Identify under-mapped areas: Include population analysis function

[cgiovando]

Overlaying population density maps from other sources could help to identify under-mapped areas in OSM. We had discussed about this function during initial brainstorming of OSM Analytics, but the function was not selected for implementation during the prototype phase. There are also limited population spatial datasets that are compatible with the OSM ODBL license.

@SimonbJohnson made an excellent implementation of this idea, using the WorldPop dataset and buildings from OSM:

https://medium.com/@Simon_B_Johnson/where-are-the-maps-missing-b22ceedb26f3#.7a18xtxu2

https://github.com/SimonbJohnson/missingmaps_whatsmissing

It appears that WorldPop data is licensed under CC-BY which would not make it compatible with OSM. It may be argued that the data is not used directly for tracing or importing into OSM, but only to identify under-mapped areas. Maybe @pnorman @simonpoole could comment whether this may escape the derivative definition or not ;)

Also worth mentioning that another higher resolution population dataset may be released under an open license towards the end of 2016

https://code.facebook.com/posts/1676452492623525/connecting-the-world-with-better-maps/

http://www.digitalglobeblog.com/2016/02/22/helping-facebook-connect-the-world-with-deep-learning/

[simonpoole]

IMHO the output of the work by @SimonbJohnson fits nicely in the ODbL "Produced Work" definition and as long as the use case is as above I don't see any issues. Naturally,.off topic, I don't agree that the precense or absence of buildings in OSM is a good metric of anything except building count to start with :-).

[nerik]

@cgiovando Facebook released a few local datasets last november:
https://code.facebook.com/posts/596471193873876
https://ciesin.columbia.edu/data/hrsl/

Also see source gridded dataset:
http://sedac.ciesin.columbia.edu/data/collection/gpw-v4/sets/browse

[TylerRadford]

Saw this recently about another method to identify under-mapped areas in OSM: http://gbdxstories.digitalglobe.com/osm-lulc/