Every decision about motion is a decision about space. When an element enters from the bottom of a screen rather than fading in place, you are making a claim about where it was before it arrived. When a modal slides in from the right, you are asserting a spatial relationship between the current view and the one being revealed. Motion is never neutral — it is always architecture.

The problem with most interface motion is not that it is too fast or too slow, too literal or too abstract. The problem is that it has no spatial model. Elements appear and disappear without any consistent logic about where the off-screen world exists. The result is a kind of spatial anxiety — a feeling that the interface has no depth, no edge, no territory beyond what is immediately visible.

Architecture solved this problem centuries ago. Every building has a spatial logic — a sequence of entry, threshold, room, corridor, and exit — that tells you where you are relative to where you have been and where you might go. Interfaces need the same kind of logic, and motion is the medium through which that logic is expressed.

When we build motion systems, we start not with animation curves or durations, but with a spatial model: where does this content live when it is off-screen? Is it below, above, beside? Is it in a different register — a layer, a level, a mode — that warrants a different kind of transition? The motion should answer these questions, not raise them.