In Japan, as in many other places, their are religious temples or shrines located on larger sites that may contain many buildings. Or to put it another way, the precincts of a temple may contain the main temple and many sub-temples, each with their own names and supporting buildings and gardens.
After searching both the Japanese wiki and the main one, and these forums I was unable to find any consensus on how to deal with these. So, I went back to study the wiki.
It looks like creating an area with landuse=religious would give me a precinct that I can name and then naming the main buildings of the sub-temples to the extent that I know them, and then creating a multipolygon relationship between all of the buildings located on the land (inner) and the surrounding area (outer) would maybe accomplish what I want. Not letting anything overlap.
But I’m new, and while it appears to show correctly on the osm, I don’t want to be screwing something else up. Anyone have any opinions pro or con, or suggestions of further reading for me?
I am also curious, the area defaults to layer 0, so should I raise all of the buildings and other features inside the area to layer 1 or higher as needed? I don’t seem to need to do that to get things to show correctly on osm on my computer, but would it help or hinder other rendering machines (for Garmin and Android, oh, just for example )
and that is what I am attempting to do. The buildings, paths, gardens, etc. are all contained in the precinct. There is often a physical barrier around it, with gates that control access according to opening and closing times. These are real world relationships and many of them have legal standing.
Since I made the original post I have found more discussions concerning “landuse=farmland” which all seem pertinent and have given me various practical ideas. However, there isn’t much discussion in relationship to religious uses, so no consensus that I have been able to find, yet. AND I don’t want to be working contrary to more experienced individuals’ and developers’ assumptions. Thus, my questions.
The easiest example would be a building with a courtyard.
Another example is where there are several related buildings, but it not practicable to survey the land that contains them (e.g. school buildings where the exact extent of the campus is unknown.
Yet another example is where an administrative region is composed of subordinate ones, or has sufficiently complex geometry that it exceeds the maximum number of nodes in a way.