[Imports] LINZ data import in the Antarctic (was: Hydrographic data import from LINZ)

[Christoph Hormann]

On Sunday 24 October 2021, Kyle Hensel wrote:
> Hi, I’m not sure what you mean by ‘geometrically incompatible
> features’ As we discussed in that changeset, I can’t see this ‘damage
> to the database’ that you’re talking about.

It seems to me that this is the core of the problem. You are unable or 
unwilling to see the damage your import is doing in the area.  I am not 
sure why though.  My guess would be that it is a combination of lack of 
familiarity with how mapping in OSM works, a lack of familiarity with 
the geography of the area (have you even looked at the imagery that is 
readily available in the area - like in [1]?), a lack of knowledge on 
how LINZ produced their data of the area (which is fundamentally 
different from how it is done in New Zealand - hence the two areas 
should not be combined in a single import) - all of that combined with 
the strong desire to bring this data into OSM.

As said in the changeset comment i will not invest time doing 
after-import QA on this.  If you are willing to accept that as is this 
import does not work and requires fundamental revision i will gladly 
provide advice on how to properly integrate data from LINZ in a way 
that supports and encourages pre-existing and future manual mapping 
(w.r.t. in particular choice of features, tagging and conflation).  But 
you would need to be prepared to accept that this would not be an 
import with a bulldozer like you have done so far but would require a 
significant amount of time in manual evaluation of existing mapping and 
LINZ data with the reality observable from images as reference.

And by the way - the old narrative of there being insufficient imagery 
sources available for the Antarctic to perform decent remote mapping in 
OSM (as an argument why imports with a bulldozer should be accepted) 
does not apply these days any more.  For much of the Antarctic the 
imagery situation in actually better than for many areas at lower 
latitude.  Like elsewhere there are frequently issues with unclear 
license terms for available data sources of course.  But the main issue 
these days is motivating mappers to familiarize themseleves with the 
region and to do diligent mapping there and not access to information 
for mapping.  Hence this kind of import (which completely discourages 
any kind of manual mapping work) is highly counterproductive for OSM.

[1] 
http://maps.imagico.de/#map=10/-77.671/162.751&lang=en&l=sat&r=osmim&o=3&ui=0

-- 
Christoph Hormann
https://www.imagico.de/