About

Who keeps this database, what goes into it and what is refused, how to say it is wrong, how often it changes, and what becomes of it if the person keeping it stops.

Who publishes this

Lap Ledger is built and kept by Alex Farley, one person, in the open. It is not a Formula One publication: it is unaffiliated with Formula One, with the FIA and with every team, nothing here is licensed from any of them, and nobody at any of them has checked it.

The build, the checks that gate it and the source data they read are all public, so what this page says about itself can be read rather than taken on trust: the repository.

What goes in, and what is refused

A fact goes in only with a source somebody else can check, and it carries the standing of that source with it — an official FIA or formula1.com source is the only thing that reaches the top of the confidence ladder, and nothing is promoted to it without one. A figure nobody has established is left empty rather than guessed at, and an empty figure is never a zero.

Two things are refused outright. A source whose licence does not allow its rows to be passed on is classified as such on the register, and the build refuses a database in which any row cites one. And Formula One race timing, which nobody publishes under terms that permit passing it on, so lap times, stints, race timing and race control messages ship as four empty tables on purpose.

When two sources disagree

Neither reading is picked quietly. The disagreement is written into the database as a row of its own — both figures, and what each source says — and published beside the data; a fact nobody has established is listed in the gap register with what would close it. Both registers ship inside the file you can download, so a question this database has not settled is one you can see rather than one you have to catch it out on.

If something here is wrong

A reader who has spotted a wrong figure is the cross-check this project cannot run on itself. The report form asks for the page, the figure and what says otherwise, and for nothing about how this project files its own work.

What a report gets is an answer on the issue it opens and one of four outcomes: a correction, a recorded disagreement, a recorded gap, or a reason it was not taken. It is read by the one person named above, who does not do this full time. There is no desk behind it and no promised turnaround, and saying so is better than promising one nobody is staffed to keep.

Found something wrong? Report it. What two sources disagree about is recorded and published rather than quietly picked, and what nobody has established is listed as a gap.

How often it changes

F1DB publishes a race within a day or two of it being run; the harvest here is re-fetched from it every morning at 06:00 UTC, and every three hours from a weekend’s first practice session until three days after its race. It is published only once the rebuilt database has passed every check, so the race record follows a race within hours of F1DB publishing it. Everything else — the eras, the regulations, the registers, the prose — moves when somebody works on it.

Every page names the database version it is running on, the date that database was built, and a digest of the file itself. The digest is what fixes which figures you were shown: the version moves on a release and the build date only when the harvest does, while the copy this site serves is rebuilt whenever anything changes — so a version and a date together can name more than one file, and a digest names one.

If this stops

The database is a pure function of its sources and a published build script: the same inputs produce the same bytes, which is why the build date inside it is a constant rather than a timestamp. Every release carries a SHA256SUMS file and this site serves one for the copies it hands out, so a copy can be checked against the digests the build produced rather than against this site still being up.

The data is offered under CC BY-SA 4.0 — with the columns named in meta.project_prose_columns, this project’s reading of each disagreement, its account of each gap and the prose of ten of its twelve reference tables, under CC BY 4.0 — and the code under the MIT licence, except the livery palette in web/src/lib/liveries.js, which is CC BY-SA 4.0; the circuit centrelines are © OpenStreetMap contributors under ODbL 1.0, and that obligation follows f1-geometry.db alone. So if the person named above stops, what has already been published stays readable, checkable and rebuildable by anybody holding a copy — which is the most this page can honestly promise, and worth more than a promise to carry on.