Teardown & Redesign Dossier

A full teardown of one captive machine, a redesign that keeps its legitimate incentives, and a memo naming who would resist the change

Due: 28 May 2027 · Weight: 50%

The brief

Take apart one machine this course has given you the vocabulary for, then put it back together fairer — without pretending it has no legitimate constraints at all.

Choose one machine from the semester’s twelve, or a close relative of one. The teardown documents how it currently behaves and who that behaviour serves. The redesign then has to do something harder than remove friction: it has to say which piece of friction was actually legitimate, and keep it. The closing memo names whoever loses out from your redesign — an operator, a vendor, a franchise holder — and says plainly why they’d push back.

What you submit

A single dossier with three clearly labelled sections:

  1. The teardown — how the machine currently behaves and who that behaviour serves, evidenced the way Assignment 2 asked you to evidence a claim: not just asserted.
  2. The redesign — sketches, flows, or a working prototype, your choice of fidelity. Whatever the fidelity, it has to be enough to show the specific piece of friction being removed and the specific piece being deliberately kept, and to name which legitimate constraint (fraud prevention, throughput, liability) the kept piece serves.
  3. The resistance memo (300–500 words) — names who loses out from the redesign and says plainly why they would push back.

Assume your reader has this course’s vocabulary but no memory of your machine: name it again here even if you already used it in Assignment 1 or 2. Say plainly what a reader needs to evaluate the dossier without you in the room.

Who benefits when the design is bad?

The spec

The brief poses the problem and leaves room for a response. The spec is the fixed part: what a marker checks. Some lines can be checked mechanically; the rest call for judgement.

  • the teardown is evidenced, not just asserted
  • the redesign names the legitimate incentive it deliberately keeps
  • the memo names who would resist the change, and why
  • the dossier names its machine and transaction on its own, without relying on a verbal walkthrough

Who benefits when the design is bad?

How it is marked

Judged as a whole against three things the brief asks for together — an evidenced teardown of the machine as it exists, a redesign that removes illegitimate friction while keeping whatever legitimate constraint (fraud prevention, throughput, liability) made the original design defensible, and a short memo naming who would resist the redesign and why.