[OSM-talk] Limitations on mapping private information
matkoniecz at gmail.com
Wed Feb 14 13:10:26 UTC 2018
On Sat, 10 Feb 2018 00:50:32 +0100
Tom Pfeifer <t.pfeifer at computer.org> wrote:
What I miss is some generic "do not map completely private
For example, while mapping amenity=place_of_worship in Europe is OK, I
would expect it to be horrible privacy violation in places where given
religion is persecuted.
There are probably more cases like this and we will never cover all of
them, so some generic rule would be a good idea.
Maybe also mention some opposite cases? For example we map military
areas, also in countries that have laws forbidding doing this.
I am unsure about "do not add the names of inhabitants to dwellings".
I would describe my position as:
In Europe/North America, information who lives at given location is
generally private and confidential. In addition it is not necessary
as we have addresses that are considered public.
But significant part of people across the world have no addresses.
These places are generally not currently mapped in OSM, so how to
describe locating schemes used by their residents remains an unsolved
I expect that at least some of these places use names of residents
instead of addresses (or use names of residents as part of location
description that has function of an address).
So: I worry that by assuming that "who lives here" is always private
information we will cause complications for mappers of less developed
(note that encoding systems described here are only starting and
traditional location description systems used there are at this moment
used instead. Also, this type of location encoding systems has some
issues described at https://wiki.openstreetmap.org/wiki/What3words for
one of the worst offenders)
More information about the talk