Week XI Lab: First-Use Constraint Audit
Share
Workshop 11 already moved everyday objects into abstract form. The lab should do the opposite: keep the objects ordinary and measure how well their knowledge-in-the-world actually works.
Question. How much of “knowing what to do” lives in the object versus in the user’s head, and which constraint type predicts first-use success?
Setup. Pick 4–6 objects you can photograph or use immediately. Mix scales:
- one household control (stove, thermostat, lamp switch cluster)
- one packaged product (bottle, pump, jar lid)
- one digital screen (settings panel, checkout, media player)
- one “designed” object from this week’s workshop photos if you already abstracted it
For each object, collect this row of data:
Constraint & Mapping Audit
Week 11 lab sheet — record one row of observations per everyday object.
| Field | What to record |
|---|---|
| Object | name + photo |
| Intended action | one verb (“open,” “heat left-rear burner,” “mute”) |
| Knowledge in the world | labels, shape, layout, color, sound — listed, not paraphrased |
| Knowledge required in the head | anything you had to already know |
| Constraint type(s) | physical / cultural / semantic / logical (Norman Ch. 4) |
| Mapping quality | 1 = arbitrary, 3 = spatial/natural, 5 = control is the thing |
| Discoverability | could a first-time user find the action without instruction? Y/N + 1-sentence why |
| Feedback | what happens after a correct vs wrong action (visual / auditory / none) |
| First-use trial | time to correct action (seconds) and error count (wrong presses, wrong orientation, hesitation >3s) |
| Failure mode | if errors: slip, wrong model, missing signifier, or missing constraint |
Run each first-use trial once as if you had never seen it, then once after 30 seconds of study. That gives you the Ch. 3 tradeoff in two columns: world-heavy vs head-heavy performance.
Analysis (the scientific part).
1. Sort objects by error count, then by mapping score.
2. Check whether physical + logical constraints predict lower first-use errors than cultural + semantic ones.
3. Note where high knowledge-in-the-world slowed the second trial (Norman’s skilled-use tradeoff).
4. Write one causal sentence per object: “Error X happened because constraint Y was absent / mapping Z was arbitrary.”
That last sentence is the lab’s product. It is design diagnosis, not vibe.
Optional Grove overlay (if you want evolution in the same frame)
Take one observatory organism from Tuesday — Acadian Flycatcher, Northern Hog Sucker, Hepatica — and treat the environment as the designed object.
- Physical constraint: perch geometry, riffle stones, forest-floor light.
- Semantic constraint: what the body “means” it can do (sally, root, open with light).
- Mapping: trait → action (concave head → stone-turning; oil-droplet retina → high-speed prey tracking).
Score the same table. The point is that evolution already put knowledge in the world; good human design is trying to do the same thing on purpose.
What not to do today
Don’t turn this into another abstract collage or sound sketch. Wednesday already did form. Today’s job is a scoring sheet, a few photos, and a short results paragraph that a skeptic could check.
