# Charter — what constitutes great work

**Status:** v3, 2026-08-11 (v2 2026-07-09; v3 adds the tier list). Used for categorizing and prioritizing. Never for keeping work out of the wiki.

## The metric

**Impact = displacement in joy**, estimated by:

1. **Depth × breadth.** Immense joy for a few and small joy for many are both high impact.
2. **Permanence dominates.** A permanent displacement outweighs any temporary one.
3. **Filter by maximum estimated bounds, not sign.** Consequential work often has enormous bounds in *both* directions (a spacefaring species: many will die, Earth may suffer — and survival-in-space is near-infinite positive displacement; or space kills us all). The reader's question is "how much could this move the world?", not "is it good or bad?" — rank by the magnitude of the bounds and describe the directions plainly.
4. **Counterfactual, not credit.** Compare the world with and without the entity; work that would have happened anyway gets the difference, not the whole.
5. Joy, not pain-removal — pain's relationship with joy is complicated; we don't optimize for its absence.

Priors that shift estimates: long-focused people have outsized outcomes; businesses whose stakeholders are aligned with a long-term mission (and whose people deeply care) displace more durably — extractive ownership structures are evidence against, overridable by specifics.

## Stance

Judge loosely; the world needs believers. But always name the bet: what has to become true, and where evidence is thin.

## The tier list

The metric above is a ranking, so every fact page carries one: **`**Tier:**`** — `S`, `A`, `B`, `C`,
`D`, `F`, or `unranked`, optionally suffixed `*`. One ladder, every `Type` on it, absolute against
the world rather than graded against Utah.

| | |
|---|---|
| **S** | world-historic — permanently different world, and the difference traces to this |
| **A** | field-defining — sets a global frontier, or is a live bet whose plausible success is S |
| **B** | structurally significant — durably changes how a sector or a state works |
| **C** | substantive and replaceable — real scale, close substitute exists |
| **D** | local — real to the people in the room, stops at the walls |
| **F** | inert — a listing, a ceremony, a duplicate, or a claim nothing corroborates |
| **unranked** | too thin to argue bounds from; the honest escape hatch, not a synonym for low |

Three calls decide most cases. Tier is the **ceiling**, not the current quarter. **Magnitude, never
sign** — weapons, surveillance, and extraction rank on how far they could move the world in either
direction, and the direction goes in the prose. Enablers get only their **counterfactual difference**,
which puts most of the `resource` shelf at `D` or `F`.

**The rubric, the nine rulings, and the worked cases are in [`tiers.md`](tiers.md) — read it before
assigning a tier.** Gems are `S` and `A`; that is how "surface all the gems" became a filter instead
of a matter of taste.

## Hooks into the machinery

- `## Impact` sections argue depth, breadth, permanence, bounds, counterfactual — prose with
  reasoning, no scores. A `**Tier:**` of `B` or above requires one, and where prose and tier disagree
  the prose is right.
- `**Bet:**` names the contrarian premise on pages that have one.
- `**Ownership:**` attribute records the checkable fact; judgment pages interpret it.
- Wanted-list gaps are ordered by expected impact.

Refined like everything else: probe with hard cases, adjudicate divergence into this doc — or into
`tiers.md` when the disagreement is about placement rather than about the metric (E11).