[R-sig-Geo] re. Creating random (but believable) geographies

Darren Norris doon75 at hotmail.com
Mon Dec 28 13:11:32 CET 2009

how about simply using what already exists?
For example,
Brazil [data freely available (but may need to cite the source which may 
defeat what you are trying to achieve?) at: 
has three sub-administrative layers (region, state, municipality).
overlay a random polygon of your new country somewhere in the country 
(should be big enough) and you will end up with at least 2 
administrative layers?
If you want can then shift the coordinates to any location (ocean, 
middle of a desert etc)......
What am I missing?

More information about the R-sig-Geo mailing list