feat(sextant): Freedom & control matrix + UPDATE cost bucket

Adds the edit/remaster/pull/permanence axis the cost model was missing:

- 4th cost bucket UPDATE: re-mastering an already-live release. Free on
  RouteNote; a new paid release on CD Baby; forces DistroKid to the Ultimate
  tier ($89.99/yr) for an in-place Audio Swap. Driven by a Remasters/year slider.
- Freedom & control matrix: per-option remaster / edit-metadata / takedown /
  if-you-stop-paying, glyph+text+colour (never hue alone). Surfaces the key fact
  that no DSP distributor swaps audio in place (a store rule), with Bandcamp the
  free-replace outlier.
- Ownership reframed: all options are non-exclusive (you keep 100% of masters);
  what differs is permanence, edit-freedom, portability.
- Softened the unverified RouteNote freeze-risk claim; flagged Premium pricing
  as contested pending verification.

armada/tasks/012-le-sextant-the-compass-gets-an-engine.md

@@ -18,11 +18,13 @@ the running as a real option (not pre-rejected on the no-sub rule).
 Rebuilt as a slider-driven economic model in `armada/ui` (Vite multi-page: the
 take-judge stays at `/`, the simulator at `/sextant.html`), Ship's Bridge tokens
 reused as-is. A pure-function model (`model.ts`) expands catalog inputs into dated
-releases, then computes PORT/KEEP/ADD cost + a decaying-spike revenue curve per
-option. Four distributors — RouteNote Free (baseline), RouteNote Premium, CD Baby
+releases, then computes PORT/KEEP/ADD/**UPDATE** cost + a decaying-spike revenue curve
+per option. Four distributors — RouteNote Free (baseline), RouteNote Premium, CD Baby
 (recommended), DistroKid — plus Bandcamp as an independent layer added on top of any
 of them. Prefilled with the real catalog (11 singles + 1 album on DSP, 41 SoundCloud
-backlog, ~6/yr forward). Hand-rolled SVG break-even chart — no chart dependency.
+backlog, ~6/yr forward). Hand-rolled SVG break-even chart — no chart dependency. A
+**Freedom & control** matrix answers the non-money question: can you remaster, edit,
+pull, and stay up — and at what cost?
 
 ## Prise (findings / artifacts)
 - `armada/ui/src/sextant/{model,data,format,Control,OptionRow,BreakEvenChart,Sextant}.tsx`
@@ -33,6 +35,15 @@ backlog, ~6/yr forward). Hand-rolled SVG break-even chart — no chart dependenc
   near-identical net because the catalog stays yours. **Bandcamp digital sales are a
   rounding error** (~$132 over 10 yr) — its worth is ownership, not revenue.
 
+- **The remaster question reframes everything.** No DSP distributor swaps audio in
+  place — that's a *store* rule, not the distributor's. A remaster = a brand-new
+  release (new ISRC, lost date & plays) on CD Baby / RouteNote / TuneCore; DistroKid's
+  in-place Audio Swap needs the **Ultimate tier ($89.99/yr)** and only for minor
+  tweaks; **Bandcamp replaces the file free, anytime**. For a remaster-heavy salvage
+  op, that flips Bandcamp from "nice direct channel" to "master-of-record."
+- Adding 2 remasters/yr pushes DistroKid to Ultimate and narrows its net lead over CD
+  Baby from ~$1.3k to ~$640 — the UPDATE bucket made a hidden cost visible.
+
 ## Sel (the shareable learning)
 - The break-even cadence everyone quotes (CD Baby vs DistroKid) **collapses to zero
   once you have a back-catalog to port** — owning-vs-renting stops being a price

armada/ui/src/sextant/FreedomMatrix.tsx

+import { Check, Minus, X } from 'lucide-react'
+import { OPTIONS, BANDCAMP_FREEDOM, type Freedom, type FreedomFact, type FreedomLevel } from './model'
+
+// Level → glyph + color, encoded redundantly (glyph + text + color, never hue alone).
+const LEVEL: Record<FreedomLevel, { Icon: typeof Check; cls: string; sr: string }> = {
+  good: { Icon: Check, cls: 'text-ready', sr: 'yes' },
+  mixed: { Icon: Minus, cls: 'text-status-wip', sr: 'partial' },
+  bad: { Icon: X, cls: 'text-blocked', sr: 'no' },
+}
+
+const COLS: { key: keyof Freedom; label: string }[] = [
+  { key: 'remaster', label: 'Remaster (swap audio)' },
+  { key: 'edit', label: 'Edit metadata' },
+  { key: 'pull', label: 'Pull / takedown' },
+  { key: 'unpaid', label: 'If you stop paying' },
+]
+
+function Cell({ fact }: { fact: FreedomFact }) {
+  const { Icon, cls, sr } = LEVEL[fact.level]
+  return (
+    <td className="border-t border-hairline px-3 py-2.5 align-top">
+      <div className="flex gap-2">
+        <Icon size={14} className={`mt-px shrink-0 ${cls}`} aria-hidden />
+        <span className="sr-only">{sr}: </span>
+        <span className="text-[12px] leading-snug text-ink-muted">{fact.text}</span>
+      </div>
+    </td>
+  )
+}
+
+/** Rows = the four distributors + the Bandcamp layer; columns = the freedom axes
+ *  that actually differ (rights ownership doesn't — see the note below). */
+export function FreedomMatrix() {
+  const rows = [
+    ...OPTIONS.map((o) => ({ id: o.id, name: o.name, recommended: o.recommended, freedom: o.freedom })),
+    { id: 'bandcamp', name: 'Bandcamp (layer)', recommended: false, freedom: BANDCAMP_FREEDOM },
+  ]
+  return (
+    <section className="rounded-lg border border-hairline bg-raised p-4">
+      <div className="mb-3 flex flex-wrap items-baseline justify-between gap-2">
+        <h2 className="text-sm font-semibold text-ink">Freedom &amp; control</h2>
+        <p className="max-w-prose text-xs text-ink-muted">
+          Can you fix, change, or walk away, and at what cost? You keep 100% of your masters with all of
+          them — these are the axes that actually differ.
+        </p>
+      </div>
+
+      <div className="rounded-md border border-hairline bg-surface p-3 text-xs leading-relaxed text-ink-muted">
+        <span className="font-medium text-ink">No DSP distributor swaps audio in place</span> — that&apos;s a
+        store rule, not the distributor&apos;s. Everywhere except DistroKid Ultimate, a remaster is a brand-new
+        release (new ISRC, lost release date &amp; plays). The exception worth knowing:{' '}
+        <span className="text-melodic">Bandcamp lets you replace the file freely, anytime</span> — reason
+        enough to keep it as your master-of-record.
+      </div>
+
+      <div className="mt-3 -mx-1 overflow-x-auto">
+        <table className="w-full min-w-[640px] border-collapse text-left">
+          <thead>
+            <tr>
+              <th className="px-3 pb-2 text-[11px] font-medium uppercase tracking-wide text-ink-faint">
+                Option
+              </th>
+              {COLS.map((c) => (
+                <th
+                  key={c.key}
+                  className="px-3 pb-2 text-[11px] font-medium uppercase tracking-wide text-ink-faint"
+                >
+                  {c.label}
+                </th>
+              ))}
+            </tr>
+          </thead>
+          <tbody>
+            {rows.map((r) => (
+              <tr key={r.id} className={r.recommended ? 'bg-magenta/[0.04]' : undefined}>
+                <th
+                  scope="row"
+                  className="border-t border-hairline px-3 py-2.5 align-top text-[12px] font-medium text-ink"
+                >
+                  {r.name}
+                </th>
+                {COLS.map((c) => (
+                  <Cell key={c.key} fact={r.freedom[c.key]} />
+                ))}
+              </tr>
+            ))}
+          </tbody>
+        </table>
+      </div>
+
+      <p className="mt-3 text-[11px] leading-snug text-ink-faint">
+        <span className="text-ink-muted">Ownership ≠ permanence.</span> All of these are non-exclusive: your
+        masters and copyrights stay 100% yours. What varies is <span className="text-ink-muted">permanence</span>{' '}
+        (does it stay up unpaid), <span className="text-ink-muted">edit-freedom</span> (can you fix it), and{' '}
+        <span className="text-ink-muted">portability</span> (you keep your ISRCs and can re-deliver elsewhere).
+        Fine print: CD Baby Pro (legacy) and TuneCore publishing-admin take a cut of <em>publishing</em>{' '}
+        royalties — opt-in, separate from distribution, never your masters.
+      </p>
+    </section>
+  )
+}

armada/ui/src/sextant/OptionRow.tsx

@@ -15,6 +15,7 @@ const SEGMENTS = [
   { key: 'port' as const, label: 'Port', color: 'var(--color-melodic)' },
   { key: 'keep' as const, label: 'Keep', color: 'var(--color-percs)' },
   { key: 'add' as const, label: 'Add', color: 'var(--color-tops)' },
+  { key: 'update' as const, label: 'Update', color: 'var(--color-bass)' },
 ]
 
 export function OptionRow({ r, isWinner, maxCost, maxProfit }: Props) {

armada/ui/src/sextant/Sextant.tsx

@@ -10,10 +10,11 @@ import {
 import { PARVAGUES_CATALOG, MODEST_REVENUE } from './data'
 import { Slider, Stepper, Group, Toggle } from './Control'
 import { OptionRow } from './OptionRow'
+import { FreedomMatrix } from './FreedomMatrix'
 import { BreakEvenChart } from './BreakEvenChart'
 import { money, compact } from './format'
 
-const STORE = 'sextant_inputs_v3'
+const STORE = 'sextant_inputs_v4'
 
 interface State {
   cat: CatalogInputs
@@ -160,6 +161,14 @@ export default function Sextant() {
               format={(v) => `${Math.round(v * 100)}% albums`}
               onChange={(v) => setCat({ cadenceAlbumShare: v })}
             />
+            <Slider
+              label="Remasters / year"
+              value={cat.remastersPerYear}
+              min={0}
+              max={24}
+              hint="Re-mastering an already-live release. Free on RouteNote/Bandcamp; a new paid release on CD Baby; needs DistroKid Ultimate for an in-place swap."
+              onChange={(v) => setCat({ remastersPerYear: v })}
+            />
             <Slider label="Horizon" value={cat.horizonYears} min={1} max={25} unit=" yr" onChange={(v) => setCat({ horizonYears: v })} />
           </Group>
 
@@ -211,6 +220,8 @@ export default function Sextant() {
             ))}
           </div>
 
+          <FreedomMatrix />
+
           <section className="rounded-lg border border-hairline bg-raised p-4">
             <div className="mb-3 flex flex-wrap items-baseline justify-between gap-2">
               <h2 className="text-sm font-semibold text-ink">The break-even, drawn</h2>

armada/ui/src/sextant/data.ts

@@ -12,6 +12,7 @@ export const PARVAGUES_CATALOG: CatalogInputs = {
   backlogTracksPerEp: 1, // all singles by default; bundle into EPs with the slider
   cadencePerYear: 6,
   cadenceAlbumShare: 0.2,
+  remastersPerYear: 2, // the salvage op is remaster-heavy — re-delivering existing takes
   horizonYears: 10,
 }