Identifying stuff in the transport field is a mess. In every certain field there are different identification methods and they are difficult to interconnect.
For example, when one dataset from a bus company is opened, it takes a lot of fiddling to make the right connections with a train company's dataset. Similar problems exist when programming things cross border (in different countries, the same stops/stations/things have different identifications) or even in the same datasets, to tell which stop is connected to another stop (sometimes they have the same name, in other cases they are a bit different (e.g. an extra id concatenated to the name)).
Using the vocabulary we have discussed at opentransport/vocabulary we will set up a site which can reuse data from various sources and link them all together.
It should have functions for three different end-users:
- Data owners: add a link for datadump of stops which can be automatically fetched. Provide mappings to the stations.io scheme.
- Data consumers: get the right datadump for your application with the right IDs from different systems
- Data enrichers: add data for one specific stop
(e.g. an open street map contributor or a transport agency employee)
- Clean datasets and do named entity recognition
- Link to global
stop_area
andstop_point
IDs - -add yours-
(e.g. an app creator, a map designer or an accessibility researcher)
- An autocomplete list generator
- Get the right
connection_link
times to create links between modes - -add yours-
(e.g. an app creator who wants to provide a feedback button in his app, an integrator who finds mistakes in the dataset, a transport company that is moderating the data, and so on)
- Being able to do PUT and POST requests on a URI
- A simple interface to view, edit and add feedback to a representation
- -add yours-