> In Raskin's sense and according to his definition, an interface is not modal as long as the user is fully aware of its current state.
Given this definition, eliminating modes is as simple as making sure that the user is aware of what mode they are in at all times. This makes your assertion problematic:
> To think that you can simplify your model of computers to the point that you can eliminate modes is to simplify your model to the point that it is useless for a great many things.
I don't think you would argue against the notion that users should be made aware of what mode they are in. A pretty common example comes from the CSS `cursor` property:
- If its `default`, clicking should have no effect
- If its `pointer`, clicking should have an effect
- If its `move`, clicking and dragging should have an effect
These cues are far from useless and help transition a modal interface into a modeless one.