| assets | ||
| Customizations | ||
| Logic | ||
| UI | ||
| vendor | ||
| .gitignore | ||
| Helpers.ts | ||
| index.css | ||
| index.html | ||
| index.ts | ||
| land.html | ||
| package-lock.json | ||
| package.json | ||
| Quests.ts | ||
| README.md | ||
| test.html | ||
| test.ts | ||
MapComplete
Let a thousand flowers bloom
MapComplete attempts to be a webversion of StreetComplete. However, we focus on 'themes', a bit similar as mapcontrib.
The design goals of MapComplete are to be:
- Easy to use
- Easy to deploy
- Easy to modify
The basic functionality is to download some map features with overpass and then ask certain questions. An answer is sent back to OpenStreetMap.
Furthermore, it shows images present in the image tag or, if a wikidata or wikimedia_commons-tag is present, it follows those to get these images too
Examples
- Buurtnatuur.be, developed for the Belgian Green party. They also funded the initial development!
- Cyclofix, further development on Open Summer of Code funded by Brussels Mobility
- Bookcases cause I like to collect them.
Have a theme idea? Drop it in the issues
License
GPL + pingback.
I love it to see where the project ends up. You are free to reuse the software (under GPL) but, when you have made your own change and are using it, I would like to know about it. Drop me a line, give a pingback in the issues, ...
Dev
To develop:
- Install
npm. - Run
npm installto install the dependencies - Run
npm run startto build and host a local testversion
To deploy:
rm -rf dist/to remove the local buildnpm run build- copy the entire
distfolder to where you host your website. Visitingindex.htmlgives you the website
Architecture
High-level overview
The website is purely static. This means that there is no database here, nor one is needed as all the data is kept in OpenStreetMap or Wikimedia (for images).
When viewing, the data is loaded from overpass. The data is then converted (in the browser) to geojson, which is rendered by Leaflet.
When a map feature is clicked, a popup shows the information, images and questions that are relevant for that object. The answers given by the user are sent (after a few seconds) to OpenStreetMap directly - if the user is logged in. If not logged in, the user is prompted to do so.
Searching images
Images are fetched from:
- The OSM
image,image:0,image:1, ... tags - The OSM
wikimedia_commonstags - If wikidata is present, the wikidata
P18(image) claim and, if a commons link is present, the commons images
Uploading images
Images are uplaoded to imgur, as their API was way easier to handle. The URL is written into the changes
The idea is that one in a while, the images are transfered to wikipedia
Privacy
Privacy is important, we try to leak as little information as possible. All major personal information is handled by OSM. Geolocation is available on mobile only throught hte device's GPS location (so no geolocation is sent of to google)
TODO: erase cookies of third party websites and API's
Attributions:
Data from OpenStreetMap Images from Wikipedia/Wikimedia
https://commons.wikimedia.org/wiki/File:Camera_font_awesome.svg Camera Icon, Dave Gandy, CC-BY-SA 3.0
https://commons.wikimedia.org/wiki/File:OOjs_UI_indicator_search-rtl.svg Search Icon, MIT
https://commons.wikimedia.org/wiki/File:Trash_font_awesome.svg Trash icon by Dave Gandy, CC-BY-SA
https://commons.wikimedia.org/wiki/File:Home-icon.svg Home icon by Timothy Miller, CC-BY-SA 3.0
https://commons.wikimedia.org/wiki/File:Map_icons_by_Scott_de_Jonge_-_bicycle-store.svg Bicycle logo, Scott de Jonge
