On Thu, Sep 21, 2017 at 9:22 PM, Martin Koppenhoefer <dieterdre...@gmail.com > wrote:
> This is one of the fields (fundamental to OSM), where wikidata is just a > mess: distinction of geographically localized communities and > administrative territorial entities. > > Just a few examples: > https://www.wikidata.org/wiki/Q123705 neighbourhood is a subclass of > "human settlement" and "community". So far so good, but then it is also > "part of municipality"? > https://www.wikidata.org/wiki/Q2983893 quarter is a subclass of > neightbourhood and administrative territorial entity. And it is an instance > of "designation for an administrative territorial entity". > https://www.wikidata.org/wiki/Q486972 human settlement looks OK at a > glance > https://www.wikidata.org/wiki/Q515 city is a subclass of "human > settlement", "administrative territorial entity" and "political territorial > entity" (are these AND or OR?). > https://www.wikidata.org/wiki/Q3957 town is a subclass of "human > settlement". It is "part of a country". It was also a subclass of political > territorial entity until today [1] > https://www.wikidata.org/wiki/Q532 village is a subclass of "rural > settlement" and part of "rural area". > https://www.wikidata.org/wiki/Q14788575 rural settlement is an instance > of "designation for an administrative territorial entity" and a subclass of > "human settlement" > etc. > > Take the town example: it has been for some years a subclass of political > territorial entity and isn't anymore since today. There are tens of > thousands of objects that are all instances of towns according to wikidata. > With one edit all of them have lost their "political territorial entity" > status. > I wouldn't worry too much about these very generic classes of human settlements or administrative areas. It would be better to focus our attention on the actual Wikidata items on settlements and administrative areas of each country, like Waldhufendorfs in Germany ( https://www.wikidata.org/wiki/Q351190) or comunes in Italy ( https://www.wikidata.org/wiki/Q747074). These national subclasses should then be classified as subclasses of whichever appropriate generic settlements or administrative areas there are. So I doubt that tens of thousands of towns in Wikidata suddenly lost their political territorial entity status. I would think that some of them are still classified as such because of a different path up the ontology tree.
_______________________________________________ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk
