Key:source

From OpenStreetMap Wiki
Jump to navigation Jump to search

Split (260036) - The Noun Project.svg

This key differs significantly between OpenStreetMap and OpenHistoricalMap. For definitions and usage instructions in OpenHistoricalMap, see OpenHistoricalMap/Tags/Key/source.
Public-images-osm logo.svg source
Description
For indicating the source of all tags added to the OSM database. Show/edit corresponding data item.
Group: annotations
Used on these elements
may be used on nodesmay be used on waysmay be used on areas (and multipolygon relations)may be used on relationsmay be used on changesets
Documented values: 32
See also
Status: de facto

The source=* tag can be used to indicate the source of some information (i.e. meta data) added to OpenStreetMap. This tag could be useful for verification and to assist future editors who may wish to understand the origins of the data they are modifying.

With modern editors, the source=* tag is typically added to the changeset when a change is made, not as a tag on a piece of information which may be updated at different times based on different sources. However it seems the historic practice of tagging objects or individual attributes has not been officially deprecated yet, so don't go around deleting those source tags indiscriminately.

If possible, try to split modifications based on the source being used to properly attribute data, and allow easier validation. If that's not possible, just use the standard OSM multivalue convention of separating them with a ";" (semicolon) e.g. source=survey;https://www.mapillary.com/map/im/... (when combining information from your own surveying and a mapillary image).

How to use on changesets

When uploading/saving your edits to the public osm database, your uploads are always marked as a changeset with your username, the time of change etc. To this changeset you should add an explanatory comment and also a source=* tag how the details were captured. When you update or change an existing object (node, way, area, relation etc.), just specify the source of your changes, since the previous changesets and their sources can still be seen in the history of the object. But do mention a source even if it is the same as in a previous edit, because this is not implied.

Historic usage on objects and attributes

When this page was originally created, it was normal practice to put the source=* tag on the objects being edited rather than on a changeset. This section describes the specifics of that usage. As usual don't start deleting those tags indiscriminately unless and until there is a general project decision to do so.

The most common way to do this old-style source tagging is/was to simply add a source=* tag on the object being edited, such as a building or a road.

It is often preferable to use separate source tags for each of the main tags, for example source:name=*, source:ref=*, source:addr=* detailing how the information for name, ref and address information was collected.

In the Potlatch editor a source tag can be added to a feature by pressing 'b'.

How to use (historic)

Add a source tag detailing how the details were captured. Please also update the source information if you check or update a tag value based on a new survey. For example, a feature may have originally been captured from out-of-copyright imagery, being tagged 'source=NPE'. If you then later improve the alignment using Bing imagery then change the source value to 'source=Bing'. If a tag value is based on information derived from multiple sources use 'source=NPE;Bing' or similar.

Single source tag (historic)

A single source tag is used when all tags associated with a feature have the same source. For example:

  1. source=survey - to indicate that all tags were added based on a survey
  2. source=Bing - to indicate that all tags were added based on Bing Imagery.

Multiple source tags (historic)

If a feature has multiple tags you may want to make use of the source namespace to indicate precisely which tag your source refers to. For example:

  1. source:geometry=Bing to indicate that the geometry (shape and position) of an object was based on Bing Imagery.
  2. source:name=survey to indicate that the name was added based on a survey.
  3. source:name=OS Locator to indicate that the name added from OS Opendata Locator.
  4. source:name=massgis to indicate that the name was added from a Massachusetts open data source
  5. source:addr=massgis to indicate that the address was added from a Massachusetts open data source

Other tags exist or can be established within the source namespace.

A note about maxspeed (historic)

Note that source:maxspeed=* is frequently used to give information about the type of the speed limit. If you wish to tag both the type of the speed limit and also source information for how the information was collected then either use source:maxspeed=* in the way described on this page and use maxspeed:type=* for the type of speed limit (for example maxspeed=40, source:maxspeed=survey source:maxspeed:date=2012-11-20 and maxspeed:type=sign would indicate that a numeric sign showed a speed limit of 40 km/h based on a ground survey done on 20 November 2012, or use source:maxspeed=*. Please note that currently most usages of source:maxspeed=* are indicating the type of maxspeed, and that by the time this tag was introduced there were 0 occurences of sources like documented here!

OpenHistoricalMap

Main article: OpenHistoricalMap/Good practice#Verifiability

Unlike OpenStreetMap, OpenHistoricalMap continues to tag source=* on individual features. In fact, you are generally expected to add source=* to each individual feature to avoid the appearance of plagiarism. It is not necessary to cite a built-in background imagery layer that you trace the geometry from, since that is already recorded in imagery_used=* on the changeset, but you should use source=* to provide a citation regarding start_date=*, end_date=*, and other attributes.

Specific values

General sources commonly used by human mappers

Value Country(ies) Meaning Website Note
survey worldwide The OSM user was there and saw it firsthand. The user may or may not have done any precision measurement of the location etc. using a device such as a GPS receiver.
extrapolation worldwide The information was extrapolated from other information in OpenStreetMap, but is still better than pure guesswork
GPS worldwide Someone measured the location using a GPS, GLONASS, Galileo, ATLAS or other satellite navigation receiver but did not copy the information from a map that might be installed in that receiver. Either the OSM user editing the object did this (and may or may not have uploaded a GPS track), or the OSM user downloaded one or more GPS track from the OpenStreetMap database (doing both is also possible). (This was the default value in ancient versions of the JOSM editor)
mapillary worldwide Information was photographed and can be checked by everyone using a plugin for Mapillary photos. Mapillary is a service for sharing geotagged photos (open data alternative of Google Streetview). It is also possible to be more specific by pointing to a specific photo id (see mapillary=*)
image worldwide Information was photographed / video filmed (not from the air or space) Indicate from which image it's from, or survey if you did take the pictures.
aerial imagery worldwide General tracing of aerial imagery Indicate from which imagery it's from, to make that verifiable and follow most sources' requirement of attribution.
common knowledge worldwide The OSM user knows this from many informal source and believes it to be commonly known by most people who care. For instance most people know that the huge ocean between Europe, Africa and the Americas is called the Atlantic. Prefer verifiable sources such as survey
local knowledge worldwide The OSM user has some affiliation with the area (resident, place of work, frequent visits...) and knows this first hand or from others in the area Prefer verifiable sources such as survey

Specific sources commonly used by human mappers

Value Country(ies) Meaning Website Note
Bing worldwide Tracing of Microsoft Bing aerial imagery; image capture date not specified. Commonly written in lower case (bing) for consistency with most other single-word tags.
Bing 2011 worldwide Tracing of Microsoft Bing aerial imagery; image capture date sometime in 2011.
Bing 2012-10--2013-05 worldwide Tracing of Microsoft Bing aerial imagery; image capture date between October 2012 and May 2013 (for example). Use ISO 8601 notation for dates and periods.
cuzk:km CZ Manual or semi-automatic tracing of Czech cadastral map http://www.cuzk.cz
digitalglobe worldwide Traced DigitalGlobe / MapBox aerial imagery. some use Mapbox as source, as suggested by MapBox themselves
Fugro 2005 DK Tracing of Fugro's aerial imagery from 2005 http://www.fugro.com/divisions/survey/geospatial-services/aerial-mapping
landsat worldwide Tracing of Landsat low resolution satellite imagery
uhul:ortofoto CZ Usually manual tracing of grey-scale aerial imagery by Czech office for economic use of forests
yahoo worldwide Tracing of Yahoo's aerial imagery
AGIV BE Tracing of quality images from Flemish governmental organisation AGIV ( "Agentschap voor geografische informatie Vlaanderen" ) http://www.agiv.be Open data license
AGIV crab BE Addressing data comes from CRAB database by AGIV agency http://www.agiv.be Open data license
GRB BE Building information (geometry, position) comes from GRB database, also possibly addressing data when included directly from this source. Alternatively addressing data can be sourced from CRAB database. (they are not equal at the moment) http://www.agiv.be/producten/grb Open data license (2016)
IGVSB VE Information obtained from the Instituto Geográfico de Venezuela Simón Bolívar, the government organization that provides official cartography in Venezuela. http://www.igvsb.gob.ve/
Schadow1 Expeditions PH Map data obtained from Schadow1 Expeditions, a mapping advocacy resource for the Philippines httpsː//www.s1expeditions.com Explicit permission

Tags used only by specific data import software

exclamation mark

Imports and automated edits should only be carried out by those with experience and understanding of the way the OpenStreetMap community creates maps, and only with careful planning and consultation with the local community.
See Import/Guidelines and Automated Edits code of conduct for more information. Imports/automated edits which do not follow these guidelines might be reverted!
Value Country(ies) Meaning Website Note
3dShapes NL 3dShapes landuse and building shapes in The Netherlands and a bit across the border
AND NL Data donated in 2007 by AND Automotive Navigation Data, see the AND Data wiki page.
artem;bing RU See the wiki page.
BAG NL Address and building data from NL
cadastre-dgi-fr FR WikiProject France/Cadastre French catastre maps. Mostly buildings, road names, administrative boundaries, addresses
cuzk:ruian CZ Import source of addresses, buildings, parcels of land in the Czech Republic http://www.ruian.cz
dibavod CZ Import source of water bodies in the Czech Republic http://www.dibavod.cz
DOP_LVA DE source=DOP_LVA_BY_01 - (pilot scheme from 2008-12-12 to 2009-03-31) extending OSM by digital orthophotos provided by the land surveying office of Bavaria, Germany
Geobase_Import CA Geobase data. Mostly place names
HELP␣SERVICE␣-␣REMOTE␣SENSING... CZ Import source of some parts of road network in the Czech Republic http://www.bnhelp.cz
IBGE BR IBGE Import. Brazilian federal data
kapor2 SK Import of ways and buildings from cadastre data in Slovakia https://www.katasterportal.sk
KSJ2 JP Japan KSJ2 Import National-Land Numerical Information based on JPGIS. Can no longer use the data since 2015.
LSIB3 Worldwide International boundary from the Office of the Geographer and Global Issues (INR/GGI) http://rockburger.com/openstreetmap/state
microsoft/BuildingFootprints Worldwide Building shapes added from RapiD or the MapWithAI JOSM plugin
NRCan-CanVec-7.0
NHD US National Hydrography Dataset. Imported rivers/lakes
NPS US US National Park Service Data.
OSAK DK Imported address data from official Danish source
PGS Worldwide Prototype Global Shoreline. Low detail coastline generated from landsat imagery.
SanGIS␣Addresses␣Public␣Domain San Diego U.S. SanGIS government data
YahooJapan/ALPSMAP JP Information donated by Yahoo in Japan

Usage statistics

For human mappers

source value instances (change direction MM-YYYY of count) meaning short source meaning contributor
Bing (and "bing") >16.7 million (increase Jun 2016) Bing Corporate (Microsoft) Ceyockey
cuzk:km >2.3 million (decrease Jun 2016) Manual or semi-automatic tracing of Czech catastral map Government (Czech Republic) Ceyockey (see Tag:source=cuzk:km)
survey >1.7 million (increase Jun 2016) See Key:source OSM editor Ceyockey

For import software

source value instances (change direction MM-YYYY of count) meaning short source meaning contributor
BAG >18.8 million (Green up.png Mar 2017) see BAGimport Government (Netherlands) value assigned at Import/Catalogue#One-Time Imports
cadastre-dgi-fr source : Direction Générale des Impôts - Cadastre. Mise à jour : 2010 >12.1 million (Red down.png Jun 2016) see FR:WikiProject France/Cadastre Government (France) Ceyockey
cadastre-dgi-fr source : Direction Générale des Impôts - Cadastre. Mise à jour : 2011 >9.3 million (Red down.png Jun 2016) see FR:WikiProject France/Cadastre Government (France) Ceyockey
cadastre-dgi-fr source : Direction Générale des Impôts - Cadastre ; mise à jour : 2012 >8 million (Red down.png Jun 2016) see FR:WikiProject France/Cadastre Government (France) Ceyockey
YahooJapan/ALPSMAP >5.1 million (Red down.png Jun 2016) see YahooJapanALPS Data Corporate (Yahoo Japan) value assigned at YahooJapanALPS Data/Highway
3dShapes >1.7 million (Red down.png Jun 2016) 3dShapes Corporate (Object Vision) value assigned at Import/Catalogue#Ongoing_Imports
PGS & PGS(could be inacurately) >3.3 million (Red down.png Jun 2016) Prototype Global Shoreline Government (United States) Ceyockey
cadastre-dgi-fr source : Direction Générale des Impôts - Cadastre ; (M_or_m)ise à jour : 2009 >3.3 million (Red down.png Jun 2016) see FR:WikiProject France/Cadastre Government (France) Ceyockey
KSJ2 >0.7 million (Red down.png Jun 2016) see Import/Catalogue/Japan KSJ2 Import Government (Japan) value assigned at Import/Catalogue#One-Time Imports
NRCan-CanVec-7.0 >2.6 million (Red down.png Jun 2016) CanVec#CanVec 7 Government (Canada) Ceyockey
NRCan-CanVec-10.0 >5.7 million (Green up.png Jun 2016) Canvec#CanVec 10 Government (Canada) Ceyockey
NHD >3.1 million (Green up.png Jun 2016) National Hydrography Dataset Government (United States) value assigned at Import/Catalogue#One-Time Imports
NRCan-CanVec-8.0 >2.1 million (Equals-sign-blue.png Jun 2016) CanVec#CanVec_8 Government (Canada) Ceyockey
extraction vectorielle v1 cadastre-dgi-fr source : Direction Générale des Impôts - Cadas. Mise à jour : 2010 >1.9 million (Equals-sign-blue.png Jun 2016) see FR:WikiProject France/Cadastre Government (France) Ceyockey
OSAK (2010) >1 million (Red down.png Jun 2016) Address data released by Danish government. Government (Denmark) rasher (personal knowledge)
CanVec 6.0 - NRCan >1.8 million (Red down.png Jun 2016) CanVec#CanVec 6 Government (Canada) Ceyockey
kapor2 >2.2 million (Green up.png Jun 2016) WikiProject Slovakia/Sources#Cadastral map Government (Slovakia) Ceyockey
Regione Emillia Romagna >1.2 million (Green up.png Jun 2016) IT:Emilia Romagna edificato Government (Italy) Ceyockey
cadastre-dgi-fr source : Direction Générale des Impôts - Cadastre ; mise à jour : 2008 >0.7 million (Equals-sign-blue.png Jun 2016) see FR:WikiProject France/Cadastre Government (France) ceyockey


Possible tagging mistakes

If you know places with this tag, verify if it could be tagged with another tag.
Automated edits are strongly discouraged unless you really know what you are doing!
If you know places with this tag, verify if it could be tagged with another tag.
Automated edits are strongly discouraged unless you really know what you are doing!

See also