A mid-air vocabulary must be small enough to count and easy enough to find

What it is

A mid-air vocabulary must have a boundary people can count and an entrance they can find. Adding “one more cool move” without bound grows discovery cost and neighbour confusion together, until nobody remembers the whole set. Bounded means both count and shape-distance have a ceiling. Discoverable means someone who does not know the whole set can still find where to look, and which items are live now. A keyboard has visible keycaps. Mid-air does not. The boundary has to be designed; people cannot exhaustively search the air. Fit decides whether to use gestures at all. Vocabulary size decides whether, once used, they can still be learned to the end.

Why it happens

Working memory will not hold dozens of arbitrary moves with no physical labels. Each added item must sit far enough in shape from existing ones, or they confuse. Teaching and in-place hints have finite area; items past that area become hidden commands. Experts want more shortcuts; the public quits after the third. Bounded is not an arbitrary “seven max.” It is: the discovery path (legend, practice, first lesson) covers the whole set, and neighbour distance holds inside the recognizer. The typical discoverability failure: a function appeared in a demo, and in the field there is no entrance that shows that move again.

Studying it

Vocabulary size as a factor. Measure time to first discovery, retention at a week, neighbour confusion, and people's estimate of “how many items are there.” Legend present or absent as a second factor. Experts and novices separately. Elicitation tells you which moves people want. It does not tell you whether the whole set is still discoverable—that is a separate measurement. Count how many functions exist only on a gesture path with no entrance, as hidden-command count.

Where it stops holding

Games may expand the set on purpose as unlockable progress; discovery is the content. Professional systems (stage control) may open up for experts, but still need a searchable legend, not only oral tradition. When system-level and app-level gestures stack, people see one larger unbounded set; the cap has to sit at system level. User-defined vocabularies ease memory and push discovery onto the user; the product still has to offer browse and search.

Applying it

• Ship the vocabulary as a printable list, sized so a legend can show it at once. Surplus goes in a menu, not in the air.
• The field must always be able to open a legend or short video of currently live moves. Ban “you saw the demo, you should remember.”
• Accept boundedness on novice discovery time and “how many did you think there were.” An estimate far below the actual set means too many hidden commands.

Related

• Same group: C4.14.2 Similar shapes must not carry very different consequences · C4.14.3 Gesture meaning is validated with target users, not designer intuition · C4.14.4 Learned gestures must be relearnable on demand
• Adjacent: C4.01 Definition and fit of mid-air gestures · C4.05 Three dimensions of gesture evaluation
• Search: gesture vocabulary · discoverability · set size