Command lines and gestures depend on recall and need visible entry points
Aliases: discoverability of hidden commands · gesture affordance · unaided interaction modes
What it is
Command-line input and gesture-based interaction share a structural trait: before the user acts, the screen typically shows nothing telling them what they can do right now or how to do it. The user must first pull the exact command spelling or gesture path out of memory before they can execute anything. This is the opposite of an interaction style that surfaces a set of visible options to pick from — command lines and gestures are inherently recall-driven, not recognition-driven.
Why it happens
This recall-dependent structure can't be tuned away with a minor interface adjustment — it's a property of the interaction form itself. A blank command prompt, or a gesture performed in empty space, carries no visual cue at all about "what options are available" unless the designer deliberately adds an extra layer of guidance. Without that layer, the user has no choice but to generate the command or motion sequence completely and accurately from memory alone — exactly the kind of interaction that taxes memory most heavily. When it fails, the failure isn't picking the wrong thing from what's visible; it's simply never managing to recall what to do in the first place.
Where it stops holding
This recall burden isn't equally heavy for every user. A user encountering a given command or gesture for the first time, or using it only rarely, carries the heaviest burden and the highest error rate; once an action is repeated frequently enough, memory for it stabilizes and can become nearly automatic, dropping the recall difficulty considerably. But that's a change of state that comes with practice — it can't be turned around into a reason that command lines and gestures don't need visible help in the first place. A product will always have new and infrequent users, and they still face the full recall burden of the initial learning stage.
Applying it
- For command-line interactions, provide candidate suggestions as the user types (matching full commands from just the first few characters), a permanently accessible list of available commands, and naming conventions built around verb-object structures that align with intuitive guessing, so that even partial memory combined with a visible prompt lets the user piece together the right command.
- For gesture interactions, provide fallback paths that don't rely on memory: a gesture hint animation on first use, a long-press or summon-able menu that lists currently available gestures and what they do, and at least one visible button or menu entry that accomplishes the same action — never let a feature be reachable only through a gesture with no discoverable alternative.
- When a command fails or a gesture isn't recognized, feedback should proactively suggest "did you mean…" corrections, turning a failed recall into a recognition opportunity rather than simply telling the user "invalid."
- How to check: test success rate and completion time, with and without the visible hint available, separately for users encountering the command or gesture for the first time and for users who have already used it many times. If first-time users' success rate improves sharply with the hint while experienced users show little difference between the two conditions, that confirms the visible entry point is mainly a safety net for new and infrequent users — and is the basis for deciding whether to keep it.