Has anyone written a UX design article on why cars are designed so that the tank shows "full" at ~90% capacity? I found https://www.npr.org/2011/04/02/135064825/the-gas-gauge-says-full-but-thats-not-quite-true but I'd like more on the psychology being served.

@tdierks I remember when I started working with network file systems and wondered why the Finder polled for changes on AppleShare volumes about every 10 seconds. I was told they’d found that if it were a longer time, users thought it wasn’t responsive (and shorter times used more network and server resources).

It seems like lots of things we use have been designed based on keeping users satisfied.

@jimluther Reminds me of the codewarrior compile-progress screen which made compiles slower (because rendering wasn't always free) but made them feel faster because WATCH IT GO.

@tdierks Remember how the Mac boot progress bar used to crawl along to about 50% and then it was done?

In Mac OS 8.1, I had to change the Startup Manager to keep loading extensions in MacRoman order even though the contents of the directory listings on HFS Plus volumes were in Unicode order (some extensions depended on load order and were named to make them load in MacRoman order). I'd also added a callback before and after each extension was executed so startup management programs, like Conflict Catcher, would continue to work.

Another engineer used those new Startup Manager callbacks to make the progress bar progress more consistently. When those changes went into a build, people testing that build started asking “What changed to make booting my Mac so much faster?!?” It wasn't any faster -- it was just a perception.

We thought about making the progress bar longer based on the display width so that people with bigger displays would get even faster boots 🤣 We didn't do that.

@jimluther @tdierks Whenever a progress spinner animates smoothly, I think ”this is interrupt driven; the process is probably hung.”
@grob @tdierks Look what MPW's beachball cursor did to your brain.