I like this although I'm going to stick my neck out and say the author is objectively wrong that sqrt is better than cubic for a toggle switch, and that cubic i

I like this although I'm going to stick my neck out and say the author is objectively wrong that sqrt is better than cubic for a toggle switch, and that cubic is in fact the better choice for this situation simply because of the way real life toggle switches generally work. Think about the breaker switches on the electrical consumer unit on your house, or the kind of switches you often find on analog synths or other audio gear, generally stuff that's going for a particular aesthetic. I have a small Hughes & Kettner guitar amp that has two of these switches that are very satisfying to use[0]. When you use these kinds of switches in real life there's some initial resistance, and then they suddenly snap into the new position because of the way they're spring loaded. This is more closely modelled by the cubic function than it is by either the sqrt or the exponential smoothing function. Other than that absolute nitpick, I really enjoyed the piece. It well illustrates how animation can enhance a user experience when done well (with, for example, an appropriate easing function) or alternatively detract from it and be quite jarring if not implemented thoughtfully (thinking particularly of the linear interpolation examples). [0] This does somewhat depend on the type of toggle switch, although even the variety that you find on something like a Minimoog (this isn't some sort of gear flex, by the way - I don't own one of these!) exhibit this "resist then snap to the new position" behaviour that makes them so enjoyable to flick.