OpenStreetMap as infrastructure - a localgov map?

04 October 2014

The Moabi project is reusing the tools of the OpenStreetMap project to map natural resource use in the Democratic Republic of the Congo. This is an example of what Mikel Maron (from the Moabi project) and Elizabeth McCartney (from the US Geological Survey) called ‘OpenStreetMap as Infrastructure’ in their recent talk at State of the Map US ie taking the OpenStreetMap tool-chain and applying them to new problems.

I got to experiment with some of the OpenStreetMap tool-chain at bit during recent work at the Land Registry and, now in its 10th year, the OSM tools are really impressive:

  • The core OSM website is a solid product providing storage of geospatial data, history of changes, user accounts etc

  • The new in-browser editor iD is very simple to use, and is pretty easy to customise (I think this is the where Moabi have done the bulk of their customisation)

  • There are various tools for rendering map tiles, checking data quality, and importing and exporting different formats

Above all though OpenStreetMap is a set of tools for building consensus around ‘things in space’.

Which brings me onto local government. Local government has lots of information about ‘things in space’: the extents of conservation areas, redevelopment zones, dates and times of bin collections by area, locations of air-quality stations etc.

Some councils helpfully publish this data online. The best have done the hard work to clean the data, convert it to open standards and figured out the licensing. Lambeth Council even publishes areas with protected vistas of St Paul’s Cathederal as GeoJSON.

The trouble is - publishing, quality and format are patchy and inconsistent between councils and that makes reuse harder.

So here’s the proposition: could the OpenStreetMap tools be used to build a LocalGovMap? Not necessarily as the definitive source of data for evermore (although I guess it could be), but to build consensus about what local government datasets should look like, using a single collaborative space.

A similar approach is working in central government, where designers from across departments are using wikis to figure out the best way of designing user interface elements. So why not across local council officers who are the custodians of geospatial data?

I doubt it would take much. Probably a couple of councils coming together and a month or so of a Ruby on Rails / javascript developer (the core OSM website is written in Rails, and the iD editor is a javascript app).