Skip to content

Correct the README claims the rebuild made false - #69

Merged
AndrewCTF merged 1 commit into
masterfrom
readme-text-corrections
Aug 8, 2026
Merged

Correct the README claims the rebuild made false#69
AndrewCTF merged 1 commit into
masterfrom
readme-text-corrections

Conversation

@AndrewCTF

Copy link
Copy Markdown
Owner

Follow-up to #68. The figures were re-shot there; the prose around them was
still describing the shell those figures used to show. Everything below was
checked against source or against a running keyless backend, not against
memory.

Wrong about the console

  • §11 "tear the workspace apart" claimed any right-rail panel detaches into
    a floating window. It does not — that control belongs to ConsoleShell,
    which only the /2d route renders. What the 3D console actually floats is
    the three map-tool panels (annotation editor, watchbox rules, field kit),
    through the same substrate, opened from the toolbar and cleared by
    Window → Re-dock every floating panel. The paragraph says that now, and the
    figure shows two of them over the live globe — this closes the one figure
    Re-shoot the README against the console that actually ships #68 could not honestly re-shoot.
  • The replay caption described a coverage chip ("recording since … · GB ·
    fixes") and a day-picker. Both belong to Timeline, also 2D-only. The 3D
    console's dock has window presets (1h–7d), a density strip you click to seek,
    1x–3600x playback and a Replay/Live header, so that is what the caption and
    paragraph describe, with a pointer to /2d for the dated picker. The
    retention bullet made the same claim and now credits /api/history/coverage,
    which is what actually reports real depth.
  • The app table listed twelve apps and omitted the AI and Markets groups.
    state/appView.ts has fourteen in six groups.
  • The selection figure's alt text described a Boeing 787-9 on the JFK–Rome
    run, which was the airframe in the 2026-07 shot.

Wrong about scale

Measured today against a keyless backend:

Claim Was Is
MCP tools 50 84 (--list-tools)
Aircraft 9–13k (~11k typical) 14,436
Vessels ~33k 50,024 tracked
Satellites ~16k 16,095
Earthquakes ~250/day ~290/day
GPS jamming ~200 cells ~220 cells
Military aircraft ~140 29–62

The aircraft figure predates the fourth ADS-B tier, the FR24 tier and the
OpenSky gap filler. The military count went down; it is lowered rather than
explained away, because the tier that fed the old number is not what feeds it
now. The news row notes the feed answers enabled: false on a keyless boot.

Provenance

The tour intro now dates the screenshots and names the script that re-shoots
them, so a reader can tell how old the figures are without checking git.

bash scripts/verify.sh green: 2255 passed + 2 skipped.

The figures were re-shot in the previous commit; the prose around them was
still describing the shell those figures used to show. Everything below was
checked against source or against a running backend, not against memory.

Wrong about the console:

- §11 said any right-rail panel detaches into a floating window. It does not:
  that control belongs to ConsoleShell, which only the /2d route renders. What
  the 3D console actually floats is the three map-tool panels — the annotation
  editor, the watchbox rules and the field kit — through the same substrate,
  opened from the toolbar and re-docked from the Window menu. The paragraph now
  says that, and the figure shows two of them over the live globe.
- The replay figure's caption described a coverage chip ("recording since … ·
  GB · fixes") and a day-picker. Those belong to Timeline, which is also 2D
  only. The 3D console's dock has window presets (1h to 7d), a density strip
  you click to seek, 1x-3600x playback and a Replay/Live header, so that is
  what the caption and the paragraph describe, with a pointer to /2d for the
  dated picker. The retention bullet made the same claim and now credits
  /api/history/coverage, which is the thing that actually reports real depth.
- The app switcher lists fourteen apps in six groups; the README said twelve
  and its table omitted the AI and Markets groups entirely.
- The selection figure's alt text described a Boeing 787-9 on the JFK-Rome run,
  which was the airframe in the 2026-07 shot.

Wrong about scale, measured on a keyless backend today:

- MCP tools: 50 in the text, 84 by `--list-tools`.
- Aircraft: "9-13k (~11k typical)" predates the fourth ADS-B tier, the FR24
  tier and the OpenSky gap filler. Measured 14,436.
- Vessels: 33k claimed, 50,024 tracked.
- Satellites 16,095, quakes ~290/day, jamming ~220 flagged cells.
- Military aircraft: ~140 claimed, 29-62 across two counts. Lowered rather
  than explained away; the tier that fed the old number is not what feeds it
  now.
- The news feed answers `enabled: false` on a keyless boot, so its row says so.

The tour intro now dates the screenshots and names the script that re-shoots
them, so the next reader can tell how old the figures are without checking git.
@AndrewCTF
AndrewCTF merged commit c83f244 into master Aug 8, 2026
2 checks passed
@AndrewCTF
AndrewCTF deleted the readme-text-corrections branch August 8, 2026 05:20
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant