Garment Intake Deprecated
Photograph a garment, and it is ingested, classified and queued for a one-keystroke confirmation. Fully specified, not yet built.
Deprecated. Retired in a fleet code-standard review because it held no runnable product: a complete spec with no code, whose first step, a throwaway prototype timing the photo step, was never built.It was fully planned on paper but never built, so it was retired.
Feasibility: 4/5
Scored 1–5 on what stands between the idea and a working version: whether the data is reachable through a supported channel, whether anything gates it (approval, cost, another person acting), whether it keeps working unattended, and how much is left to build once the data is in hand.Scored 1 to 5 on what is actually in the way: whether the information can be got hold of properly, whether anyone has to approve it or charge for it, whether it will keep working on its own, and how much work is left once the data is available.
A 4, and unusually for a backlog project the uncertainty is not technical. Twenty design documents and nine ADRs already settle the capture path, the domain model, the schema and the taxonomy. What is unresolved is the one assumption the whole design rests on: that a Back Tap shortcut really captures a garment in under four interactions. The plan opens with a two-day spike to find out.
Four out of five. The design is finished in unusual detail. The one real unknown is whether the phone gesture that starts the whole thing is actually as quick as assumed, which a two-day test would settle.
- Helps Design already done Twenty documents and nine ADRs, down to the transition table and the HEIC decoder chain. The usual backlog risk (nobody has thought it through) does not apply.The thinking is already done in detail, which is normally the risky part of an unstarted project.
- Helps No exotic dependency Postgres, TypeScript, an Apple Shortcut and a cloud vision call. Nothing here needs an API that might not exist.It needs only ordinary, widely available pieces, nothing that might turn out to be unavailable.
- Holds it back The capture gesture is unproven The eight-second target assumes Take Photo with count=2 does not demand a Use Photo tap per shot. If it does, the capture design changes.The speed target rests on the phone behaving a particular way, which nobody has actually timed yet.
- Holds it back Six weeks is an estimate, not a commitment The roadmap is six weeks of full-time work for one experienced developer, and estimates that size are relative rather than committed.The plan is about six weeks of full-time work, and plans that size tend to stretch.
What would raise the score
- Run Phase 0, the two-day capture spike → 5/5 A throwaway Shortcut and a twenty-line server, timed over twenty garments. It is the only phase whose outcome could be 'change the plan', which is exactly why it goes first.Build a rough version of just the photo step and time it on twenty garments. It is the one test that could change the whole approach.