On Wed, May 7, 2025 at 10:16 AM Illia Marchenko <illiamarchenk...@gmail.com> wrote:
> Strictly speaking, "land use" and "named places" is orthogonal entities. > Don't mix them. Just use node with place=* and name=* and area with > landuse=* but without any name=*. > You're never going to get people to stop using "landuse=residential; name=x". In fact, that would be a horrible move. People want to know where the borders are when known, not just an arbitrary location which could either be the centroid or just the "heart" of the area. Not to mention this is the same treatment we use elsewhere: when you want to label a shopping center, we use `landuse=retail; name=Shopping Center Name`. In the case of a named retail landuse, there is no designated `place=shopping_center; name=foo`. But rather than finding an example that substantiates my proposal (which I can think of many), in what other applications of OSM do we create duplicate ways for "orthogonal concepts" that are semantically the same in the real world? Not to grant your foundation though I don't get why you would say they're "orthogonal entities." I'd like to see that explained. I don't see it: a neighborhood is a subclass of residential landuse that has a name. This is documented and in alignment with the wiki and common use. OSM gets this right. Given the following, * "this is residential landuse which *may* have a name" * "this is a subclass of residential landuse that *must* have a name." They seem *NOT* to be orthogonal entities to me. This seems to be a unique case in OSM. -- Evan Carroll - m...@evancarroll.com System Lord of the Internets web: http://www.evancarroll.com ph: 281.901.0011 <+1-281-901-0011>
_______________________________________________ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
