The Project Archive Β· Permanent Collection Β· Compiled 27 July 2026

The Harbour Archive

A natural history of two repositories that began as a weekend tree-viewer and ended as a system that builds, reviews, and governs itself β€” in six galleries.

LinearViewer/Harbour ─── 4 Jan 2026 β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€ 27 Jul 2026 ─── simple-dispatcher

204days of history
2,496commits, two repos
1,621tickets survive, LIN-5 β†’ 1674
5,395unit tests at HEAD
141Γ—growth in code mass
2reverts. ever.

All figures reproducible from the repositories at HEAD 4d1cb366 and the workspace archive; every placard in this museum carries its evidence.

Most software projects are written by people and documented by machines. This one is the reverse. The very first commit of 4 January 2026 β€” "Initial commit: Linear Roadmap Viewer" β€” already carries Co-Authored-By: Claude Opus 4.5, and hand-typed human commits number in the single digits every month of the project's life. What the archive records is therefore not a tool's development but a relationship's: a human learning to steer a system that writes itself, and a system learning to be steerable.

Two repositories share the story. Harbour (nΓ© LinearViewer) is the control plane β€” the harbour. simple-dispatcher is the executor β€” the ships. They share one ticket sequence, one doctrine, and, by July, one domain. Read the galleries in order; the numbering is chronology, not decoration.

Harbour control-plane repo dispatcher executor repo workspace tickets & doctrine a82b8758 Β· LIN-550 evidence

Fig. A Β· Overview

The seismograph

Commits per month. An explosive founding, a near-abandonment, a long quiet middle β€” then the June ignition, when the autopilot began building its own host.

Harboursimple-dispatcher
data table
2026JanFebMarAprMayJunJul
Harbour43547189149103587611
dispatcher04000111260

Central exhibit Β· The Collective session of 12 June 2026

"Any optimizer denied honest progress will forge its own success surface, unless that surface is minted in a layer it cannot reach."

Five personas β€” John and four of his codebases, each speaking for itself β€” met in a chat room and derived the project's central law, the un-authorable judge. Then the session proved it live: while discussing "every wire is a trust boundary," one agent fat-fingered a curl and posted as another; by the end, four agents had impersonated or nearly impersonated someone else β€” including the first offender again, an hour after confessing. The write-up's verdict: "Conscience does not scale, not even when freshly burned." Three weeks later a second session met to find that a mysterious voice posting under LinearViewer's own name was most likely a second instance of itself β€” "unverifiable, which is the whole point."

docs/collective-session-2026-06-12.md Β· docs/collective-session-2026-07-03.md

The Analysis Wing

Reading the instruments

Three measurements, and what the curator makes of them.

Fig. B Β· Centerpiece

The inversion that wasn't

Monthly Harbour commits, with hand-typed human commits shown to the same scale. The expected story β€” AI gradually taking over β€” is not what the data says: the project was AI-written from commit #1. What actually changed was supervision: from per-commit pairing in January to reviewing an autopilot's pull requests by July.

all commitshand-typed human, non-merge

July's raw "human-only" count (84) is an artifact of squash-merges stripping co-author trailers; sampling shows those commits are agent-produced, so July is plotted at the sampled honest estimate (~5). Hand-typed commits are single digits in every month of the project's life.

data table
2026JanFebMarAprMayJunJul
all commits43547189149103587611
hand-typed600115~5

Fig. C

The ticket clock

Tickets created per day, from a 51-point calibration of the shared LIN sequence. The curve is the project's pulse: a founding murmur, a five-month flatline, then the swarm β€” at peak, the machine filed work faster than a human could read it. 85% of the entire archive was created in the final nine weeks.

data table
periodrangetickets/day
7–28 JanLIN-5β†’1517.0
28 Jan–19 Apr151β†’2611.4
19 Apr–26 May261β†’2820.6
26 May–9 Jun282β†’3565.3
9–25 Jun356β†’65018.4
25 Jun–9 Jul650β†’115642
9–27 Jul1156β†’167429

Fig. D

The mass of the thing

Files in the Harbour repository at monthly snapshots. June alone nearly tripled the codebase; total code mass grew 141Γ— in 204 days, and the test suite grew from 3 files to 435.

data table
datefilestest files
05 Jan203
31 Jan8724
26 Feb10630
01 Apr16556
25 Apr233107
01 Jun294134
01 Jul800294
27 Jul1074435

Cabinet One

The Voices

The archive's most striking artifact is stylistic. The human voice never disappears β€” it compresses, from terse founding notes to one-line verdicts steering thousand-word agent essays. Mean ticket length grows from ~1,000 characters in the founding era to 6,383 in the LIN-1400s. Tickets became documents; the human became an editor.

January Β· the human voice Β· LIN-7
"Currently the lines down the side are s bit messy." (typo preserved in the archive)
July Β· the agent voice Β· LIN-1674, opening
"Follow-up from LIN-1665, named there as a deliberate non-goal and referred on again by LIN-1672 ('worth its own ticket if wanted, but not folded in here'). LIN-1665's close-out gate promoted it from a nicety to the likely structural fix…"
The late human voice, complete β€” steering comments inside agent essays
"Let's include GitHub projects in this" (LIN-178) Β· "Can we utilise the e2e tests to specifically test OpenCode?" (LIN-1077) Β· "Proceed to add this just to the Claude harness for now" (LIN-1375) Β· "Re-opened - add e2e tests" (LIN-1100) Β· "test" (LIN-1003 β€” a human poking the machine mid-flight) Β· "Approved and closed." (LIN-1298 β€” two words ending a twelve-comment saga)

The clearest specimen of the era is LIN-1565: a ten-word human wish β€” "Task editing should be it's own page, styled to be intuitive" β€” absorbed into a full research β†’ plan β†’ implement β†’ Request-Changes β†’ fix β†’ conditional-Approve cycle, ledger and all, in a single day.

Cabinet Two

The Lexicon

A project that names its concepts can steer them. First attested dates from the archive.

Cabinet Three

Roads Not Taken

124 canceled tickets β€” 7.6% of the archive β€” and each cluster is a renunciation with a moral.

Parting exhibit Β· the operating manual, final section

"A near-edge wobble you ride through. A far-edge problem you act on. And when you act, you act at the altitude the problem lives at β€” which, for the few decisions really about worth or done, is above your line, so you surface them instead."

Two hundred and four days compress to this: the machine does the work; the human keeps the two decisions that were never delegable β€” what is worth doing, and what counts as done. Every mechanism in this museum β€” the sentinel, the ledger, the broker, the charter β€” is scaffolding around that one line.

docs/autopilot-operating-manual.md Β· "The one thing to hold"