# Decision log: Swap & Handoff

Each entry records who decided, when, why, the evidence, and what was **not** chosen. It also records the conditions under which the decision should be reopened. Nothing is deleted.

---

## DL-1 · Week 4 · Restructure v1 "Meet now" into v2 "Swap Spots"
- **Decided by:** Ana (product). Consulted: Ravi (trust & safety) and Jo (platform)
- **Proposal (v1):** members see each other's live location after acceptance and meet wherever they are (`prd-v1.md` UC-4, `sim-v1.html` scenario `meetnow`).
- **Analysis:** `03-analysis-v1.md`: F2 (no real-time channel, 9–12 weeks), F3 (policy P-7), I1 (conflict with the Quiet profile PRD), I4 (home location inferable after about 3 swaps).
- **Decision:** replace live location with fixed public Swap Spots, 1-hour time windows and a boolean check-in. Keep the handoff code from v1.
- **Rejected alternatives:**
  - *Live location with coarse blur (±300 m).* Still breaks P-7, and the blur still triangulates over repeated swaps.
  - *Live location only inside a 200 m radius of a venue.* Still needs the streaming channel, so the cost stays the same.
  - *No meeting help at all (chat only).* Doesn't move the 61% failure rate.
- **Assumptions behind the rejection:** streaming infrastructure doesn't exist; P-7 has no consent exception; members' homes are near where they swap.
- **Revisit v1 when ALL of these become true:**
  1. The platform has a real-time channel for other reasons (the Lantern Clubs "live reading room" roadmap item, if built).
  2. Legal approves a time-boxed, consented P-7 exception with per-region consent flows.
  3. A proximity method exists that doesn't reveal positions (e.g. "you are both within 100 m: yes/no" computed without exchanging coordinates).
  4. Swap Spots don't reach a 70% completion rate (if they do, v1 isn't needed).
- **Where v1 lives:** `prd-v1.md`, `sim-v1.html` (frozen) and the idea archive tag `live-proximity`.

## DL-2 · Week 5 · Add swap credits (UC-2b)
- **Decided by:** Ana. **Source:** competitive analysis (PageTrade).
- **Why:** removes the "nothing you want on my shelf" deadlock (Q2).
- **Not chosen:** paid credits (out of scope; changes the community's tone).
- **Revisit when:** credits pile up (more than 20% of members with 5+ unspent), which would suggest an exchange imbalance.

## DL-3 · Week 6 · No public ratings
- **Decided by:** Ravi + Ana. **Why:** retaliation risk; private reports instead (UC-9).
- **Revisit when:** report volume is too low to identify repeat bad actors.

## DL-4 · Week 12 · Fixed 30-minute slots (reconciliation)
- **Decided by:** Jo (engineering), accepted by Ana after release validation.
- **Why:** venues asked for predictable shelf traffic; fixed slots make the `spots.slots` capacity simple.
- **Effect:** the 1-hour windows from PRD 2.0 UC-4 are replaced by 30-minute slots. PRD 2.1 is updated, and the window picker in SIM v2 carries a "Reconciled" note. The 1-hour window design is kept as history.