OpenStreetMap

Illustration of OpenStreetMap data primitives (nodes, ways and relations)

OpenStreetMap uses a topological data structure, with four core elements (also known as data primitives):

• Nodes are points with a geographic position, stored as coordinates (pairs of a latitude and a longitude) according to WGS 84. Outside of their usage in ways, they are used to represent map features without a size, such as points of interest or mountain peaks.[25]
• Ways are ordered lists of nodes, representing a polyline, or possibly a polygon if they form a closed loop. They are used both for representing linear features such as streets and rivers, and areas, like forests, parks, parking areas and lakes.[25]

• Relations are ordered lists of nodes, ways and relations (together called "members"), where each member can optionally have a "role" (a string). Relations are used for representing the relationship of existing nodes and ways. Examples include turn restrictions on roads, routes that span several existing ways (for instance, a long-distance motorway), and areas with holes.[25]
• Tags are key-value pairs (both arbitrary strings). They are used to store metadata about the map objects (such as their type, their name and their physical properties). Tags are not freestanding, but are always attached to an object: to a node, a way or a relation.[25] A recommended ontology of map features (the meaning of tags) is maintained on a wiki. New tagging schemes can always be proposed by a popular vote of a written proposal in OpenStreetMap wiki, however, there is no requirement to follow this process. There are over 89 million different kinds of tags in use as of June 2017.[26]

The OpenStreetMap data primitives are stored and processed in different formats. OpenStreetMap server uses PostgreSQL database, with one table for each data primitive, with individual objects stored as rows.[27][28] From this, several database dumps are created, which are available for download. The complete dump is called planet.osm. These dumps exist in two formats, one using XML and one using the Protocol Buffer Binary Format (PBF).

OpenStreetMap data and derived tiles were originally published under the Creative Commons Attribution-ShareAlike licence (CC BY-SA) with the intention of promoting free use and redistribution of the data. In September 2012, the licence was changed to the Open Database Licence (ODbL) published by Open Data Commons (ODC) in order to more specifically define its bearing on data rather than representation.[29][30] As part of this relicensing process, some of the map data was removed from the public distribution. This included all data contributed by members that did not agree to the new licensing terms, as well as all subsequent edits to those affected objects. It also included any data contributed based on input data that was not compatible with the new terms. Estimates suggested that over 97% of data would be retained globally, but certain regions would be affected more than others, such as in Australia where 24 to 84% of objects would be retained, depending on the type of object.[31] Ultimately, more than 99% of the data was retained, with Australia and Poland being the countries most severely affected by the change.[32]

All data added to the project needs to have a licence compatible with the Open Database Licence. This can include out-of-copyright information, public domain or other licences. Software used in the production and presentation of OpenStreetMap data may have separate licensing terms.

Map tiles provided by the OpenStreetMap project were licensed under CC-BY-SA-2.0 until 1 August 2020. The ODbL license requires attribution to be attached to maps produced from OpenStreetMap data, but does not require that any particular license be applied to those maps. "©OpenStreetMap Contributors" with link to ODbL copyright page as attribution requirement is used on the site.[33]

Editing with JOSM after a ground survey

Map data is collected by ground survey, personal knowledge, digitizing from imagery, and government data. Ground survey data is collected by volunteers traditionally using tools such as a handheld GPS unit, a notebook, digital camera and voice recorder. Software applications on smartphones (mobile devices) have made it easy for anybody to survey. The data is then entered into the OpenStreetMap database using a number of software tools including JOSM and Merkaator. Mapathon competition events are also held by local OpenStreetMap teams and by non-profit organisations and local governments to map a particular area.

The availability of aerial photography and other data from commercial and government sources has added important sources of data for manual editing and automated imports. Special processes are in place to handle automated imports and avoid legal and technical problems.

Surveying routes with a satellite navigation device

Ground surveys are performed by a mapper, on foot, bicycle, or in a car, motorcycle, or boat. Map data was typically recorded on a GPS unit or on a smart phone with mapping app. Once the data has been collected, it is entered into the database by uploading it onto the project's website together with appropriate attribute data. As collecting and uploading data may be separated from editing objects, contribution to the project is possible without using a GPS unit, e.g. by using Paper mapping.

Some committed contributors adopt the task of mapping whole towns and cities, or organising mapping parties to gather the support of others to complete a map area. A large number of less active users contribute corrections and small additions to the map.

Maxar,[14] Bing,[13] ESRI, and Mapbox are some of the providers of aerial/satellite imagery which are used as a backdrop for map production.

Data from several street-level image platforms are available as map data photo overlays. Bing Streetside 360° image tracks, and the open and crowdsourced Mapillary and KartaView platforms provide generally smartphone and windshield-mounted camera images. Additionally, a Mapillary traffic sign data layer, a product of user-submitted images is also available.[34]

Some government agencies have released official data on appropriate licences. This includes the United States, where works of the federal government are placed under public domain. In the United States, most roads originate from TIGER from the Census Bureau.[35] Geographic names were initially sourced from Geographic Names Information System, and some areas contain water features from the National Hydrography Dataset. In the UK, some Ordnance Survey OpenData is imported. In Canada Natural Resources Canada's CanVec vector data and GeoBase provide landcover and streets.

Globally, OpenStreetMap initially used the prototype global shoreline from NOAA. Due to it being oversimplified and crude, it has been mainly replaced by other government sources or manual tracing.

Out-of-copyright maps can be good sources of information about features that do not change frequently. Copyright periods vary, but in the UK Crown copyright expires after 50 years and hence old Ordnance Survey maps can legally be used. A complete set of UK 1 inch/mile maps from the late 1940s and early 1950s has been collected, scanned, and is available online as a resource for contributors.[36]

Field survey in various parts of the Guagua by a group of mappers. They took notes and photos, and recorded GPS tracks. Shown in the photo is the Betis group standing beside one of the Death March trail monuments.

The project has a geographically diverse user-base, due to emphasis of local knowledge and "on-the-ground" situation in the process of data collection.[37] Many early contributors were cyclists who survey with and for bicyclists, charting cycleroutes and navigable trails.[38] Others are GIS professionals who contribute data with an extension for ArcGIS.[39] Contributors are predominately men, with only 3–5% being women.[40]

By August 2008, shortly after the second The State of the Map conference was held, there were over 50,000 registered contributors; by March 2009, there were 100,000 and by the end of 2009 the figure was nearly 200,000. In April 2012, OpenStreetMap cleared 600,000 registered contributors.[41] On 6 January 2013, OpenStreetMap reached one million registered users.[42] Around 30% of users have contributed at least one point to the OpenStreetMap database.[43][44]

As per a study conducted in 2011, only 38% of members carried out at least one edit and only 5% of members created more than 1000 nodes. Most members are in Europe (72%).[45] According to another study, when a competing maps platform is launched, OSM attracts fewer new contributors and pre-existing contributors increase their level of contribution possibly driven by their ideological attachment to the platform. Overall, there is a negative effect on the quantum of contributions.[46]

Some companies freely license satellite/aerial/street imagery sources from which OpenStreetMap contributors trace roads and features, while other companies make data available for importing map data. Automotive Navigation Data (AND) provided a complete road data set for Netherlands and trunk roads data for China and India. In June 2018, the Microsoft Bing team announced contribution of 125 million U.S. building footprints to the project – four times the number contributed by users and government data imports.[47] Amazon uses OpenStreetMap for navigation and has a team which revises the map based on GPS traces and feedback from its drivers.[48] As of February 2021, Apple was the most prolific corporate editor, responsible for 80% of edits to existing roads.[48]

According to a study, nearly 17% of all edits to the map came from corporate teams during 2019–2020. The top 13 corporate contributors during 2014–2020 include Apple, Kaart, Amazon, Facebook, Mapbox, Digital Egypt, Grab, Microsoft, Telenav, Developmentseed, Uber, Lightcyphers and Lyft.[48] There was some vandalism on some occasions attributed to corporate editors.[49][50]

Humanitarian OpenStreetMap Team (HOT) is a nonprofit organisation promoting community mapping across the world. It developed the open source HOT Tasking Manager for collaboration, and contributed to mapping efforts after the April 2015 Nepal earthquake, the 2016 Kumamoto earthquakes, and the 2016 Ecuador earthquake. The Missing Maps Project, founded by the American Red Cross, Doctors Without Borders, and other NGOs, uses HOT Tasking Manager. The University of Heidelberg hosts the Disastermappers Project for training university students in mapping for humanitarian purposes. When Ebola broke out in 2014, the volunteers mapped 100,000 buildings and hundreds of miles of roads in Guinea in just five days.[51]

OSM application architectural components

OpenStreetMap applications utilize multiple components to provide services. The map data is rendered using pre-generated tiles for various levels of zoom. Editing applications typically support display of imagery, and field mapping data in the form of GPS traces and voice, photo, video annotations to aid in editing map. JOSM, ID, StreetComplete, Rapid, Potlatch are the top 5 editing tools for contributions during 2018-2023 according to a study by Heigit.[52]

OpenStreetMap of Soho, central London, shown in "standard" OpenStreetMap layer

Raw OpenStreetMap data of India loading in QGIS for analysis and mapmaking

The web site provides a slippy map interface based on the Leaflet JavaScript library (and formerly built on OpenLayers), displaying map tiles rendered by the Mapnik rendering engine[53] The basic map views offered are Standard, Cycle map, Transport map and Humanitarian. In February 2015, OpenStreetMap added route planning functionality to the map on its official website.[54][55]

iD[56] is used for editing on OSM website. It was originally developed by Mapbox, with initial financing from the Knight Foundation and is available under open source.[57] It uses "Ruby on Rails" to enable users to edit maps and view changelogs. The application interfaces with OSM PostgreSQL database for storage of user data and edit metadata. The default map is rendered by Mapnik, stored in PostGIS, and powered by an Apache module called mod_tile. The map data can be edited utilising the satellite/aerial imagery, GPS traces, and local knowledge.

Nominatim[58] (from the Latin, 'by name') is a tool to search OpenStreetMap data by name and address (geocoding) and then to generate synthetic addresses of OpenStreetMap points (reverse geocoding). Data from Nominatim and GeoNames are used for the search functionality in OSM website.[59]

"RapiD" is a web based editor derived from iD. It is developed and used by Facebook for "map with ai" project to add artificial intelligence (AI) detected maps of roads to OSM.[60] JOSM, Potlatch,[61] and Merkaartor[62] are more powerful desktop editing applications that are better suited for advanced users.

StreetComplete asking user a question. User filled in the answer. After tapping "OK" this answer will be added to an OpenStreetMap database.

Vespucci[63] is the primary full-featured editor for Android; it has been regularly released since 2009. StreetComplete, an Android app launched in 2016,[64] allowed users without any OpenStreetMap knowledge to answer simple questions for existing data in OpenStreetMap, and thus contribute data.

As OSM is a crowd sourced project with complex tagging scheme, there is potential for introduction of unintentional errors and intentional errors. Contributors use history menu on the OSM website, tools like OSMcha, OSM Inspector and Osmose to monitor, review and fix errors.

OpenStreetMap data has been favourably compared with proprietary datasources,[65] although as of 2009 data quality varied across the world.[66][67] A study in 2011 compared OSM data with TomTom for Germany. For car navigation TomTom has 9% more information, while for the entire street network, OSM has 27% more information.[68]

OpenStreetMap Philippines GPS map, an end-product of over a thousand crisis mappers that contributed almost 5 million map updates during the 2013 Haiyan humanitarian activation[69]

The 2010 Haiti earthquake established a model for non-governmental organisations (NGOs) to collaborate with international organisations. OpenStreetMap and Crisis Commons volunteers used available satellite imagery to map the roads, buildings and refugee camps of Port-au-Prince in just two days, building "the most complete digital map of Haiti's roads".[70][71] The resulting data and maps have been used by several organisations providing relief aid, such as the World Bank, the European Commission Joint Research Centre, the Office for the Coordination of Humanitarian Affairs, UNOSAT and others.[72]

After Haiti, the OpenStreetMap community continued mapping to support humanitarian organisations for various crises and disasters. After the Northern Mali conflict (January 2013), Typhoon Haiyan[73][74] in the Philippines (November 2013), and the Ebola virus epidemic in West Africa (March 2014), the OpenStreetMap community in association with the NGO Humanitarian OpenStreetMap Team (HOT) has shown it can play a significant role in supporting humanitarian organisations.[51]

Several open collaborative mapping projects integrate with the OpenStreetMap database or are otherwise affiliated with the OpenStreetMap project:

• OpenHistoricalMap is a world historical map based on the OpenStreetMap software platform.[75]
• OpenRailwayMap is a detailed online map of the world's railway infrastructure, built on OpenStreetMap data. It has been available since mid-2013 at openrailwaymap.org.[76]
• OpenSeaMap is a world nautical chart built as a mashup of OpenStreetMap, crowdsourced water depth tracks, and third-party weather and bathymetric data.
• Wheelmap.org is a portal for mapping, browsing, and reviewing wheelchair-accessible places.

Mapbox is one of the earliest company to provide OSM based services. Custom maps can also be generated from OpenStreetMap data through various software including Jawg Maps, Mapnik, Mapbox Studio, Mapzen's Tangrams.[77] OSRM, GraphHopper, MapQuest and Mapbox's Valhalla are some of the route planning application providers.

Since 2007, the OpenStreetMap community has held an annual, international conference called State of the Map (SotM) where all stake holders gather to share progress and discuss issues.[6] There are also various national, regional and continental SotM conferences, such as SotM U.S., SotM Baltics, SotM Asia & SotM Africa.

Foody, Giles; et al. (2017). Mapping and the Citizen Sensor. London: Ubiquity Press. ISBN 978-1-911529-16-3. JSTOR j.ctv3t5qzc.