[OSM-talk] HOT Community Webinar - Wed July 26th 3:00 PM – 4:00 PM UTC

Christoph Hormann osm at imagico.de
Tue Jul 25 15:28:50 UTC 2017


On Tuesday 25 July 2017, althio wrote:
>
> As I read your message, I failed to see your points.
> I might be missing something obvious, or at least clear for some
> advanced mappers.
> Maybe you could explain in some more details the problems with this
> mapping guide?

I indeed find this obvious.  The presentation does not in any way even 
touch the meaning of the tag landuse=residential in OSM.  It 
essentially despises some of the core principles of OSM, in particular 
verifiability and on-the-ground rules and it specifically instructs the 
reader to enter a non-verifiable geometry with a non-verifiable tag in 
complete disregard of the actual situation on the ground.

> What is so shocking in these mapping instructions [1], "in light of
> the projects in question"?

If i assume the project has a true humanitarian goal, i.e. to improve 
the living conditions of those living in these areas in a sustained way 
and if i further assume that accurate map data representing the actual 
situation on the ground is useful for this goal, mechanically drawing 
wrapper polygons around clusters of buildings and entering this into 
the OSM database where it is further manually maintained and managed is 
a waste of human ressources.  The task can easily and in much better 
quality be performed by a computer program on the fly for whatever 
purpose the data is being used for and doing this would free mapping 
capacity to do more actually verifiable mapping work where it is 
needed.

> What is so shocking in these mapping instructions [1] "because of
> what is presented [...] as good practice"?

Because this practice of drawing non-verifiable wrapper polygons is in 
blatant violation of OSMs core principles.

Not that it really matters but in several of the sample images shown you 
can actually quite reliably identify non-building structures that could 
be mapped (like pens for lifestock).  So the takeaway for the observant 
reader is to map abstract geometries with no visible basis on the 
ground but specifically ignore what you can actually see in the images.

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



More information about the talk mailing list