Different parts of one interface can use different locking modes

What it is

A canvas pinned to a wall, a tool strip at the hip, battery in the corner of the view — not three products, one interface binding different parts to different frames. Mixed reference frames let world-lock, body-lock, and head-lock coexist in one layout, chosen per part’s job, rather than picking a single default for the whole UI. What each lock solves is not the topic here. The question is whether, once mixed, the relations among parts remain readable.

Why it happens

Jobs map onto frames: content that must align with room objects and be pointed at by others follows the world; content that should travel with the person but not yaw with every head turn follows the body; content that must be glanceable without occupying a place in the room follows the head. The cost of mixing is that parts no longer share a rigid transform. On a turn, the hip tools come along, the wall canvas stays, and the spatial sentence “the tools were on the left of the canvas” breaks. Taking an item from a body-locked strip onto a world-locked canvas is a frame-to-frame carry: grasp, transition, release. Without a carry path, people think the tool vanished, or that something on the canvas was taken with them. A head-locked layer also covers silhouettes of the other two, so grouping is misread.

Studying it

Run the same task in three layouts: all world-locked, all body-locked, and canvas-world plus tools-body plus status-head. Record time, retrieval errors after a turn, and false reports that a control is “gone”.

Independent variables: assignment of frames to parts, yaw amplitude, whether a grab-and-pin path is provided. Dependent variables: task time, first-localisation accuracy after the turn, success of cross-frame drag-and-drop, a rating of “still one UI or scattered”.

Egocentric / exocentric switching tasks in the reference-frame literature can be rewritten at widget grain rather than scene grain. Collaboration should be run separately: a body-locked strip is invisible to a bystander and changes deixis.

Where it stops holding

With few parts and no turning, the break from mixing is unmeasurable and a single frame is cheaper. A 3DoF device has no stable world origin, so do not include a world-locked branch in the mix. When the body is planted in a seat and barely walks, world-lock and body-lock feel similar to the wearer — “two metres ahead” versus “on that wall” — but still differ for someone who enters the room later. Blind users do not group by these visual frames.

Applying it

• Assign by job: environment-aligned content to the world; carry-on tools to the body; standing readable status to the head — and provide an explicit carry (grab down, pin up) among the three.
• Do not let the silhouette of a head-locked layer cut the hit areas of the other two.
• For a flow that must survive a turn, do not leave the next button on a world canvas the person has walked away from unless that button itself is re-parented to the body.
• How to check: after a stretch of work, turn 180° and ask, for each part, where it is and whether it can still be used. Every wrong point or “it’s gone” is a mix that failed to state the relation.

Related

• Same group: N3.08.2 A change of locking mode needs a visible transition · N3.08.3 Lazy follow lets body-lock keep up without shaking · N3.08.4 Only world-locked content can be jointly pointed at by several people
• Nearby: N3.01 Locking Modes · N2.07 The Body as Reference Frame · N5.05 Shared Space
• Search terms: mixed reference frames · heterogeneous locking · egocentric exocentric