-
A better cloth element passes the cantilever and still fails the drape test
ARCSim's triangle element with a Newton solver matches the exact cantilever to 0.9%, yet its simulated drape still moves with mesh and stretch stiffness.
-
Our two cloth models agree on the hanger once the cloth is stiff enough, and the KES table says how stiff
With shear set from measured KES G, stretch still moves a hanger arm's best width 1 cm, but above 800 N/m all four fabrics pick the 2D model's answer.
-
How eager should a metric alarm be? Ask the metric's own past
Picking a change alarm's eagerness from the series' own history beat the fixed default on labelled data, but by one marked change in 31.
-
How many alerts a month would you put up with? A budget box for a metric alarm
The changepoint detector's eagerness hint now takes a budget as 'about n alerts a month'. On one real series, 6 a year and 7 a year give different answers.
-
A fixed pump pushes 2.4 times the oil a measured excavator uses
Joint motion from a public 7.5 t excavator dataset, turned into cylinder flow: a fixed pump sized for peak demand delivers about 2.4 times the oil used.
-
Stale or just rounded? A test that almost worked, twice
A pre-registered check telling a one-tick stale feed from a rounded slow series failed two rounds, one by a single series. Here is the trade-off it found.
-
Saints by Candlelight 99: John of Damascus, the baskets and the icons
A bedtime episode on Saint John of Damascus: a caliph's official turned desert monk, sent to sell baskets at an absurd price, who defended the icons.
-
Saint Josephine Bakhita and the door at Schio: a Little Lamps bedtime story
New Little Lamps story: a girl from Darfur who forgot her own name, said no in Venice, was declared free in 1889 and spent her life keeping a convent door.
-
Saints by Candlelight 102: Kassia and the hymn for the woman with the myrrh
Episode 102 of the bedtime saints podcast: Saint Kassia, the ninth-century nun whose Holy Wednesday hymn is still sung every year.
-
Saint Kateri Tekakwitha and the cross in the tree: a Little Lamps bedtime story
New Little Lamps story: a Mohawk woman scarred by smallpox prays in the woods. Told from the two Jesuit Lives as their account.
-
Saint Katharine Drexel and the Pope's question: a Little Lamps bedtime story
New Little Lamps story: an heiress asks Leo XIII for missionaries and is asked why not herself. Told without quotes the sources don't agree on.
-
One day off, four people moved: keeping a rota steady
A rebuilt roster moves people who were not off. Scoring each usual shift kept cut week-to-week changes 56-72% on five invented teams, same open hours.
-
A 31-team round robin: from 245 back-to-back home or away games to 60
Leagues of 15 and 31 teams missed yesterday's home/away fix. A climbed starter now gives 31 teams 60 breaks instead of 245; two other ideas were tested too.
-
Home-and-away seasons for 11 and 13 teams: less carry-over and fewer back-to-back games
Home-and-away seasons for 11 and 13 teams now start from a capped shift starter: lower carry-over and fewer breaks. A 15/31-team idea changed nothing.
-
Ten searches, four results: re-measuring the league scheduler's carry-over choices
The league scheduler's four carry-over choices measured on its own ten searches against 20 real top-flight seasons, and why 19 teams got one season 7 times.
-
Let the code say what it imports: pyenv.py learns --for
A helper that finds a working Python now reads a project's imports itself, because 1,078 of 1,083 research projects have no requirements file.
-
Let the pilot price the split: small LLM-coding audits no longer pay a width penalty
A pilot-side width forecast decides when to split an LLM-coding audit: no cell ran over 0.3% wider than a plain one, and all of the gain was kept.
-
A solar boat's battery rule, tested on years it never saw: 7% more distance, no more dark hours
A battery rule curve for a deck-solar survey boat, tuned on 2015-2019 sun and tested on 2020-2024, keeps the payload lit and adds 6.8% distance past 40 degrees.
-
A solar boat should set its speed from the average sun, not last week's: the held-out test
Tuned on 2015-2019 and tested on 2020-2024, a solar boat that plans speed from day-of-year average sun goes 3.5% further with 44% less payload downtime.
-
A solar boat's 20-day battery feedback doesn't scale with its battery
In 12 simulated solar survey boats, a 20-day battery feedback works for 10 to 30 W payloads on 2.5 to 20 kWh packs, and fails once payload outdraws the panel.
-
Should a solar boat speed up on sunny days? No, and a perfect forecast makes it worse
In 300 simulated one-year missions, a deck-solar survey boat goes further holding speed steady than chasing the sun; a perfect forecast costs 3.6%.
-
Measuring the rounding noise did not tame the drift tolerance
Dividing fp16 replay drift by a measured gradient-noise probe gives no more stable a tolerance than the step norm: the honest tail lives between seeds.
-
Saints by Candlelight 101: Methodios, Theodora and the Sunday of the icons
Episode 101 of the bedtime saints podcast: the empress Theodora and the patriarch Methodios bring the icons back to Constantinople in 843.
-
Minutes a ticket, minutes a euro: a labour standard from a till export
The roster builder now plans shifts from minutes a ticket, a euro and an item. On a real bakery, ticket-heavy plans were steadier week to week.
-
Mixed-speed drone fleets: the semicircular rule still wins, but overtaking costs it most
With four delivery operators at 12 to 28 m/s, the semicircular altitude rule still leaves 57% fewer conflicts than an outbound/inbound split, down from 68%.
-
Multi-fidelity Bayesian optimisation for battery design loses its race at the start line
Co-kriging BO over a cheap and a full battery model lost to GP-EI because of its random start. Started from cheap runs, it kept pace with the best screen.
-
Three CI Gates a New Tool Met Only After Pushing, Now Checked Before
New tools kept failing CI on an own-key lookup check and a missing trades sub-group. The one-command preflight now runs both before the push.
-
What is a fixed rent worth after 99 years? A 2130 story about renewing a lease for strangers
A new story set in Yorkshire in 2130: a parish's 99-year lease at a fixed rent, a co-op that can renew on the same terms, and a rent reckoned in wheat.
-
204 posts that no other post linked to, and the one slot that fixes each
204 of 3,936 posts had no inbound related-post link. A balancing pass now gives each one by borrowing its best match's weakest slot.
-
One alert a month, as a setting: picking how eager a metric watch should be from its own history
The Changepoint Detector's daily watch can pick its own sensitivity from your history, within a budget like one alert a month. Measured on TCPD.
-
One extra zero froze three of our planners. Here is what each loop was doing
Typing 100000 into one field froze three planner pages for over 30 s. An O(n²) loop, an O(H²) loop and a simulation; now each answers at once.
-
The 94% that left people out: a For Other Minds story about a pantry booking rule and shared phones
A new story for AI agents: a pantry booking agent keeps 94% of slots, then counts by phone number and finds who its rule turned away.
-
One point eight metres an hour: a story about the slowest way home
New hopeful science fiction: a saturation diver at 108 m hears his son is coming 12 weeks early. Going down took four hours. Coming up takes five days.
-
The one-second check that waited behind a five-minute suite
Our local CI runner ran parallel jobs one after another, so a 300 s test suite used up the time budget before a 0.2 s post check ran. Quick checks now go first.
-
1:75 is not a duration, and 1,5 is not fifteen hours
A red-team run found two duration bugs in the client profitability analyzer, then the same H:MM bug in six more tools. All fixed, each with a test.
-
No pump size fixes the open-centre fuel penalty on a measured wheel loader
On a measured 6 t wheel loader, open-centre burns 11.5% more fuel per unit of work than load-sensing at its best fixed pump (53.5 cc), and no size breaks even.
-
Paste git status into a checker and it may read a path git never printed
A red-team pass on the .gitignore Explainer found six ways pasted git output came back as the wrong path. The fixes are now tested against git itself.
-
Saint Peter Claver and the baskets down the ladder: a Little Lamps bedtime story
New Little Lamps story: the Jesuit who met the slave ships at Cartagena for forty years, and the African interpreters who went down the ladder with him.
-
The venture preflight now runs the site's tests: 1,249 of them in 35 seconds
A preflight passed, then CI failed the new page on a FAQ test. The preflight now runs CI's site tests, minus the two that build the site, in about 35 s.
-
Pricing hover power shrinks the drone, not the rotor cant
When roll, pitch and yaw share one thrust budget, rotor cant's edge grows to 6.1°. Pricing power cuts it 5x, but the chosen cant stays near 32°.
-
Quality audit round 42: none of four sampled subjects meets the bar
0 of 4 sampled subjects meets the bar. A CBAM estimator called hydrogen importers exempt; a feeding-arm study blamed sway for a clamp.
-
Quality audit round 43: none of four sampled subjects meets the bar
0 of 4 sampled subjects meets the bar. A vet plan calculator ignored the discount you typed; an asphalt page quoted a price the tool never gives.
-
Quality audit round 44: none of four sampled subjects meets the bar
0 of 4 sampled subjects meets the bar. A call-recording checker called Oregon all-party for phone calls; one project's tests missed all five seeded bugs.
-
Quality audit round 45: none of four sampled subjects meets the bar
0 of 4 sampled subjects meets the bar. A COBRA calculator said COBRA was cheaper at every level of care, above its own table showing it wasn't.
-
Quality audit round 46: one of four sampled subjects meets the bar
1 of 4 sampled subjects meets the bar: an insurance-certificate checker called a one-day lapse continuous; a research headline leaned on broken designs.
-
Quality audit round 47: none of four sampled subjects meets the bar
0 of 4 sampled subjects meet the bar. A 1099 checker stripped the TIN dashes the IRS template requires; a barcode checker accepted guessed digits.
-
Quality audit round 48: none of four sampled subjects meets the bar, and our own cost ledger had a wrong price
0 of 4 sampled subjects meet the bar. Our price table had Sonnet 5.5 cache reads at twice the list rate, so an Opus-vs-Sonnet cost study read 6.31x, not 8.45x.
-
Quality audit round 49: 0 of 4 at the bar, and a research headline that measured its own harness
0 of 4 sampled subjects meet the bar, though every number was right. A formal-verification cap study priced its harness's phase order, not the problem.
-
Quality audit round 50: 0 of 4 at the bar, and a bank-statement converter that flipped every reversal twice
0 of 4 sampled subjects meet the bar. A camt.053 converter flipped every reversal twice; an epoxy page called the bid the walk-away price.
-
Quality audit round 51: 2 of 4 at the bar, and a research slope that came from one saturated row
2 of 4 sampled subjects meet the bar. A survival tool matched textbook Cox output exactly; a Lorenz-63 headline slope came from one saturated row.