all the open road ahead. Obviously it doesn't work: Speed limits are more like speed suggestions, and the speed setters take into account drivers going a bit over the limit.

Even on the Ontario G test you're encouraged to exceed the speed limit with the examiner telling you to go the flow of traffic.

However, if an officer was explicitly tasked with watching speeders, they'd do something. But not if they were just going about their business.

A cop sees you jaywalking in downtown Toronto - they don't care. They'd do the same if they weren't in uniform.

Setting hard rules and enforcing them softly is also common at the office. Imagine you're at your desk doing something in the grey zone like using AI tools, and your supervisor passes by. They glance at your screen. They won't call you out but it's not like they would encourage you to do it when the rules say "NO GENERATIVE AI." And don't ask them if you can do it because they would have to say no to save themselves.

Following the principle, Portugal significantly decreased drug use and drug-related deaths. You can outlaw drugs, but only with an impossible amount of enforcement could you fully stop their use. By decriminalizing drugs instead, Portugal admit that drug abuse existed despite the laws, but invited those struggling with it to seek help.

In 2001, their drug-related death rates were similar to the EU average but plummetted and remained well below the EU average upon the change. A compromise that really wasn't.


Naive businesses don't survive because they forget that the people react. Only on the surface it makes sense: Maximize prices and minimize costs through quality to fulfill the mission of optimizing for money.

More realistic models like Spotify's try to be attractive enough for enough people to not pirate it. That's a real option for their customers and something many streaming services struggle with. Not many people who download their own music; Spotify's done a good job on this end. They provide a complete library of music and a good user experience.

But Netflix, which grew early by beating pirate sites, has been struggling.

How would you increase your streaming company's profit margin? Put in more ads and lock video quality to higher tiers and avoid costly deals with big shows.

You don't want to put up with that as a customer when piracy sites give you ALL those benefits with a smoother experience.

The consumer, the streaming service, and the piracy site all have some leverage over one another. But not enough. So although the streaming company ruthlesslessly seeks to increase shareholder value, they must accept the truth that a push for more is rarely without its costs.

>>  For the curious