Mediaviewer selects the static part of the description as caption when the changing part is the relevant. Problem exists with all languages and all WLM photos.
Example:
Current result for finnish :
"Tämä on kuva suomalaisesta monumentista, jonka tunniste on"
Expected result for finnish:
"Katedralskolanin liikuntasalin peruskorjauksessa löytyneet rauniot"
So basically question is how to say in template (like in https://commons.wikimedia.org/wiki/Template:Lang ) that this text is not primary text and show me only if there is no better text.
MediaViewer uses data from CommonsMetadata, which is on the verge of being deprecated in favor of StructuredDataOnCommons. So probably not worth putting effort into fixing non-critical bugs now.
Testing today, I obtain the following caption: "Tämä on kuva suomalaisesta monumentista, jonka tunniste on 'Turun tuomiokirkko ja Turun historiallinen ydinalue' (Q29974183) - RKY: 4620"
@Zache, this is the expected behaviour? It seems like the issue has been solved in the meanwhile?
Best and wanted caption would be. still "Katedralskolanin liikuntasalin peruskorjauksessa löytyneet rauniot" .