Hi,

Stephan Knauss wrote:
The mapper who agreed did not only move part of the nodes replacing their information with new one and confirming the existence.

I think that's the key point here. We cannot know whether the new mapper actually had a valid source that would have let him place these nodes.

Imagine: The mapper has collected a GPX track for a 10km long cycleway that, partly, followed a riverbank. When mapping, he finds that the current river geometry in OSM seems to be 20 metres offset because it meanders in and out of his cycleway. He doesn't have a full new river geometry; he only knows that in 3 locations along the 10km track, the river is obviously, and consistently offset - maybe the river is tagged "source=landsat" which would explain that -, and thus the mapper simply moves the whole river and all its nodes 20 metres into one direction.

This is a contrived example but not totally unrealstic; I have definitely done similar things myself!

Now according to your logic, the new mapper gains sole copyright (provided such a thing exists) for the 10km stretch of river, even though he never even looked at the Landsat images or whatever.

On the other hand, had the new mapper traveled along the river in a boat and collected a GPX track which later led him to do the exact same thing - move the whole river by 20 metres in one direction -, then that could be said to constitute a "confirmation of the existence" and a replacement of the geometry information with new, originally collected information.

Now if my example was really outlandish and something like that almost never happens, then one could probably say, to hell with it, let's assume any moving of a geometry can only be made from original sources. But if there is reason to believe that this happens often, then we must err on the side of caution and flag the river (in this example) for deletion.

Now if the mapper comes along and sees the river flagged for deletion, and remembers that he traveled the river in a boat, and maybe even has the GPX track, there's nothing to keep him from simply overriding the standard assumption of "we will have to delete this river". We don't yet have a mechanism for that; currently the mapper would have to delete and re-create the river but personally I am in favour of a special, temporary license override tag that people could add to an object, something like "i_have_personally_investigated_the_history_of_this_object_and_i_can_vouch_for_it_being_odbl_clean=true".

Bye
Frederik

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

_______________________________________________
legal-talk mailing list
legal-t...@openstreetmap.org
http://lists.openstreetmap.org/listinfo/legal-talk

Reply via email to