The extra placement freedom is also a tidying cost

What it is

Every panel can be put at almost any point in the room, and that is sold as the dividend of spatial UI. Ten minutes later the same dividend is: finding the last board takes a scan of the room, closed does not mean as gone as on a desktop, and walking back to the original spot is how one discovers it is still there. Spatial housekeeping cost grows with placement freedom — every extra dimension of pose is a dimension that later has to be filed, retrieved, and decided keep-or-not. Freedom is not free layout. It is tidying owed.

Why it happens

Desktop windows are boxed by a screen and a taskbar; the cost of going missing has a cap. In space the pose is continuous, and close does not mean into a known container — often it only means “still on that wall”. Working memory does not store a six-DoF history; people use coarse codes like “somewhere on the left”, which die on a turn. Retrieval becomes search, and search spends the time that space had “freed up” by letting more be placed. Tidying also lacks ready desktop metaphors: no clear-desktop, no virtual desktops, no minimise-all. Without containers (a carry-on drawer, a one-shot gather-in-front, a per-task clear), tidying lands on the user’s grabs and walking, an order of magnitude above desktop effort. The more freedom a product offers, the sooner it must design filing, or the extra points become obstacles in the late session.

Studying it

Let people place several panels freely during a stretch of work, interrupt, then return, comparing presence versus absence of filing containers.

Independent variables: gather-in-front / per-task clear / drawer present or not, panel count, whether the interruption includes a turn or a room change. Dependent variables: time to retrieve each panel after the break, number of panels missed, count of spontaneous tidying walks and grabs, a rating of “the desk is a mess”.

Filing versus piling from personal information management maps on: space defaults to a pile, and the pile is scattered in 3D. Do not measure only satisfaction with the first placement — the dividend shows in the first three minutes, the cost in the tenth.

Where it stops holding

Demos and one-shot experiences barely need tidying; freedom can be maxed. Professionals grow a spatial archive (this wall is reference, that wall is draft) and skill suppresses the cost, but the system still must not scatter their archive after relocalisation. World-locked persistent content survives across sessions, so the cost becomes “the room gets dirtier with use”; body-locked carry-on dirties the personal surround. Children and casual users will not file on their own; defaults should lean to auto-gather, not permanent pins.

Applying it

• For every free placement offered, ship a filing move: a carry-on drawer, a one-shot gather-in-front, a leave-task prompt “pin here or put away”.
• Panels that have sat unread for a stretch get a low-interrupt “still here — bring it in?” Do not vanish them silently, and do not let them occupy that volume forever.
• After relocalisation or a room change, restore containers first, then scattered pins, so the archive does not collapse into a search task.
• How to check: let someone work freely for fifteen minutes, turn away, come back. If time to find each board approaches time to open a new one, freedom has no housekeeping. What can be recalled in one shot from a drawer or from in front of the body is cost that got caught.

Related

• Same group: N3.14.1 Space got larger; the readable angular range did not · N3.14.2 Stacking panels in depth does not increase effective capacity · N3.14.3 Spatial layout has no automatic typesetting; overlap is constrained only by rules
• Nearby: N3.09 Interface Distance and Comfort Viewing Distance · N3.07 Content at the Edge of the Field of View · N5.09 Spatial Anchors and Persistence
• Search terms: spatial housekeeping · layout entropy · retrieve cost