"place nodes close" does not take higher level admin borders into account?

[hartmut-mariadb]

Getting a warning here about two place nodes named "Bruchmühlen" being close to each other.

https://nominatim.org/qa/#map=12.95/52.21/8.43&layer=place_nodes_close

This is a false positive as the village is (for reasons I never really fully understood) split by a state border, the southern and western part belonging to Lower Sachsony, and the north-east part to North Rhine Westfalia

So the two are indeed two separate administrative entities known by the same name, but having different ZIP codes (32289 vs. 49328), are in different states and districts, and have different car license plate letters ("HF" vs. "OS")

[lonvia]

For this particular case one might argue that you should be setting only one pace node. While there two separate administrative entities (represented by two distinct boundary=administrative), there is only one human settlement (to be represented by one place node). Even Wikipedia says "Kulturell und städtebaulich bilden das niedersächsische und nordrhein-westfälische Bruchmühlen eine Einheit. "

That said I have seen a couple of other cases where your suggestion has merit. For example the villages in Glarus all come with their own "Allmend" (e.g. https://www.openstreetmap.org/node/7943278583).

We might simply add another check that if the two place nodes also have the same parent.