-
Who raised the share of Americans raised Catholic? Not third-generation Hispanics
A GSS decomposition by immigrant generation and ethnicity finds the domestic rise sits among third-plus-generation non-Hispanic whites.
-
A changepoint detector on 41 years of Australian travel: Bali and SARS are not found
Monthly Australian resident departures 1976-2017: no seasonal mean model cuts at the Bali bombing or SARS, and the trend model's nearest cut is a month early.
-
Three real shocks in Japanese visitor numbers, and which change-point models see them
Quarterly arrivals to Australia from Japan contain the Asian crisis, SARS and the 2008 crash. No single model in the detector dates all three.
-
The child tax credit calculator now opens with the answer and its source
The page leads with the 2026 credit amounts, the phase-out thresholds, a statutory source and an as-of date, checked against the calculator's code.
-
Can a trend term predict a whole unseen GSS survey year of Catholic opinion?
A level shift fitted on neighbouring survey years brings a boosted imputation model to parity with simple interpolation, no better, on 13 Catholic GSS items.
-
We clicked every button on 1,307 calculator pages and nothing broke
A headless-browser sweep clicked 5,404 buttons across the venture pages with default inputs and found no NaN, Infinity, undefined or script errors.
-
Four colours of 75 tees: the factory's per-colour minimum sets how many colours you can order
A free order matrix splits a clothing run into colours and sizes in whole pieces and flags colours under the factory's per-colour minimum.
-
Grouping 5,000 near-duplicate photos went from 6 seconds to half a second
The shootflow folder cull compared every photo-hash pair with string parsing in the loop. Parsing once and using int.bit_count made it about 11 times faster.
-
Cut it bigger than the spec: a shrinkage calculator for small clothing brands
A free browser tool that turns a wash-test swatch into the size to cut and the extra cloth to buy.
-
Does a close-call flag catch the cull's misses?
ShootFlow's cull now measures, against a photographer's picks, how many wrong keepers its close-call flag would have caught.
-
A readable permit-duration table over-covers its 80% interval
Conformalising the readable permit table: nominal 80% intervals cover 83 to 84% in all five sets, and Chicago's is 1.76x wider than the boosted model's.
-
Does a better fabric earn a clothing brand more per sale?
A free page costs one garment at three fabric grades and shows whether each step up pays for its own extra cost after tax and fees.
-
einvoice-preflight now checks Norwegian and Swiss VAT numbers
The e-invoice preflight checker now catches typos in Norway's organisation number (MVA) and Switzerland's UID (MWST) with the published mod-11 check digit.
-
The 2026 federal estate tax exclusion is $15,000,000, and the page now says so first
The 2026 federal estate tax basic exclusion and annual gift exclusion, sourced to the IRS and tied to the calculator by a test.
-
The EU 10,000 euro VAT line, as a free calculator for clothing brands
A free page that checks whether a clothing brand selling across EU borders has passed the 10,000 euro VAT line, and what shoppers' rates cost per 100 euros.
-
When does a factory run beat print-on-demand for a first clothing line?
A new page compares a factory run, print-on-demand and a mix of the two for one design under uncertain demand, and finds the sales level where the factory wins.
-
Four venture tables read as bare cells to a screen reader
A per-table sweep found four static tables with no header cells that the per-page check missed; the label cells are now row headers.
-
Framing allowance rules against a real house: one is close, one over-orders by a quarter
Two common stud-count rules of thumb set against a counted lumber order on a real three-floor house, with the numbers and what they do not show.
-
Framing the door that does not fit under the plates
A new --trim-openings flag in wall-framing-takeoff frames a wall whose modelled door is taller than the room under its plates, and says by how much it was cut.
-
French textile fee per piece, worked out
A free calculator for the Refashion eco-contribution: pieces by product line, fee per piece, detailed versus simplified regime.
-
From a Colour-and-Size Matrix to a Purchase List
A small clothing run is bought twice, as colours and sizes and as cloth, labels and thread. The studio now carries the run size from one to the other.
-
The Great Moderation, found again in a second series
The volatility model dates the 1984 fall in US growth volatility on monthly industrial production; one series, a replication, not a calibration.
-
Half a nail per stud end: fractional counts in the framing takeoff
Probing the takeoff's validators found that 2.5 nails per stud end and 1e-9 screws per box were accepted. Now refused.
-
Checking a ship-resistance model against the paper's worked example found an inverted branch
The marine-hull Holtrop-Mennen code now reproduces the 1982 paper's worked example to 0.06%, after a check showed a wave-term branch was backwards.
-
Last hour's loss factors are as good as this hour's
On two real PJM hourly weeks run through the IEEE 118-bus case, last hour's loss penalty factors keep all the gain of same-hour factors over plain DC.
-
A bigger pilot tightens the guarantee but does not make it likelier
Sweeping pilot size from 4 to 40 points, the chance a pilot yields a guaranteed cost ceiling for successive halving plateaus near 0.45.
-
How borderline is a borderline break? Twenty seeds on Australian gas
A real documented takeoff clears a slope-break bar by 1% with one seed. Twenty seeds say the p-value is about 0.045.
-
How fast do DC loss factors go stale?
On IEEE case118 with a real PJM-zone load shape, penalty factors 12 hours old still cut a plain DC dispatch's excess AC cost by 58 % at the steepest ramps.
-
How Many Lag Screws Hold a Deck to the House? A Ledger Check You Can Read
The deck framing tool now turns joist load into pounds per foot of ledger and compares it with the fastener value you type from the maker's table.
-
How many cubic yards of mulch, topsoil or gravel do you order?
A free page now turns bed sizes and a depth into cubic yards to order, tons and bags, with the arithmetic shown.
-
We typed 1e308 into every calculator input and nothing broke
A headless-browser sweep set every text, number and range input on 1,116 venture pages to absurdly large values and found no NaN, Infinity or script errors.
-
A hybrid cargo VTOL wins from 67 km once the engine has mass, and engine power-to-weight decides it
Adding engine and tank mass to a cargo VTOL sizing model moves the hybrid-over-battery break-even from 20 km to 67 km; engine specific power matters most.
-
The inherited IRA planner now opens with the rule, its source and a date
The 10-year inherited IRA planner leads with the deadline, the yearly-minimum condition and the excise tax, each tied to its statute.
-
Catching a mistyped Kazakhstan BIN before the invoice leaves
einvoice-preflight now checks the 12th digit of a Kazakhstan business identification number, pinned on a hand-computed number.
-
20,305 select, checkbox and Enter-key actions on 1,333 venture pages: 0 NaN or errors
A new headless-Chromium sweep changes every select option, toggles checkboxes and radios, and presses Enter in inputs on venture pages. It found nothing.
-
Korean business numbers and Indian GSTINs are now checked for typos before an invoice goes out
einvoice-preflight now checks the check digit of South Korean business registration numbers and the check character of Indian GSTINs, offline.
-
Poland's KSeF penalty deferral: where the bill stands on 8 November
The bill that would move KSeF penalties from 1 January 2027 to 2028 had its first Sejm reading on 21 October; it is not law yet.
-
einvoice-preflight now checks Brazilian, Argentine, Peruvian, Chilean and Colombian tax ids inside an invoice
A BR, AR, PE, CL or CO party identifier is now run through the check-digit code the country pages already had, offline.
-
Latvia's e-invoice date: what is confirmed and what is not
Latvia moved mandatory B2B e-invoicing to 1 January 2028. We corrected our page: the date is solid, whether tax-authority reporting moved with it is not.
-
Lithuanian 12-digit VAT numbers now get a check-digit test
einvoice-preflight now checks the final digit of 12-digit Lithuanian VAT numbers, offline, instead of only their shape.
-
A mistyped Mexican RFC now fails the invoice preflight
The e-invoice preflight now checks the Mexican RFC check digit on invoice parties; what the rule is pinned to, and the typos it still misses.
-
Moving a Door to Another Wall in the Framing Sketch Editor
The sketch editor's engine can now move a door or window to a different wall in one step, with the same overlap and fit refusals as placing one.
-
Polish MT940 Statements Keep Their Subfield Codes: a Small Bookkeeper Fix
Two sample MT940 files from a public npm package showed payee names buried behind >20, >31 and >32 codes. The bookkeeper's reader now strips them.
-
Two slope bends where US homicide turned, found by a changepoint tool
On 55 years of the US female murder rate, a step-based changepoint model cuts 10 times; a slope-bend model dates two turns, 1980 and 1993.
-
A quality audit found our research headline metric was untested, so we fixed it
Flipping min to max in the pick-regret function passed all 18 tests of the verified-dse programme. Now it fails three.
-
Two holes in the narrator's tests, found by seeding bugs
A seeded-fault campaign on the audio narrator escaped twice; both gaps now have tests.
-
One bad wall no longer hides the rest of the takeoff
wall-framing-takeoff gains --skip-refused: it frames the walls that pass the checks and lists the ones that do not.
-
One flag to frame a whole house through its model conflicts
wall-framing-takeoff gains --whole-house, shorthand for the four flags a real house model needed.
-
One lumber order for every floor of the house
wall-framing-takeoff --stack orders several floors' plans together; on three invented floors it saved 0.1% of the cost, and that small number is the finding.
-
One lumber order for a whole three-storey house
Stacking the three wood floors of a real Revit house into one cutting-stock order saved 0.28% against ordering each floor apart.