Yes the node approach is only done when the boundaries are unknown (and so this is an approximation) but in this case, as we have the boundaries as open data, we could add them indeed. You can see it in the wiki page too <https://wiki.openstreetmap.org/wiki/Key:place>. Thus, creating a polygon using the exact boundaries of the open data with the tag "place=neighbourhood" + "name=*" should be fine. You can just import a shapefile or similar format in JOSM for example, and tag it correctly.
Le lun. 19 août 2019 à 10:35, Pieter Colpaert <pie...@openknowledge.be> a écrit : > Hi all, > > Neighbourhoods in Ghent are mapped as a point. See for example > https://www.openstreetmap.org/node/2274965455 > > The problem with this is that nominatim will add the neighbourhood to an > address string, based on the nearest point. An address still in the > Brugse Poort therefore will get an obvious wrong neighbourhood attached > to it. See for example > > https://nominatim.openstreetmap.org/search.php?q=appelstraat+27%2C+9000+Gent&polygon_geojson=1&viewbox=. > > This point is in the Brugse Poort, but it is closest to the point of the > neighbourhood Malem. > > Should we change these point to be polygons instead? > > There is also an official Open Data list of “wijken” (which is not > entirely the same though) on the open data portal of the city of Ghent: > https://datatank.stad.gent/4/grondgebied/wijken → Should we map these > bounds in OSM as well? And how? > > Kind regards, > > Pieter > > _______________________________________________ > Talk-be mailing list > Talk-be@openstreetmap.org > https://lists.openstreetmap.org/listinfo/talk-be >
_______________________________________________ Talk-be mailing list Talk-be@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk-be
