[Tagging] The endless debate about "landcover" as a top-level tag

Marc Gemis marc.gemis at gmail.com
Wed Jun 13 07:23:01 UTC 2018


On Wed, Jun 13, 2018 at 9:15 AM Warin <61sundowner at gmail.com> wrote:
>> won't work, see e.g.
>> https://www.openstreetmap.org/search?query=waasmunster#map=16/51.1215/4.0932&layers=N
>> that's not a forest, that are a lot of private gardens with trees in it.
>
> Exclude area with landuse=residential ??


but what if you want to map the presence of trees. I would do that
with landcover=trees.
But those using landuse=forest will have to overlap it on
landuse=residential. As I see it, landuse=forest on top another
landuse still means the "other" landuse but with some trees on it.
landuse=forest not overlapping any other landuse means "forest", and
e.g. a small landuse=retail overlapping on landuse=residential means
retail (at least that is how carto-css present things now).

>> >
>> >
>> > No, you cant. As there are conflicting tagging methods
>>
>> If everything was "properly" mapped with those 3 tags I could come up
>> with an algorithm. Not with the current mess of course.
>
> Proper? Who says what is proper?

Proper for me means clearly separate landuse from landcover, so that
one can see the use of the land and how it is covered from different
tags.
Not proper is e.g. one mapper using landuse=forest to indicate an area
for timber production and another mapper to map trees in a private
residential garden.
The latter mapping is fine if you just want to colour a map. :-)

m



More information about the Tagging mailing list