[Imports] Florida Landuse Import

[Frederik Ramm]

Hi,

On 3/8/22 01:39, Matheus Gomes wrote:
> Regarding the fdep:landuse_code, I have the opinion that this is a valid 
> and important information to be retained on OSM.

I disagree with that. It increases the data volume greatly and leads to 
"WTF" moments for future mappers when you have two adjacent polygons 
sharing a boundary and both are (e.g.) "landuse=forest" in OSM - just 
because they used to have different fdep:landuse_codes originally.

I remember one county on the west coast where I grew quite a few grey 
hairs trying to simplify residential areas, where relations had been 
constructed to create holes in residential areas and these holes were 
then filled with residential areas again, just because the original 
source had different zoning definitions for these residential areas.

It also means that when a mapper modifies an existing area they will 
puzzle over that "fdep:landuse_code" tag ("so I'm extending this forest 
to include a bit of new growth I can see on the aerial imagery... but 
the existing forest as this landuse code, does that apply to the new bit 
as well or am I adding wrong information now claiming that the new bit 
has this landuse code...").

> Future mappers can 
> check the “true” landuse of that polygon, and may do further refinements 
> if needed in the future (maybe some more accurate tags will be invented, 
> who knows). 

By then the polygons will have been modified and a wholesale change will 
likely be wrong.

> The usage of this tag can also serve as a “source” 
> information, given it is directly related to this import.

It will be possible to determine the data source from the v1 changeset 
though - and if the area is modified later according to aerial imagery, 
then what, should mappers then remove the code because FDEP is not the 
(only) source any more?

Bye
Frederik

-- 
Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"