# 2022 Railway station access
###### tags: `projects osm be`
## Goals
Collect data on accessibility of all railway stations
## Data model
What do we need in terms of data:
Pictures of objects:
- station in general
- every steps
- every elevator
- an indoor map
STATION LEVEL
* Add a picture of an indoor map (information board)?
* Are bicycles allowed in the railway station?
ELEVATOR DATA
* Which platforms are reachable with an elevator? (highway=elevator of elevator=yes)
* Do regular bicycles fit in the elevator?
* Are bicycles forbidden?
* How wide is the door? (wheelchair)
* Which floors does the elevator reach?
* Wheelchair-friendly? (not relevant at the station level)
* Does the elevator work?
STEPS DATA
* Which platforms are reachable with an escalator?
* Which platforms are reachable with normal steps?
* Escalator or normal steps?
* Bicycle wheel ramp available? (oprijgoot)
* Does the escalator work?
## Tools / method
* start with random sample of railway station and try to map them as good as needed
* replicate during an event
* collect data by surveying
* clean up the result
* Railway station theme MapComplete
* Query logic to translate to "is the railway station accessible"
## Can we add the pictures to [Wikimedia Commons](https://commons.wikimedia.org/)? How does this work?
Yes! We can use Wikimedia Commons for this project. There's still some work to be done to figure how we can efficiently upload the pictures and structured data to the platform though. Below several ideas which came out of a discussion with Geert from Wikimedia Belgium (thanks for your advice, Geert!).
Freedom of panorama allows us to take pictures of all accessibility objects in railway stations. There's one exception: it's not allowed to upload pictures of indoor maps due to copyright restrictions.
At the start of this project, we have created a new Wikidata item Q109973555 called "Railway station access" so that we can link all pictures to this item.
In order to be able to easily query the data afterwards, it's best to add the following statements as structured data to each picture:
* Items portrayed (depicts) in this file (P180) referring to the Wikidata item (Q number) of either the railway station or the newly created Wikidata item Q109973555 for this project
* Add the location of the picture (Add SDoC property Location: Name of the train station). Example: P276:Q660792 (location:station Brussel Noord). SDoC = Structured Data on Commons (tab Structured data on Wikimedia Commons).
* Add a qualification "Applies to part" (P518) to specify what accessibility object (e.g. steps, platform, elevator) is on the picture
* I believe we could also add P17:Q31 (land:Belgiƫ) to qualify the global "Railway station access" object type.
Query to search for applicable train station images: https://commons.wikimedia.org/wiki/Special:MediaSearch?search=haswbstatement:P276=Q660792
As a general example, https://commons.wikimedia.org/wiki/Special:MediaSearch?search=haswbstatement:P170=Q2602433 gives all pictures on Wikimedia Commons with Michiel Hendryckx (Q2602433) as creator (P170).
Last but not least, it's also good to know that:
* There's a Wikimedia Common app which allows users to directly upload pictures to the platform.
* If we ask the username of each contributor, we could check out the pictures they uploaded (with a specific name given by the surveyor) and add structured data to the pictures.
* If there are GPS coordinates attached to pictures, then they will also be uploaded to Wikimedia Commons.
* It's possible to add links of Wikimedia Commons images to specific objects on OSM ([more details on wiki](https://wiki.openstreetmap.org/wiki/Key:wikimedia_commons)).
## Related projects
- https://wiki.openstreetmap.org/wiki/OpenStationMap
- https://github.com/iRail/stations#facilitiescsv
## First meetup to explore our data model
We basically know the starting and ending point of the project:
* Starting point: with a group of volunteers, we aim to crowdsource accessibility data for all Belgian railway stations.
* Ending point: there's open data available on the accessibility of Belgian railway stations. Based on the open data, it's possible to create an overview of the accessibility per railway station.
What's missing: a simple data model and tool for the data collection.
In order to find out how we can easily collect the data, here's what we're going to explore during our first meetup:
* Meet in a railway station and collect the first data and pictures
* During the first data collection, we'll learn what's easy to collect (basic data) and what takes more time to collect (advanced data)
* We'll add the first data to OpenStreetMap and Wikimedia and discuss how we can make it easy for volunteers to collect data themselves.
## Result of our first our data model exploration
During our next first meetup, we collected the first data in a railway station and discussed what data model would allow us to collect accessibility data of all railway stations with volunteers. Here's a recap of the **data challenge** we want to set up to make this happen.
What we need for this data challenge: a map editor without showing a map
What this would look like - basically 3 pages:
1. An overview page with a list of railway stations
* The overview page shows the total progress (in %) of our data collection.
* There are 3 categories of railway stations: easy (1 volunteer needed), regular (2 volunteers needed) and complex (3 volunteers with at least one expert mapper needed).
* Volunteers can click on their preferred railway station and claim their spot by registering with their OpenStreetMap and Wikimedia account
2. A list with all objects per railway station
* Based on the exisiting OSM data, volunteers can see what's already mapped and what's missing.
* Volunteers are asked to complete the existing data and verify the missing data by clicking on the specific objects.
3. A list of all attributes per object
* Attributes are specific questions per object that volunteers need to answer.
* In this attributes section, we collect both structured data and pictures.

The data challenge is completed as soon as we reach 100%. With 100% of the data collected, we could create a clear overview of the accessibility per railway station.
As a next step, we want to discuss our proposed data model with NMBS/SNCB, the Belgian national railway company.
Verdere opvolging:
https://docs.google.com/document/d/1y3QNZTOOWZLDDbsdZZBM_WIoDFQXNOUmh6CpSA8W-6U/edit?usp=sharing