My first time going to a library study room, I was confused on how to get out because I didn’t know you had to click a button to leave. Although it has a sign that says “click to exit,” I was still lost because other study rooms on campus didn’t have this, and the button is on the side of the door where you usually don’t look, something Norman calls a classic signifier failure. The more I used the study rooms, I found myself still struggling as sometimes the button to leave or the ID scanner to get inside just don’t work. When I swipe my ID and nothing happens I have no idea if my card failed, the reader failed, or I did something wrong, which is exactly the feedback problem Norman describes. Getting in and getting out also use two completely different interactions with no consistent logic and it breaks any conceptual model I try to build of how the door works.
I think I can apply Norman’s principles to my own work in p5.js sketches. One thing I want to start doing is making it clearer to users how they are supposed to interact with my sketches, whether they need to click, long press, or drag to interact with my piece. Right now someone could open my sketch and have no idea what to do with it, which is the same signifier problem Norman talks about with doors. I want to incorporate small visual cues like a pulsing cursor, an animated hand icon, or a brief on-screen hint that disappears after a few seconds to guide the user naturally into the interaction without overwhelming them with instructions.