Using OSMCha to understand bad edits
2019-09-22, 14:00–15:00, Mathematikon C

Protecting OpenStreetMap is a continuous process performed by Mapbox to secure maps from displaying erroneous edits. Any edits that raise suspicion are flagged in OSMCha, an open service that allows to check low-quality changes that are made by the members of the OSM project in a shared database. This not only helps to report our findings to the community but examine them in aggregate and draw conclusions to improve our data quality processes.


It is important that the reviewer has the necessary information about the changeset and understand the diff of modifications in the OSM data.

For this purpose, OSMCha offers an interface similar to OpenStreetMap but with additional tools such as changeset-map to visualise the edits, information about the mapper, other information related to the changeset and OSM user history that can help the reviewer identify problematic edits on OSM.

During this workshop the following aspects of work with OSMCha will be highlighted:
- the basic principles of work with the interface,

  • analysing data on the map view
  • flagging changesets as bad or good with varying levels of severity
  • getting information about changeset, data and contributor

  • the possibility of filtering out the data in a granular approach that helps to limit the amount of information about changes in OSM, in terms of

  • contributors, where you can pick up the data for a particular mapper(s) or mapping teams

  • the date range, you can limit a number of changesets by the dates of submission to OSM and by the dates when they were reviewd in OSMCha

  • the reasons for suspicion, OSMCha has a range of internal functions that assign different reason for suspicion to changeset that you can use for reviews

  • area, where you can set up area of interest for edits inspection

The main goal of the workshop is to engage the community to evaluate the existing data using OSMCha and improve its quality.


Co-authors – Wille Marcel Talk keywords – OSMCha, data analysis/quality, vandalism