On Tue, Apr 26, 2011 at 8:33 AM, Jorge Gustavo Rocha <[email protected]> wrote: > Does it make sense to have administrative boundaries in OSM, since it > isn't a "crowdsourced editable" data? Or is it?
It is. Or, at least, it can be. Chances are the best of the imported data is going to be more accurate than any casual editor is going to be able to produce, but that's true of the best imported road data, or building polygons, or whatever. It's also true of the best OSM-surveyed data. Right now the best way I can think of to indicate to others that "this is really accurate data that you probably shouldn't be editing unless you're really sure you're improving things" would be to use an appropriate source tag. The kind of imported data that I don't think should be imported is the data that doesn't have a real world meaning that can be verified by OSM mappers. Administrative boundaries *can* be verified by OSM mappers. It's often a very difficult process, and something that inexperienced individuals may mess up more than they help, but it is something which can be done. An example of data which doesn't have a real world meaning would be the TIGER TLID. That would be data which isn't crowdsourced editable. Whatever TIGER says the TIGER TLID is, that's what it is. That kind of stuff shouldn't be imported. _______________________________________________ Imports mailing list [email protected] http://lists.openstreetmap.org/listinfo/imports
