From OpenStreetMap Wiki
Jump to: navigation, search

Bigger islands which consist of several coastline segments

I've just tried tagging like this on an island which is encompassed with several coastline segments, and since it is categorised as an area tag JOSM complains. Should a single point rather be used in such cases or is JOSM wrong to complain?--Guttorm Flatabø (talk) 16:22, 17 July 2013 (UTC)

I think JOSM is wrong to complain. I think the problem is that no strict 'area' data type exists in osm. Some back ground in this video: --Hawkeyes (talk) 20:09, 8 October 2014 (UTC)

Multipolygon and coastline tagging

Can anyone more experienced correct the following for group of islands?

"Mf Relation.svg add the natural=coastline into a multipolygon."

This does not seem to be true and contradicts with what is written under "Do not tag nodes or relations with natural=coastline."

Thanks --Frank jr (talk) 14:57, 14 May 2015 (UTC)

Island vs. Islet

The criteria listed on the main page are:

  • "with max. area of 1 km²" - this could be easily calculated by the renderer, but it cannot be that easily estimated by a mapper. I just came across an island with about 1 km² and wondered which tag to use. This criterion does not make any sense to me.
  • "only a few buildings (or none)" - what is "only a few"? 5? 10? Can't the number of buildings be determined by the renderer, again? And is it really intended that big islands within the polar circle or with military use - in which cases they might be unhabited but nevertheless important - shall be tagged as islets?
  • "located near by a bigger island or to the onshore" - just as above, this can be automatically determined.
  • "if an island is in the middle of the sea and far away of other islands, cause the prominent meaning..." - The word "prominent" tells it all. It's just the same as with peaks, villages etc. The calculation of prominence is a rendering task, not a tagging task! Otherwise we would need to invent tags like "peaklet", "houselet", "meadowlet", and maybe the other way too using an augmentative suffix like the Italian "-one".

Therefore, the distinction island/islet is bogus and should be dropped IMO. --Fkv (talk) 05:36, 2 March 2016 (UTC)

Does it hurt? Apparently people feel like using it. RicoZ (talk) 13:36, 13 March 2016 (UTC)

Why natural=coastline for an island?

An island (or islet) boundary represents exactly a coastline.

natural=coastline is useful (and required) for really large territories and with several administrative divisions that demands many breaks on the land border with the ocean. And by its own definition natural=coastline is only applicable to ways and not for areas (closed ways) so the use of this tag should not be proposed as a complement for islands.

An island is semantically complete in itself, so asking mappers to include the tag natural=coastline just creates a redundancy and adds an useless complexity.