Knowledge dashboard
How to run a cross-cutting campaign
The essay's knowledge, dashboard-shaped — graph, eyes, actions. The graph is the source of truth; the prose follows.
The essay's claims, typed and connected. Synthesis is the author's contribution. Citation grounds in a named source. Derivation follows from prior nodes. Click any node to jump to its card.
S S01-cross-cutting-finish-or-not · opening
Cross-cutting changes finish or they don't.
S S02-context-in-hand · the principle
Every human in the loop has the context they need to act on the change in front of them.
D D03-cascade-loop
Skipping the constraint produces Campaign → Stall → Escalate → Directive → New Campaign.
derives_from: S02-context-in-hand
S S04-three-paths
Three valid responses — automate, centralize, enable.
derives_from: S02-context-in-hand
D D05-dead-zone
The pattern fails in the middle — work done for the team, stop short of merge. Human-in-loop without context.
derives_from: S02-context-in-hand , S04-three-paths
C C01-codemod
Codemod — a program that rewrites source code mechanically.
grounds: facebookarchive/codemod
Claims that need more grounding. The essay's confident synthesis — but where external corroboration would strengthen the case, or an open question is implicit.
The cascade loop (Campaign → Stall → Escalate → Directive → New) is the author's observation. No external citation naming this exact failure mode.
Why eyes: The pattern is plausible from first-hand engineering experience, but external corroboration would strengthen the claim. Adjacent literature exists.
ground_with:
The three-paths taxonomy (automate / centralize / enable) is original framing. No cited prior taxonomy.
Why eyes: Useful as the author's synthesis. If a published org-design taxonomy already names a similar trichotomy, citing it would let the essay sit alongside that frame instead of re-naming it.
"Dead zone" is the author's naming. No empirical data on how often this specific failure mode occurs in real orgs.
Why eyes: The naming is sharp; the prevalence claim is implicit. An open question: what fraction of stalled cross-cutting initiatives sit in the dead zone vs. fail elsewhere?
What the essay implies for the reader. Etudes drill specific claims; apply items put the principle to work; next points at sibling essays.
etude Cascade simulator →
Press launch. Watch four campaigns spawn. The original migration creeps. Demonstrates D03.
etude Review this PR →
30 seconds in the seat where the cascade lands. Demonstrates D05.
etude Three paths →
Pick a path for four scenarios. Drills S04.
apply Audit your initiatives
For each cross-cutting initiative your org is running, name which of the three paths it's on. Anything you can't cleanly assign is in the dead zone.
apply Find your dead zone
Which migrations have you done the work for (PRs created) but not finished (other teams holding merge)? That's the dead zone. Pull the human out of the place they cannot add value.
next Same shape, different domain →
The know-thyself series argues the same principle — substrate must match the reader — at a different scale (personal memory in LLM conversations).