This is a short blog post about a narrow UI design decision — icons in macOS menus — that lands a genuinely useful rhetorical point. The author had previously argued against the practice of putting icons in every menu item. When Apple shipped macOS Tahoe with exactly that pattern, the author lost a key piece of ammunition. Now Apple has reversed course in Golden Gate, and the author's argument is sharper than before. The core move is elegant: "Choosing not to do something you've never done is one thing. Choosing not to do something you've done is experience." That's a real insight about how design knowledge accumulates. It's the difference between theoretical objection and empirical rejection. Apple's Human Interface Guidelines now explicitly cite Tahoe's menu icons as an example of what not to do — a rare public self-correction from a company that usually pretends mistakes never happened. John Gruber covered the reversion back in June, noting the updated HIG guidance. The author here is less interested in the design rationale (icon density, visual noise, scanning speed) and more interested in the argumentative leverage the reversal provides. This is a post about persuasion as much as design. The limitation is obvious: this is a micro-post, not a design essay. There's no discussion of why the icons were added in the first place (likely accessibility or discoverability goals), no engagement with the counterargument that some menus genuinely benefit from icons, and no analysis of what Apple learned from the year of living with them. The rhetorical point stands on its own, but it's doing about 30% of the work a full treatment would require. Still, it models something valuable: how to update your priors when a major player runs an experiment you couldn't run yourself. The author doesn't gloat. They note that the reversal is stronger evidence than the original absence, which is epistemically honest. This is the kind of post that circulates among UI designers and developers not because it's deep, but because it gives them a ready-made argument for their next design review. "Apple tried it and backed off" is a sentence that ends meetings.
