Hi there, I work for a Local Authority, and we are going through the process of collecting data through site visit etc for our public transport network. We’d like the data to be freely accessible by everyone.
The data is currently being stored in Excel (noted not the best repository but a start), confident it can be exported into which ever format is needed py or json.
We wish to create a process to update and maintain nodes across our region in OSM. It is essential that we can do the process ourselves without 3rd party support - and installing new software etc is unlikely to be easy.
My proposal is to:
check for items (likely just nodes) in OSM online and add them (manually) if they are missing
take the OSM ID and match it in our excel “database” which contains all the possible tags
take our updated tags, now matched to a node ID
generate a profile with updated tag, i.e. “capacity=100”, “wheelchair=yes” etc
Just as additional bit of info, the poster above has already mentioned this to the DWG and I suggested to discuss here. I’ve already mentioned the import guidelines as describing the process by which these sort of things are done.
The OSM wiki page may be out of date, but although it mentions that geojson.io can import OSM xml files it doesn’t mention uploading them to OpenStreetMap. I also can’t see it in the help pane there either. Is this functionality documented somewhere else?
There was a recent(ish) thread by someone at the Scottish Borders Council that was looking to conflate their NaPTAN data with what has previously been created in OSM. I don’t know how uniform the tooling is among local authorities, but they may have a partial workflow that could be adapted rather than creating one from scratch. Of course NapTAN’s only a small subset of things that might be considered public transport data, so the two threads might be talking about apples and oranges.
It’s not entirely clear what types of features you will be trying to match with OSM data. For some things nodes will almost certainly be fine, but if you are looking to conflate e.g. capacity=* for a car park (amanity=parking) associated with a park and ride you’ll probably have to handle ways as well even if you only do the matching based on centroid. Something like Overpass can give you centroids for areas if needed.
As you may expect, different LA, different problems!
Would you have any other suggestion to upload to OSM? I’m sure I read this step somewhere but now I can’t find it! This is the one last step I can’t find a clean answer for - the actual update of the map!
Just to recap, the import procedure can be found at https://wiki.openstreetmap.org/wiki/Import/Guidelines. I don’t think I’ve seen a mention of a licence yet? I’m sure it won’t be a problem, since the LA wants to add the data to OSM, but they’ll need to say what licence they are making the data available as.
The best software to use will probably depend on the quantity of the data and what it represents.
The representation of Public transport information in OSM has varying levels of complexity depending on what is being recorded and only some of it is represented as nodes.
The records in your internal dataset are unlikely to have identical column names and values to what we use for our key=value pairs so some sort of conversion will be required. There will probably be people here who can offer suggestions where the wiki is unclear or out of data, but we will need to know more information first.
