-
Olympus raw files, without the converter
ShootFlow now develops Olympus and OM System ORF files itself. On 18 real bodies, from a 2002 compact to the OM-1, every sensel matches LibRaw.
-
One utility bill flipped every sign in a bank file
ai-bookkeeper read 19 more real US bank exports. A 528-row Ally file came in with every sign backwards, and Stripe payouts at Chase were suggested as transfers.
-
One Mercury download, three bank accounts: what a real LLC's file taught the bookkeeper
A real LLC's Mercury download held two checking accounts and a card in one file. Now split, autopays paired, 90% right on the owner's labels.
-
One set of ballots, three winners: a free ranked-choice counter
A free tool counts ranked ballots by instant runoff, STV and Schulze, every round shown. A separate Python recount caught a parsing bug first.
-
Two shops, one staff list: what pooling people is worth
Shift Roster Builder plans several shops from one staff list. On 40 made-up weeks, letting people work at both shops left 55% fewer hours open.
-
Panasonic raw files, read from the top down
ShootFlow now develops Panasonic Lumix RW2 files itself. On four real bodies, GF1 to GF7, all 50 million stored sensels match LibRaw.
-
Pentax raw files, and where the camera really crops
ShootFlow now reads Pentax PEF files itself: all 36.6 million sensels of a K-1 Mark II frame match LibRaw, and the camera's JPEG is not centred.
-
Five chant notes that disagreed with the book: one typo, four the scanner's
Five chant notes disagreed with the 1908 scan: one GregoBase typo, four one scanner fault. Fixing that fault cleared fourteen differences.
-
Shift scheduling: a 60-person week's proof gap went from $20,010 to $30
The roster builder proved a 60-person week's open hours but counted the whole pay bill as unproven. Pricing its clique LP cut the gap to 0.16% of payroll.
-
A shortlist is not a short ranking: finding your top 3 by picking pairs
Ranking a whole list and keeping its top 3 often gets the cut wrong. Asking only about options near the cut got it right twice as often in simulation.
-
A show of hands picks the wrong shortlist more often than the model does. Combining a team's top 3
On 432 simulated teams, a show of hands on everyone's top 3 named the group's true shortlist 72% of the time; the Bradley-Terry model, 78%.
-
A wandering pace is not boredom: telling a careless block from a random walk
A rushing-respondent test flagged 17% of honest people whose pace wandered. A step-vs-random-walk test, on only when the panel wanders, cut that to 3.4%.
-
Close calls are not a settled order: fixing a ranking page that stopped too soon
The free list ranker said 90% settled when a near-coin-flip answerer's order was barely better than chance. The cause was the prior, not stopping.
-
How many pairwise comparisons does it take to rank a list? We measured it
Every pair once is n(n-1)/2 questions. An adaptive ranker needed 93 picks for 20 options and 204 for 40, for an order nearly as good. Short lists differ.
-
"Much better" or "a bit better": graded picks on a decision matrix, and the stop rule they nearly broke
Graded picks on a decision matrix: on 300 simulated people the true best came first 182-203 times against 165, once the stop stopped trusting 'much' too far.
-
Rank a list by picking pairs, in your browser, and the stop rule that never stopped
A free page ranks any list from this-or-that picks, with rank ranges and team results. Its first stop rule never finished; here is the one that does.
-
The coin flips before the run: cutting a bored respondent from where they sped up
A slot-run test missed the random clicks a bored respondent gives first. Cutting from their speed-up instead gained up to +0.0045 Spearman, flagging nobody new.
-
The constant that was not constant: why our team ranking read disagreement at a third of its size
Our team ranking read disagreement at 30-60% of its true size. One dropped term made a Laplace fit into PQL. Kept, a true 2 reads 2.0-2.3.
-
Your team, or people like your team? Two rank ranges for one combined ranking
A combined team ranking needs two ranges: one for these people, one for the group they stand for. Read the first as the second and 90% ranges can hold 74%.
-
Reading Nikon Raw Files Without a Converter: 130 Million Sensor Values, All Matching LibRaw
ShootFlow now reads Nikon NEF files directly. On seven real bodies, D70 to D810, every one of 130.6 million sensor values matched LibRaw, in 0.1-0.6 s per file.
-
Samsung raw files, and a filter tag that lies
ShootFlow now develops Samsung NX SRW files itself: all 90 million sensels on four bodies match LibRaw, and two files get their own filter layout wrong.
-
A cheaper map for the roster search: two more shifts filled
When the pricier pay LP no longer fits the time budget, the open-hours LP now steers the scheduler. It filled one or two more shifts on 2 of 10 test weeks.
-
Her own rate downtown: per-person site wages in a two-shop roster
One line on a staff row now gives a person their own rate at one shop. On made-up weeks, ignoring those rates understated payroll by 2 to 4%.
-
Let the lower bound steer: the LP that proves a roster's cost also helps find a cheaper one
Steering a staff schedule search with the LP behind its lower bound halved the remaining pay gap and filled open shifts on 3 of 9 test weeks.
-
Per-shop pay rates should come with the export, not be typed in
Shift Roster Builder now reads per-shop pay from a staff CSV or 7shifts. On 40 made-up weeks, imports matched hand-typed rates to the cent.
-
Ten open hours back on a 100-person week, by giving the slower proof less time
On the biggest weeks the roster builder spent 9 s on a proof that never finished. Giving that time to a cheaper LP left 376 open hours instead of 386.
-
Start the second LP where the first one ended: a warm start that cuts the pay floor's pivots up to sixteenfold
The roster builder's pay floor solved its LP from scratch after a sibling LP on the same rows. Starting from that basis: same bound, 107 pivots, not 1,860.
-
The city shop pays more: pricing a higher minimum wage into a two-shop roster
A roster that knows only each person's own wage understated payroll by about 5% on made-up weeks with a Seattle-rate shop. Pricing the floor cut 2.2%.
-
The drive is on the clock: what a two-shop roster costs when the trip is paid
A two-shop roster that ignored paid travel sent people across town 6 to 8 times a week, about 2.5% of payroll it never showed. Pricing it fixed that.
-
The gap was the search: an integer program, a missed roster, and re-solving one day at a time
A 60-person week sat $30.11 above its bound. HiGHS found a valid roster $0.36 above it; re-solving one day at a time closes most of that in 2 s.
-
The last seven dollars: why local moves can't finish a 60-person roster
A 60-person week sits $7 above a proven floor. Pair swaps, cyclic trades and an LP dive failed: the gap is one trade through 52 people on every day.
-
The swap the call-out check forgot, and why fixing it made a café worse
A quick call-out cover check never let the caller take a later shift. Fixing it sharpened the count, then made a short-staffed café's weeks worse.
-
Thirty minutes across town: the trip a two-shop roster never checked
On made-up split-shift weeks, half the days someone worked at both shops left them 30 minutes for the trip. A travel rule fixes it, sometimes at a price.
-
Twenty weeks was not enough: the call-out result that flipped at 200, and a price for the second call-out
A café roster that looked worse over 20 simulated weeks of call-outs was better over 200. A new check prices call-out pairs covered alone but not together.
-
Sony raw files, without the converter
ShootFlow now develops Sony ARW files itself: compressed, uncompressed and lossless. On six real bodies every sensel matches LibRaw.
-
Sony writes its lens names in hex
Three of ten real Sony and Samsung raw files name no lens in EXIF. Their maker notes do, and reading them settles three more range ties too.
-
sshd keeps the first value: a free sshd_config checker
A free tool reads sshd_config and its drop-ins in OpenSSH's order: the value that wins, lines sshd ignores, unread drop-ins and Match-block traps.
-
Straightening a tilted chant page
Half a degree of tilt made the chant scan reader lose every staff on a drawn page. A shear found from the row profile brings them back.
-
Taking out the daily cycle did not make the alerts point at anomalies
On NAB's 47 real hourly series, removing the daily cycle left the change watch's alerts near chance. A drift model still points at anomalies best.
-
Ten teams cannot beat 94: a plain search closed what a SAT solver could not
A 150-line C branch and bound proves no 10-team round robin has non-cyclic carry-over below 94, which the league schedule generator already reaches.
-
The 400 amp click when the tennis feeder switches on
Closing the tennis feeder's e-stop contactor onto empty ESC capacitors would peak near 394 A on assumed figures; a 4.7 ohm precharge cuts it to 20 A.
-
Shift scheduling: the steadier plan's extra churn came from its anchor, not its weeks
Keep-usual-shifts plans moved 4-20% more shifts when solved column generation first. The cause was their anchor week. Fixed: same plans, 1.3-1.8x faster.
-
The chant clef in the middle of the line
A clef printed mid-line made Chant Forge misread whole chant verses. A rule tested on 14 unseen pieces cut differences from 129 to 13 and found a slip.
-
The chant-scanner gap that wasn't there
A chant scan reader was said to miss liquescents that end a porrectus. Measured on 19 before building a fix, it already read 16.
-
The chant notes a scanner could not see
Liquescents, the small half-sung notes of Gregorian chant, were invisible to Chant Forge's scan reader. A new rule reads 20 of 26 on pieces it had never seen.
-
The chant page that curled at the top
Nine unexplained differences against the 1908 Graduale: three slips in the files, two reader faults, and four from a page curling at its head.
-
The cheapest Texas plan at 1,000 kWh was the dearest for most homes we billed
We billed 200 simulated Texas homes on three example plans. The one with the lowest 1,000 kWh price was the cheapest for 54 and the dearest for 138.
-
The contractor on the payroll journal: three ways a pay report broke our books
A Gusto-style payroll journal found three faults: contractor pay booked as W-2 wages, an employee dropped over a reimbursement, employer taxes never booked.
-
Our Privacy Footer Was 499 Words, and Half of It Was No Longer True
The privacy footer listed 54 tools and 40 hosts inline and called them all public lookups. It now counts them, folds the list, and says what they call.
-
The Green Button File Your Utility Gives You, and Two Ways a Tool Can Misread It
Our demand-charge analyzer refused a common Green Button CSV layout and could not open the XML. Both are fixed, with daylight saving read from the file.