This is a testing site only. See the live Public Lab site here »

Public Lab Wiki documentation

MapKnitter Cloud Exporter

2 | 10 | | #23345

« Back to MapKnitter

The new Cloud Exporter for MapKnitter uses a cloud-based process to produce full-resolution exported images and map tiles from maps.

To use it, shift-drag, shift-click, or long-press (on tablets) the images you want to export, then press the "download" button in the upper left of the screen, as shown here:


Your export will appear under the Exports tab alongside previous exports.


From the Export tab, you can export a map in a variety of formats. If you've changed the map at all since last exporting, you can re-run the export and the previous export outputs will be saved. All formats are generated sequentially, so if you're waiting for a JPG you will have to let all the other formats complete first.

This may take some time - a large map could take as long as an hour or more. Closing the page will not stop the export.

Comparison to the previous export system

What is the same?

  • Same: all previous formats are still exported
  • Same: closing the page will not stop the export.
  • Same: privacy or lack thereof in anonymous maps

What are the differences?

  • New: each export is now saved in a list, rather than overwriting previous exports
  • New: you must select images in order to export
  • New: exports run only on selected images, not all images
  • New: where to find the “start export” button

Get Involved

Please help provide feedback here:

Here is another demo at higher resolution:



Mapknitter Cloud Exporter was one of four main coding projects of 2019's Google’s Community Atlas Grant to Public Lab. These four parts are:

  • Redesigning MapKnitter’s overall UI: tour feature, gallery, etc.
  • Rewriting the foundation of MapKnitter’s backend to 2019 standards (up from 2012 standards)
  • Overhauling and rewriting Leaflet Distortable Image Library for the rubbersheeting we use
  • Cloud Exporter!

Sasha, Vladimir, Varun, Sebastian, and others spent several months working on this system in 2019. It connected with this global collaboration involving over a dozen people: :D