OpenTransport Belgium is a project to extract data from transport corporations all over Belgium.
It's goals are to scrape (or to get by ourselves) :
- routes
- stops
- timetables
- real-time
And then provide a nice, unified API fot thoses. We would also like to store all thoses data over time to provide historical data
##Usecases Usecases would/could be :
- Nice websites to replace often old one from the corporations
- a Raildar.be (see Raildar.fr)
- predict timetables (eg: show somehow that train 1234 if always 5 min late (or early))
- bind with OSM to provide multimodal, realtime nice routing (for eg with http://navitia.io/)
- push back stations and routes to OSM
- GTFS and GTFS realtime exports
- ?
I think it should be done that way : (for each corp)
- Gather stops and routes and provide a nice API
- Gather static timetables and provide a nice API
- Provide GTFS with both of the above
- Get realtime data(=last known position and if provided, lateness) (often only the last station is provided), store it and provide an API
- Make something that extrapolates the last position to the current position and that fires events whith positions (with, if possible, (arrival) predictions from the coorp) (note: if we have realtime gps position, this step is trivial)
We should try to make abstractions so that each "layer" could be usable with another coorp if applicable
##Available datasets : None of thoses are yet standadized or stored Maybe we could find some more at http://data.irail.be/
Static:
- stations
- lines
- timetables
Dynamic:
- realtime (=last stop) python lib
Static :
Dynamic:
- Last stop for every current train
Maybe :
- Every 6 min, the actual past arrival/departure time for past stops + theorical arrival/departure time and planned lateness for future stops for every train (http://archive.irail.be/)
Static:
Static:
None
Make this data very accessible and well documented because it was a real pain in the ass all thoses sources