From OpenStreetMap Wiki
Jump to: navigation, search

OpenChargeMap ( is not a potential datasource for positions of electric vehicle charging points in the UK due to its incompatible licenses (see below). But, as every other commercial geodatabase, you may use it to find missing objects in OSM, visit them in field and add them to OSM if they are existing.

Common Abbreviation

Open Charge Map is commonly known as OCM. This abbreviation normally includes the database, the contained data and the web and software apps used to maintain and access the data.


From the OCM About page:

Our aim is to work with the community to develop and provide a high quality, public, free, open database of charging equipment locations globally.

Our goal is to avoid the proliferation of independent conflicting charging location maps/websites/applications and to provide (as much as possible) a reliable single point of reference for charging equipment location information. Instead of competing with other data providers we aim to cooperate with and provide services to them, in exchange for data sharing and de-duplication efforts.

There are many recognised data sources around the world holding electric car charging data. Many of them are all holding the same data just duplicated. The problem is that not all of them holds all the data and so right now it is a nightmare for an electric car driver to get reliable and complete information on where he/she can charge their EV. The concept behind OCM is to pull together the multitude of electric vehicle charging data bases into a single, open source, community maintained database and so creating a single point of reference worldwide for electric vehicle charging data.

OCM has a web site where new charging stations can be added or existing ones edited by the public. Those edits go onto an edit list where one of the country volunteer editors can review the data entered, make changes and then publish. Only data that is reviewed and published by a volunteer country editor is available to the public.

Data can be accessed via the OCM web site or through the various phone apps developed by OCM.

A public API is also available for software developers wishing to access OCM data. This can be used to develop third-party applications such as maps and route planning.

In addition to data entered manually through the web site/apps OCM also imports data from commercial and other sources. Once agreement has been obtained from the source owner the data is imported into OCM. The attribution and licence for each of the imported locations is set according to the source owner requirements. This data is not included in the open licence and is licensed individually per location. However, should an imported location be edited by a member of the community through the web site/app then the licence for that location is reassigned to the open licence and is then considered open data.

As of Jan 2015 there are 27,821 EV charging locations listed worldwide on OCM.

Importing from OpenChargeMap?

Data is "derived from a wide variety of public sources and contributions" and so the issue of what licence is applicable cannot be determined for the OCM database as a whole. Instead, each location is licensed independently and a licence and attribution is attached to the data for each location. Here is an example of the data attached to each location. This is a cut and paste of the JSON data retrieved through the API...

   DataProvider": {
           "WebsiteURL": "", 
           "Comments": null, 
           "DataProviderStatusType": {
               "IsProviderEnabled": true, 
               "ID": 20, 
               "Title": "Automated Import"
           "IsRestrictedEdit": false, 
           "IsOpenDataLicensed": false, 
           "IsApprovedImport": true, 
           "License": "This data is provided by the National Renewable Energy Laboratory (\"NREL\"), which is operated by the Alliance for Sustainable Energy, LLC (\"Alliance\"), for the U.S. Department of Energy (\"DOE\"), and may be used for any purpose whatsoever.", 
           "DateLastImported": "2015-01-10T01:01:56.373Z", 
           "ID": 2, 
           "Title": ""

Given this API it is easy to retrieve only those locations that are licensed under the CC BY-SA 3.0 licence.

So, much of the data on OCM is imported from copyrighted sources and are not included in the CC BY-SA 3.0. For those data items the copyright is owned by the supplying data source and they are attributed on a per location basis. When using the API you can select just open data (included in the CC BY-SA 3.0 license) or all data which will also include that data copyrighted by third parties.

Annoying legal problems

  • It looks on the surface as if loading those locations that are not included in the CC BY-SA 3.0 license is not likely to be viable. However, if the CC BY-SA 3.0 is compatible with the OSM licensing then importing OCM data for those locations may be viable.
  • CC-BY-SA 3.0 is not compatible with OSM Contributor Terms.
  • Can we re-release this data under a OSM-CT-compatible license? Permission need from openchargemap probably
  • Charge locations which have been geolocated using Google services must not be imported to OSM.

It was the case that in the early days that, like several other community contribution sites, OCM used Google for geolocation. This is now not so and OSM is now used for geolocation. Google is just used for display purposes.

Other issues

  • Duplication with existing OSM entries. We have some electrical charging points in our database already. A straightforward import could result in duplications.
  • Duplication within OCM data. As many of the locations are imported into OCM from third-parties it is inevitable that the same location will be imported into Open Charge Map several times. Duplicate detection within OCM itself is sketchy at best and from inspection it is clear that within those 27,00+ locations there are many duplicates. Perhaps it would be a good idea for OCM to first embark on a de-duplication exercise prior to any potential import into OSM.
  • Accuracy. Some of their data may be located only by postcode (figure out if this is the case) This would make those locations rather poorer accuracy than is normal for OpenStreetMap. Note: I am not sure this is now true. AFAIK all OCM locations require a lat/long.

Both of these issues (and perhaps the legal concerns too) can be tackled by using this only as a reference data set, and allowing OSMers to cross reference to check if the data has been added yet in a similar approach to Bike Shop Locator.

I am not sure if the Bike Shop Locator uses the same approach but perhaps all the legal issues, and the bulk of the concerns regarding data imports, could be resolved by not actually importing any non-geo data from OCM into OSM. Instead, just a OCM reference back to the OCM database could be stored in OSM using the ref:ocm=* tag. This would have the benefit of allowing the charging station to be mapped through OSM but the details held on OCM.

Actually, given the nature of the data concerned this is by far the better solution. The detailed data regarding EV charging stations is of a specialist nature and so it is more suited to maintenance through a bespoke system. In addition, in the future, real-time, dynamic, status updating will become available and so real-time database updates will become a necessity and OSM is not ideal for such a dynamic data environment.

So this is my proposal...

  • that the licensing issues be reviewed initially and if deemed acceptable to OSM then an import project be started in OSM
  • that the import project have the primary aim to initially import all 27,000+ locations into OSM as a single node per charging station location with tags:
  • all detailed data referring to the charging station would remain on OCM and be maintained by OCM through their normal processes.
  • that the import project be approached on a country-by-country basis and be broken down into country tranches as per normal import guidelines
  • that there be included in the import specification methods to:
    • detect possibly duplication especially with existing charging_station nodes within OSM (currently number about 4300 of amenity=charging_station
    • detect removal of OCM locations and trigger removal of the matching OSM node (automation? - discuss).
    • detect addition of new amenity=charge_station tags on OSM and flag them up as possible new additions to OCM (automation? - discuss).
    • detect a change of geolocation on OCM and flag a change of location on OSM (automation? - discuss)

Other OpenChargeMap - OpenStreetMap Integration

If the OpenChargeMap developers are interested, there's plenty of potential for use of, or deeper integration with OpenStreetMap:

  • Use of OpenStreetMap as the base map. Already done when generating location coordinates. Google is used for some map displaying but this should have no licensing implications for OSM.
  • Use of OpenStreetMap data. Extract our charging points data and add it to the OpenChargeMap dataset. Technically possible, but lots of legal issues there (the main one being that OpenChargeMap re-releases their data under CCBYSA3.0) There's only about 100 nodes tagged with fuel:electricity=yes at the moment, and about 4300 tagged with amenity=charge_station.
  • If data was imported into OpenStreetMap, the deepest level of integration would be for OpenChargeMap to follow and approach as seen with OpenEcoMaps of becoming a sort of symbiotic data portal into OpenStreetMap. Contributions feed directly into OSM, and updates from OSM feed directly onto the OpenChargeMap displays. This is perhaps the preferred method and may answer many, if not all, of the licensing issues


Open Charge Map has made a great start to integrate the diverse electric car charging location data sources. It has a rapidly growing community maintaining and adding to the core data and as it has Open Source at its heart it has a lot of potential for the future.

However, OpenStreetMap will want to map the locations of electric car charging stations and it seems utter madness that this be done in isolation and independently.

See Also