Add depicts to Commons uploaded media #43

GA-Kevin-Codes commented 2026年04月17日 15:58:00 +02:00

Wikimedia Commons strongly encourages the use of depicts in structured data when possible. This is especially true when working with images of the real-world, such as those uploaded via survey on Map2Go. A depicts field should be available to be added with a Wikidata search and suggestions based on OpenStreetMap tags already present on the element. Once chosen, depicts and the OSM ID (once created, if a new element) should be added to the Commons structured data automatically.

Wikimedia Commons strongly encourages the use of `depicts` in structured data when possible. This is especially true when working with images of the real-world, such as those uploaded via survey on Map2Go. A `depicts` field should be available to be added with a Wikidata search and suggestions based on OpenStreetMap tags already present on the element. Once chosen, `depicts` and the OSM ID (once created, if a new element) should be added to the Commons structured data automatically.

Shisma commented 2026年04月18日 12:07:49 +02:00

Here is how you could automatically suggest depicts statements:

Lets say the user takes a photograph of this node
www.openstreetmap.org/node/4803636899

it has the following tags

key	value
amenity	bench
backrest	yes
check_date	2023年06月28日
material	wood

Let's just request an item of each of these things from Wikidata:

SELECT ?thing ?thingLabel WHERE {
 SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],mul,en". }
 {
 ?thing wdt:P1282 "amenity=bench".
 } UNION {
 ?thing wdt:P1282 "backrest=yes".
 } UNION {
 ?thing wdt:P1282 "check_date=2023年06月28日"
 } UNION {
 ?thing wdt:P1282 "material=wood"
 }
}

Result

thing	thingLabel
wd:Q204776	bench
wd:Q287	wood
wd:Q3037336	backrest

You can safely assume that all those things are depicted on the photograph the user is about to submit.

Perhaps the user should be able to select one prominent value and also dismiss false positives.

If the item in question actually has a wikidata tag, then the referenced item should be preselected as prominent.

Here is how you could automatically suggest `depicts` statements: Lets say the user takes a photograph of this node [www.openstreetmap.org/node/4803636899](https://www.openstreetmap.org/node/4803636899) it has the following tags key | value ----|----- amenity | bench backrest | yes check_date | 2023年06月28日 material | wood Let's just [request an item of each of these things](https://w.wiki/LYqi) from Wikidata: ```sparql SELECT ?thing ?thingLabel WHERE { SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],mul,en". } { ?thing wdt:P1282 "amenity=bench". } UNION { ?thing wdt:P1282 "backrest=yes". } UNION { ?thing wdt:P1282 "check_date=2023年06月28日" } UNION { ?thing wdt:P1282 "material=wood" } } ``` Result thing | thingLabel ------|----------- [wd:Q204776](http://www.wikidata.org/entity/Q204776) | bench [wd:Q287](http://www.wikidata.org/entity/Q287) | wood [wd:Q3037336](http://www.wikidata.org/entity/Q3037336) | backrest You can safely assume that all those things are depicted on the photograph the user is about to submit. Perhaps the user should be able to select one [_prominent_](https://commons.wikimedia.org/wiki/Commons:Depicts#Prominence) value and also dismiss false positives. If the item in question actually has a `wikidata` tag, then the referenced item should be preselected as _prominent_.

Shisma commented 2026年05月23日 18:22:48 +02:00

I love how it turned out. thank you ❤️🙏

I love how it turned out. thank you ❤️🙏