I’ve a question regarding Granice_obwodow_spisowych and Granice_obrebow_ewidencyjnych … are these valid relation and/or Administration sources?
Also
Granice_jednostek_ewidencyjnych
Granice_obrebow_ewidencyjnych
Granice_rejonow_statystycznych
We probably all prefer relation because of Nominatim but it’s not always that easy.
Since the shot down of the polish EMUiA (places & addresses record) version 1 (somewhere in 2017) the official administrative borders go as far as to the level of gmina (commune/Gemeinde). As far as I know beyond that there are no official administrative borders, so pio2_122 is right saying Obręb geodezyjny (ewidencyjny) is - de iura - not an administrative unit. But, on the other hand…
In many cases Obręb geodezyjny (ewidencyjny) are consistent with the - de facto - borders of places (villages), where one Obręb geodezyjny is one village. That is for example the case in the commune where I live=> they exactly match the pattern of addresses and even with road signs when driving from one place to another. So, they are present in OSM as borders with admin_level=8 even though they are not official, because if something does function as a border, it is a border.
If you take a look on Lower Silesian Voivodeship (a province there 100% of villages have borders) you’ll notice villages are mostly based on Obręb geodezyjny.
But, in the commune neighbouring to mine one Obręb geodezyjny covers two villages and in this case - they are in fact - as pio2_122 writes - useless for OSM.
So no easy & obvious conclusions.
In the case of gmina Chojna I checked every Obręb geodezyjny and the place (village) within them, in the National Register of Geograhical Names (PRNG).
As far as I can see, one place=village matches one Obręb geodezyjny. Other places (hamlets, isolated dwellings) are sub-parts of villages, so in this particular case - in my opinion - Obręb geodezyjny can be used as relations for borders.