Hello,
I have a problem with filtering. More specifically, with Polish characters. When we search without Polish characters (ą, ę, ń, ć, ź, etc.), different results are displayed than without them (e.g. Gdańsk and Gdansk). My question is is there any solution to standardize this or solve this problem?
Nominatim does understand diacritics and takes them into account. It usually searches for versions with and without diacritics and then reorders the result list taking diacritics again into account. How much weight diacritics should get, is a matter of debate. There certainly is room for improvement.
Your example query “Dworzec PKP Łódź” is causing trouble because of the inconsistent naming of Łódzkie Voivodship. The translations mostly contain the “voivodship” part except for French and Islandic translations, which just contain “Łódź”. This leads to false positive results inside the voivodship for queries that use the diacritics.
Hi and welcome to OSM and the forum.
OSM is primarily a database which can be searched through different interfaces.
Where do you search?