[Tagging] Landcover... not again?....!

[Marc_marc]

Hello,

Le 06.05.22 à 09:36, Peter Elderson a écrit :
> What about the idea of just moving the proposed values into the natural key?

- semantically it makes no sense to use A to describe B... the best key 
to describe a landocover is landcover... except for the "we made a bad 
choice in the past that we don't want to go back to" argument)
- this produces situations like with trees: some people don't want to 
use natural=wood for artificial forests.
I admit that I don't think it's a good idea to use natural=* to describe 
an asphalt cover
same with a short cut grass having another landuse, using 
natural=grassland is a poor proposal compared to landcover=grass
- the key natural=* does not only describe covers. e.g. natural=coastline
I'd like to see a more detailed description of the problem, but I don't 
think it's a good idea to use the key natural=* to describe only covers, 
because it's not the same as natural=*.
In short, create or amplify 3 problems to avoid solving another one

> The main objection against the proposed landcover key was that massive 
> retagging of the enormous installed base is not going to happen, mainly 

if there was a will to do it, it would be 2x1h to put it in automatic
in iD/josm and assimilated and/or make a mecanical edit

> because it accomplishes nothing,

check a render style... datauser already have to deal with a list of 
keys corresponding to landcover. this shows that the absence of this key 
is a nonsense

Regards,
Marc