Scenarios
A control point inside a control point
Editor rewrite in stages: new renderer first, collaboration mode built on top of it. Outcome: 2 ships · 1 graph. 5 of 5 held · dependency never violated
one control point a woven topology
“a one-line bugfix” ●shipped · 41 min“a CLI release to every laptop” ↶re-pinned in 51 s · then shipped“a mobile feature stuck behind store review” ●shipped · one binary“a rewrite that claims it’s 10× faster” ●proven · 2.1M comparisons“a risky change to a feature most orgs ignore” ●verdict · 5 h“a new skill for your agent fleet” ●fleet-wide · 0 broken sessions“a feature that spans API and UI” ●shipped · 3 days“one contract across three services” ●zero drift · one 40-min hold“a feature built on another feature” ●2 ships · 1 graph“the change that looked safe” ↶reversed · 43 s
renderer-collab-stacked
TWO SHIPS, ONE GRAPHA control point inside a control point
Editor rewrite in stages: new renderer first, collaboration mode built on top of it.
The plan — drafted before anyone saw it
Strategy Stacked control points — the child cannot outrun its parent
Control points renderer (parent) · collab mode (child, nested inside)
Gates parent stable at each step before the child opens
Guardrails render errors · doc-sync conflicts
Rollback parent reverts → child goes dark with it, automatically
Ramp — planned outline, actual fill
25%
50%
child
100%
The run — from the decision log
T+0m read: collab mode compiles only against the new renderer — hard dependencyheld
T+2h chose: stacked control points — collab is a child, cannot outrun its parentheld
T+2d ramped: renderer to 50%; collab held dark inside itheld
T+4d opened: collab to renderer users only — the graph enforced the orderheld
T+6d done: two features shipped on one dependency graphheld
5 of 5 held · dependency never violated
learnedstacked topology stored as the template for renderer-dependent features.