so sometimes it's just l l l l l, but other times that does a different movement, depending on what mode you're in? how many different things can l l l l l do? are there situations where two of them look the same so you might continue to believe you are in a different mode than you are?

this kind of thing might be a good ux tradeoff but I think you'd have to validate it empirically to see how much the extra cognitive load slows user task completion