Give the link's liveness one owner

todo.md

@@ -200,16 +200,11 @@ next work ([decision](docs/decisions.md#internals-and-tests)).
 ### Locality — next, ahead of everything below; 5–6 today, and the gate is 7
 
 A second four-seat run on 2026-09-28, after #35–#40, read 6, 6, 7 and 6 again, Locality 5, 5, 6
-and 5. Every seat ranked the session's lifecycle hardest and least wanted to modify it.
+and 5. Every seat ranked the session's lifecycle hardest and least wanted to modify it. A third,
+after #46, read 6, 6, 7 and 7, Locality 5, 5, 6 and 6; the link's liveness now has one owner.
 
 - [ ] **Lift Locality to 7, and confirm it with a scoring run.** A run reading 7.0 or above also
-      retires the #30 and #46 decision. The sub-items are what the 2026-09-28 run named, most seats first.
-- [ ] **Give the link's liveness one owner.** A third run the same day, after #46, read 6, 6, 7 and
-      7, Locality 5, 5, 6 and 6; all four seats ranked `drain()`/`end()`/`teardown()`/`retrying()`
-      in `session.ts` hardest, because whether the link lives is kept in `Session.lifecycle`,
-      `ReconnectLoop.halted`, `LinkGate.up`/`returning`, `OutgoingRequests.draining` and
-      `IncomingRequests.linkGeneration`, held in step by statement order and the comment above
-      `retrying()`. Four seats.
+      retires the #30 and #46 decision.
 
 ### Correctness
 
@@ -369,9 +364,9 @@ and 5. Every seat ranked the session's lifecycle hardest and least wanted to mod
       cross-check in MIGRATION.md; the planned work in `interop-tests/AGENTS.md` (an expected
       malformed count per peer) and `benchmarks/README.md`.
 
-- [ ] **Make `LinkGate.isUp()`'s doc true or its state match it.** It says a link attached but not
-      yet bound cannot carry a request, while `up` starts `true`, so the first link and a server
-      session are up before any bind. The gate decision in `docs/decisions.md` makes the same claim
+- [ ] **Make `LinkLife` start unbound, or its decision's title true.** A link attached but not
+      yet bound cannot carry a request, while `phase` starts `up`, so the first link and a server
+      session are up before any bind. The `LinkLife` decision in `docs/decisions.md` makes the same claim
       in its title, and carries the same fix. From the comprehension panel of #25.
 
 - [ ] **Move `checkSessionOptions()`'s doc comment to what it describes.** It explains why a count
@@ -485,9 +480,9 @@ and 5. Every seat ranked the session's lifecycle hardest and least wanted to mod
       support Gitea, so mirror each Gitea pull request to GitHub for it to review there.
       Maintainer's ask, 2026-09-14; not started until asked.
 
-- [ ] **Count what is left of a budget one way in `leftOf()` and the link gate.** Today they are one
+- [ ] **Count what is left of a budget one way in `leftOf()` and `LinkLife`.** Today they are one
       concept counted twice. `idle-waiters.ts` reads what is left of a budget as `Math.max(1,
-      deadline - now)`, because 0 means "forever" there; `link-gate.ts` runs the same subtraction
+      deadline - now)`, because 0 means "forever" there; `link-life.ts` runs the same subtraction
       and calls `<= 0` expired. Neither is reachable from the other, so nothing can disagree today,
       but a reader who learns one and applies it to the other is wrong. A budget type both take
       would close it. Raised by review, 2026-09-01.