This kind of post annoys me. It's driven by the fact that the author has some <i>theories</i> about UI design, and wants to share them with us. There's nothing wrong with that. But this post pretends to be like a real usability audit, which it is not.<p>> <i>When I tap it, it turns red — but wait — it doesn’t give me the visual feedback. Did I break it?</i><p>> <i>The alarm is shown as a red filled square, so when I tap it, it should perform a negative action, right?</i><p>These comments are disingenous. You didn't get confused by these things.<p>You, as a person who's analysing a UI, have an almost totally different mindset to someone who is just using the thing for real. You know this, so here, you're just play-acting. You're thinking, what might I get confused about if I was a normal user? That's a bad way to look for what's really wrong in a UI design.<p>Yes, when you present points like these, they sound reasonable, but that doesn't mean they're right. Perhaps the alternative (e.g. banning the use of red anywhere in the OS other than for a negative action) might be over-restrictive as a requirement, which could cause excessive complexity elsewhere in the UI and lead to a worse overall result. Or perhaps you're right, the colour–meaning thing should be 100% consistent above all other considerations. Either way, you don't know, and speculation about what <i>might</i> confuse a real user should not be presented as fact.<p>Theories just guide us in what to <i>test</i>.