SYMCA's contract figures come from Vix. This platform measures the same buses from the same BODS feed and reaches its own answer. Here they are side by side, journey by journey, for whatever export has been loaded. Journeys are matched on the scheduled departure from the first stop — the only key the two systems share, since Vix names a journey by its start time and a running board, and we hold a GTFS trip id.
Both systems name a journey by its departure from its first stop, which is a residential street where the bus starts rather than the point a printed timetable is headed on. Origin and destination are given so a row can be placed without reading the route, and they come from the timetable rather than from what was observed — a journey picked up forty stops in still starts where it starts.
The reference columns are the operator's own name for the work — 0512B, 0519C —
Vix's taken from its export and ours recovered from the raw SIRI-VM archive, where BODS
publishes it as DatedVehicleJourneyRef in the same vocabulary. They are not
used to match the two sides, so where both are present and identical they are independent
confirmation that the journeys were paired correctly.
The rows Vix sent, after the pivot has been unpicked: it writes twenty-one lines of filter names, a header that is mostly blank, and a cell only when it changes, so a stop row carries no date, no service and no journey of its own. Every heading below explains what the column actually holds — hover it — and those notes come from reading this export, not from a specification. Deviation is shown in seconds rather than Vix's decimal minutes, because −0.1833 of a minute is not a number anyone reads.
Tracked is how much of the journey was recorded, against the stops the timetable gives it. It is the difference between a journey and a trace: on 10 September five of the fifteen journeys only we have were tracked over 97–100% of their stops, and the rest were fragments of 1–57%, which say more about that day's recording than about the other system. Only the complete ones are counted as journeys Vix has no record of.
A journey one side has and the other does not is the column worth reading. It is not an error to be reconciled away: one of the two systems saw a bus the other did not, and which is right has to be settled case by case. The same goes for the stop counts — where they differ on a matched journey, one side tracked the bus over more of its route.