From OpenStreetMap Wiki
Jump to navigation Jump to search


On F1 tracks the finish line is usually in a different place to the start line - presumably these want tagging too?

we probably need start, finish and start_finish. we also may need to consider the case where different start and finish locations are used for different races, club vs pro, etc. Nfgusedautoparts (talk) 19:22, 24 March 2015 (UTC)

other potential roles

we might want a role for the timing line, which on modern circuits is the spot where the transponder loop is buried in the pavement

id tag

we may want an id tag to allow queries for specific relations; the application might be variants on a circuit (e.g., Watkins Glen long course vs short course, with and without the inner loop chicane), or to distinguish historical variants no longer used (e.g. Watkins Glen first circuit, second circuit, and third circuit.)

i have been using start dates and end dates to sort out historical circuits, and have been putting them in OHM when appropriate (e.g., WG 1st through 3rd are in OHM, 4 is in OSM, with identical tagging style so the same overpass query and processing code can be used in either case.) Nfgusedautoparts (talk) 15:13, 26 August 2019 (UTC)

end and start dates

we should list these as optional tags.

These are presently used for physical objects regarding their construction. As this relation is for an activity/use I would think start and end dates are not appropriate? If the idea is to specify the operating dates then open_hours tag would be suitable? Warin61 (talk) 23:05, 19 May 2017 (UTC)
in OSM, i think using start_date to indicate when tracks go into operation is fine. in OHM we use both start_date and end_date for historic objects, and there is value in consistency. see my watkins glen/leaflet test for an example of the value:


Would the proposal include using this for e.g. horse-racing tracks?

i would doubt it. horse tracks have their own tagging distinct from motor racing.

Motor_vehicle not motorcar

A better, more inclusive, term is motor vehicle. Motorcycles are lumped in with motor car as are trucks, electric and solar vehicle as these use an electric motor. Warin61 (talk) 23:01, 19 May 2017 (UTC)


Tend to vary the course from one race to another .. may not be of much use? Warin61 (talk) 23:01, 19 May 2017 (UTC)

This is also the case for most rally circuits for racing cars, trucks and motorcycles, whose circuit changes at every edition of the race, the circuits being used only one day with no other planning for its reuse.
Even for formula 1, the circuits may be change at every edition, and dedicated circuits for racing are rare, and exist only for training in a closed private area with much shorter loops.
Competitions will occur in larger areas, on a circuit where standard highways will be used (e.g. streets in Monaco for Formula One races, important intercity roads in Le Mans for the 24-hours car or motorcycle races) ; note also that the racing circuit will use the highways in a oneway direction which is not necessarily the same (oneway or bidirectional) direction of these highways when they are returned to the public trafic. — Verdy_p (talk) 01:46, 20 May 2017 (UTC)

Every track type

I found this proposal very interesting, and think that couldn´t be limited to motor vehicles racing tracks (highway=raceway). I see that leisure=track combinated with suitable sport=* fail to verify in JOSM (and OSMose) showing a warning, in case that the track has got multiple variants, and one is a closed loop, and the rest are unclosed way (connecting two sections of main track, like chicanes). I see this in a recent edition of a cycling race track with a near to oval loop main track and some shortcuts and chicanes. About roles, may be a main role (for the main or most used variant) and variant role (with name like "short course" or "formula track"). Some role or identifiers could be sport or category (due in some cases acording of category using the circuit is the long of the track or the variant used) --Camello AR (talk) 06:16, 2 June 2020 (UTC)

circuit relation vs route subtype

would the need be better served by adding new types to route? just a thought Nfgusedautoparts (talk) 18:43, 11 January 2020 (UTC)

I like this, but there may be concerns about potential conflict with the definition of type=route and compatability on route=*. -- Kovposch (talk) 20:39, 23 March 2020 (UTC)