Week 2 Session: Translating the Push
What a physical button gives you for free — resistance, travel, a click, a physical return to its resting position — and what a digital button has to fake or simply do without.
1 March 2027
Press a doorbell and your finger feels it push back. There’s resistance before the contact closes, a short distance of travel, often an audible click at the moment of contact, and when you let go, a spring returns the button to where it started. None of that is decoration. It’s how you know, without looking, that the press registered.
A rectangle on a screen has none of it. There’s no resistance, because there’s nothing to push against — your finger is on glass, not a switch. There’s no travel, because the rectangle doesn’t move. There’s no click, unless the software fakes one. There’s no physical return, because there was no physical departure. Every one of those cues has to be simulated in some other channel — a colour change, a sound file, a vibration — or the interface does without it and hopes the user trusts the tap anyway.
Translating one button from Week 1#
Take one button from your Week 1 log — one you know well enough to describe from memory. Build the smallest possible digital version of it: one clickable element, one visible response. It doesn’t need a backend. It needs to stand in convincingly for the physical button you chose.
While you build it, keep a running list of every physical cue the original button gave you that your digital version had to either simulate or drop:
- Resistance — did you fake it with a colour or shadow change on press, or skip it?
- Travel — does your element move at all, or is it flat the whole time?
- Click — is there a sound, and if so, is it doing the same job the mechanical click did?
- Physical return — how does a user know the interaction is over, if nothing sprang back?
- Tactile confirmation — what replaces the feeling of the press succeeding, for someone who can’t see the screen change?
The list#
Submit your digital button alongside the cue list. For each cue you dropped rather than simulated, say what you think was lost by dropping it — not in general, for your specific button. Dropping tactile confirmation is a different kind of loss on a panic button than it is on a doorbell.
The spec
What you bring to the session. Written so a reader can tell whether it has been met without asking you.
- you have picked one button from your week 1 log and built a minimal digital version of it
- you can name, for that specific button, every physical cue the digital version had to fake or drop — resistance, travel, click, physical return, tactile confirmation
- the digital version gives some signal, however small, that the press registered
Teaching team
- Idris Fenn — tutor