The progress of progress bars

I was waiting for Claude to answer a token-heavy question at work the other day, the little “Thinking…” label flickering into “Wibbling…” and then “Moseying…”, and I caught myself doing the thing I always do when a spinner takes too long, staring at it, trying to divine meaning from the word choice. Is it nearly done? There’s no percentage, no bar, just a word standing in for “trust me, something is happening.” It sent me down a rabbit hole into where progress bars came from, and it turns out they were never really about progress at all. In 1985, Brad Myers ran a study at the University of Toronto. He gave participants identical tasks on identical computers (some with a progress bar during the wait, some without). Users who saw the bar rated the computer as faster and more trustworthy, even though the wait was identical. More tellingly, the bar didn’t even need to be accurate: users preferred an inaccurate one to no bar at all. From the beginning, the progress bar was about anxiety…