diff --git a/daily-finds.log.md b/daily-finds.log.md index d825be6..fd06eb5 100644 --- a/daily-finds.log.md +++ b/daily-finds.log.md @@ -4665,3 +4665,22 @@ Amazon: n-a (a manuscript in a museum). - **saint-olafs-church-wasdale-head** was the fourth-ranked candidate and the closest call. Held on the same principle that held `mac-racken-bridge` on 07-29: the piece rests on "England's smallest active parish church" and on the Viking-longship roof beams, and the source for both is an Atlas Obscura entry. The church has a real parish and the Lake District has no shortage of published local history, so this is a good candidate for a thin day with time to dig — but a review that can only cite the entry it came from is a paraphrase. - **fulton-steamboat-willie-playing-cards** clears the bar and was held for a duller reason: the interesting review is about the printing — the holographic foil process, the English handmade tuck stock, why no two emboss alike — and none of that is on the product page. It wants a second source or the deck in hand. - **chipbuilder** and **chronos-terminal-games** both hold up as finds and both would need to be *played* through to review honestly, which is not something this run can do. + +## 2026-08-01 — copy edit + +Fresh-eyes sonnet subagent over all 28 pieces of prose written this run (19 find bodies, 6 bundle blurb-plus-intro pairs, 3 review bodies). **Two prose changes — the first non-clean run since 07-25 — and the pass earned its keep twice over, because it also caught three cross-piece inconsistencies and correctly declined to fix any of them.** + +The two prose edits, both accepted: +- `fulton-steamboat-willie-playing-cards` — cut "this is one of the more charming things anyone has done with it" and put the fact back where the praise had been: "this twenty-dollar deck puts the 1928 mice and their shipmates on the court cards." Exactly the principle-1 fix the charter asks for — generic praise replaced by the specific thing. 74 → 62 words. +- `agloe-paper-town` (review) — "airtight and rather beautiful" → "airtight", and "a very good story" → "a good story". Two vague intensifiers, neither carrying information. 367 → 363 words. + +**The three catches, all resolved in the main session, because all three were factual calls rather than prose ones:** +1. **`champion-lodgepole-pine` contradicted itself.** Subtitle said the tree "germinated fifty years before Plymouth Rock"; the body says it started growing around 1560. That is sixty years, not fifty. The error came from compressing Atlas Obscura's "more than 50 years before the Pilgrims landed" into a round number that then read as exact. Subtitle now says sixty. +2. **The `archimedes-palimpsest` subtitle undersold the body, on both the find and the review.** It read "the only surviving Greek copy of three Archimedes works", which is a weaker and different claim than what the bodies actually support: two of the three (*The Method*, the *Stomachion*) survive in no copy in *any* language, and only *On Floating Bodies* is a Greek-only survival. Both subtitles now read "the only known copies of two lost Archimedes works" — precise, and the three-work nuance stays in the bodies where there is room for it. +3. **`praia-de-santa-monica` and its bundle cited opposite ends of the same range.** The find's subtitle says eleven kilometres, the bundle description said eighteen; the find's body explains that sources disagree and gives the range. Nobody reading one page would notice, but the two are adjacent in the feed. The bundle description now says "eleven-to-eighteen kilometres", which matches the body and keeps the honest uncertainty visible. + +The subagent's reasoning for escalating rather than fixing was right in all three cases: which end of a disputed range to cite, and whether a subtitle's claim is weaker than its body's, are editorial-fact decisions, not wording ones. + +**Spot-check honoured rather than skipped**, per the standing 07-25 instruction. The subagent reported measuring rather than estimating and gave per-file counts; four were re-counted by hand in the main session — `saint-olafs-church-wasdale-head` (81), `archimedes-palimpsest` find (82), `cheapfoodmap` (77) and the `archimedes-palimpsest` review (405) — and all four matched its figures. Every find body lands in 62–82 against the 40–90 budget; the three reviews at 367 / 363 / 405 against 250–450, none of them reaching for the narrative extension the charter allows history subjects. + +Format exception recorded again so it is not re-litigated: `joke-two-eyes` is 14 words against a 40–90 budget. Flagged, judged, left alone — the joke-find format is the setup and the punchline and nothing else, as with `joke-broken-pizza` on 07-29. diff --git a/src/content/bundles/2026-08-01-the-smallest-the-largest-the-emptiest/index.mdx b/src/content/bundles/2026-08-01-the-smallest-the-largest-the-emptiest/index.mdx index 15eae12..943cdf3 100644 --- a/src/content/bundles/2026-08-01-the-smallest-the-largest-the-emptiest/index.mdx +++ b/src/content/bundles/2026-08-01-the-smallest-the-largest-the-emptiest/index.mdx @@ -2,7 +2,7 @@ title: "The smallest, the largest and the emptiest" date: 2026-08-01 tags: [daily-finds, travel-and-places, outdoors-and-nature] -description: "Three superlatives you can walk up to: England's smallest active parish church, California's largest recorded lodgepole pine, and eighteen kilometres of almost empty beach." +description: "Three superlatives you can walk up to: England's smallest active parish church, California's largest recorded lodgepole pine, and eleven-to-eighteen kilometres of almost empty beach." blurb: "Three records, all of them visitable, none of them behind a rope." items: - saint-olafs-church-wasdale-head diff --git a/src/content/find/2026/08/archimedes-palimpsest/index.mdx b/src/content/find/2026/08/archimedes-palimpsest/index.mdx index 44a7ffc..2e23140 100644 --- a/src/content/find/2026/08/archimedes-palimpsest/index.mdx +++ b/src/content/find/2026/08/archimedes-palimpsest/index.mdx @@ -1,6 +1,6 @@ --- name: "The Archimedes Palimpsest" -subtitle: "A prayer book written across the only surviving Greek copy of three Archimedes works" +subtitle: "A prayer book written across the only known copies of two lost Archimedes works" date: 2026-08-01 shortlist: true promotedTo: "archimedes-palimpsest" diff --git a/src/content/find/2026/08/champion-lodgepole-pine/index.mdx b/src/content/find/2026/08/champion-lodgepole-pine/index.mdx index 97d04a8..4cbc732 100644 --- a/src/content/find/2026/08/champion-lodgepole-pine/index.mdx +++ b/src/content/find/2026/08/champion-lodgepole-pine/index.mdx @@ -1,6 +1,6 @@ --- name: "Champion Lodgepole Pine" -subtitle: "A 110-foot pine that germinated fifty years before Plymouth Rock" +subtitle: "A 110-foot pine that germinated sixty years before Plymouth Rock" date: 2026-08-01 link: "https://www.atlasobscura.com/places/champion-lodgepole-pine" linkText: "atlasobscura.com" diff --git a/src/content/find/2026/08/fulton-steamboat-willie-playing-cards/index.mdx b/src/content/find/2026/08/fulton-steamboat-willie-playing-cards/index.mdx index 03c97a0..80c4567 100644 --- a/src/content/find/2026/08/fulton-steamboat-willie-playing-cards/index.mdx +++ b/src/content/find/2026/08/fulton-steamboat-willie-playing-cards/index.mdx @@ -11,4 +11,4 @@ tags: [playing-cards, public-domain, giftable, printing] description: "A $20 deck built from the public-domain Steamboat Willie characters, with fully holographic foil backs in an embossed castle-grey tuck of English handmade paper." --- -Steamboat Willie entered the public domain in 2024, and this is one of the more charming things anyone has done with it: a twenty-dollar deck whose court cards are the 1928 mice and their shipmates. The backs are holographic foil end to end. The castle-grey tuck is handmade English paper, embossed, and takes the pressing differently every time — so the box is slightly different deck to deck, which a printed one never is. +Steamboat Willie entered the public domain in 2024, and this twenty-dollar deck puts the 1928 mice and their shipmates on the court cards. The backs are holographic foil end to end. The castle-grey tuck is handmade English paper, embossed, and takes the pressing differently every time — so the box is slightly different deck to deck, which a printed one never is. diff --git a/src/content/reviews/agloe-paper-town/index.mdx b/src/content/reviews/agloe-paper-town/index.mdx index b2cf8ea..4996273 100644 --- a/src/content/reviews/agloe-paper-town/index.mdx +++ b/src/content/reviews/agloe-paper-town/index.mdx @@ -19,11 +19,11 @@ import hero from './hero.png'; > Two cartographers cannot independently invent the same fake town — which is why one of them planted Agloe on a map, and why the story of what happened next is only about half verifiable. -The logic of the copyright trap is airtight and rather beautiful. Two mapmakers surveying the same county will draw the same rivers and the same roads, so a shared feature proves nothing. But a settlement that never existed can only get onto a rival's map one way. Around 1925, Otto G. Lindberg and his assistant Ernest Alpers of the General Drafting Company — who made the road maps distributed by Esso — put a dot at an empty junction in the Catskills and labelled it Agloe: their own initials, O.G.L. and E.A., shuffled. +The logic of the copyright trap is airtight. Two mapmakers surveying the same county will draw the same rivers and the same roads, so a shared feature proves nothing. But a settlement that never existed can only get onto a rival's map one way. Around 1925, Otto G. Lindberg and his assistant Ernest Alpers of the General Drafting Company — who made the road maps distributed by Esso — put a dot at an empty junction in the Catskills and labelled it Agloe: their own initials, O.G.L. and E.A., shuffled. Then the trap sprang the wrong way. In 1930 a company called Agloe Lodge Farms was incorporated, acquired a fishing lodge nearby and named it the Agloe Lodge. A thing invented so that it could be proven not to exist had acquired a business address. -Here the record thins, and it is worth being straight about where. The version that circulates — that Rand McNally later printed Agloe, that General Drafting sued, and that Rand McNally won by pointing at the Agloe General Store as proof the place was real — is a very good story that the documentation does not carry. Wikipedia's account, sourced to the cartographer Frank Brown, has Rand McNally answering that its information came from **Delaware County records**, which was enough to head off action. Local residents dispute that a general store ever stood there at all. Several links in the popular chain are marked as uncited. +Here the record thins, and it is worth being straight about where. The version that circulates — that Rand McNally later printed Agloe, that General Drafting sued, and that Rand McNally won by pointing at the Agloe General Store as proof the place was real — is a good story that the documentation does not carry. Wikipedia's account, sourced to the cartographer Frank Brown, has Rand McNally answering that its information came from **Delaware County records**, which was enough to head off action. Local residents dispute that a general store ever stood there at all. Several links in the popular chain are marked as uncited. That leaves the durable part, which is strange enough without embellishment. A dot placed to catch a thief attracted a business that took its name from the dot. The name then propagated through county records and later through digital maps, and Agloe kept appearing and disappearing from Google Maps decades after anyone had a reason to draw it. Nothing was ever built to make Agloe real. It simply got referred to enough times. diff --git a/src/content/reviews/archimedes-palimpsest/index.mdx b/src/content/reviews/archimedes-palimpsest/index.mdx index c2cc0e7..1e5d3bf 100644 --- a/src/content/reviews/archimedes-palimpsest/index.mdx +++ b/src/content/reviews/archimedes-palimpsest/index.mdx @@ -1,6 +1,6 @@ --- name: "The Archimedes Palimpsest" -subtitle: "A prayer book written across the only surviving Greek copy of three Archimedes works" +subtitle: "A prayer book written across the only known copies of two lost Archimedes works" date: 2026-08-01 category: history topics: [history-and-archives, learning-and-reference]