The author's big miss, IMO, is discarding the number of design choices that may appear to be aimed at helping new users, but also simplify mobile use-cases (big

The author's big miss, IMO, is discarding the number of design choices that may appear to be aimed at helping new users, but also simplify mobile use-cases (big tap targets, one clear way to get from A to B) and prevent the accumulation of detritus and the effects user self-sabotage, that tend to accumulate on PCs and Android devices. iOS remaining large problems (no 'services', no way to change defaults, awkward inter-app workflows) are unrelated to 'teachability' of the core interface, as they're almost all concerns that only crop up for power-users or normal users who are months or years into their new device. And they're solvable even if Apple clings to the big candy-like buttons, no widgets, skeumorphic app design, etc. So that bit is neither here nor there.