On Tue, 2012-01-17 at 15:25 +0100, Volker Schmidt wrote: > Regarding Landuse=residential I do not agree with the approach of the > two examples > http://www.openstreetmap.org/edit?lat=35.323225&lon=-119.077089&zoom=18 > http://www.openstreetmap.org/edit?lat=45.301967&lon=8.444596&zoom=18 > > > Apart from the aspect of overcrowding any map produced from this data, > it is simply not helpful from a practical point of view. What > additional information do I gain from excluding the road from the > landuse area,
Well, in the Bakersfield example, thanks to the residential outlines and trees, I spotted a median and an alley way that hadn't been mapped without having to open an editor and WMS first. > Let me use other landuse examples: military. There it is more obvious > that the roads in the military area are used for military purposes. In > the same sense residential roads are used for residential purposes. True, but you can have military activity in a street. Heck, a military could take "Occupy Wall Street" literally. > Or look at the industrial landuse - would you exclude the service > roads and any similar roads not dedicated to through-traffic? No, but that would be analogous to a private/permissive living street in an apartment complex. Same owner, area and ways used for other uses. Granted, there's a bit of a judgement call in play, but I'm pretty sure anybody who is intent on using landuse=* knows how nitpicky zoning is; ideally landuse=* and zoning should be as congruent as practical in regards to property lines and actual land use goes.
signature.asc
Description: This is a digitally signed message part
_______________________________________________ Tagging mailing list [email protected] http://lists.openstreetmap.org/listinfo/tagging
