My experience from all the humanitarian responses. Yes, it is important to know
the territory and adapt.
For the North of Mali humanitarian response in early 2013, people had
difficulty to identify villages, flooded by water on the available images with
the rainy season that last 6 months. In desertic regions it is also very
difficult sometimes to spot houses if images are not clear enough.
For me this offers a good challenge, not to delete nodes but to find houses and
add to OSM !
And yes, often, there are no roads in small african villages. Also in some
villages, we cannot identify what is private, what is public, and people walk
in all directions. We need to observe an area, understand how it is organized.
Le dimanche 8 avril 2018 17 h 08 min 47 s HAE, Warin
<61sundow...@gmail.com> a écrit :
In Papua New Guinea there are villages without roads ... people there travel
by foot, plane or boat!
The terrain is such that vehicle roads, even for bicycles, is impractical.
I don't have detailed knowledge of Africa to say if these villages could be
real or not... but I would hesitate to delete them without that knowledge.
Possibly question the original mapper or a mapper active locally or
failing any of that producing results demote them to locality and await a
mapper with on ground experience.
On 09/04/18 06:31, Frederik Ramm wrote:
> On 04/08/2018 10:26 PM, Frederik Ramm wrote:
>> Not only that, someone has already picked them out:
> Looking a bit more at the list, I wonder if we should maybe delete all
> nodes that
> * were imported before 2010 from GNS
> * were never used since
> * have a "fixme=no population estimate available, defaulted to village"
> * and have no mapping in the vicinity
> I have the impression that many of these nodes are stranded in the
> wilderness between several, meanwhile-mapped, populated places, and the
> danger of getting lost while trying to reach one of these places might
> outweigh the advantage of having them.
talk mailing list
talk mailing list