On 5 December 2012 14:23, Martin Koppenhoefer <[email protected]>wrote:
> 2012/12/5 Markus Lindholm <[email protected]>: > >> 2012/12/5 Markus Lindholm <[email protected]>: > >> > I just pointed out two practical problems with overloading addresses > >> > upon POIs. My main argument is that I see addresses as a separate map > feature > >> > in their own right. > >> +1, I agree with that, but isn't the logical consequence to tag them > >> on polygons and not on nodes? > > I don't see that. Polygons are more laborious to create than a node and > > don't provide for a M-N relationships between addresses and POIs. > > > So your conclusion is to map addresses as nodes because it is less > work than a polygon (of which you might already have a preliminary > version: the building outline), and than you suggest to create > relations between this address-node and every POI with this address? It all depends on the level of ambition. It should be easy and quick to do the most basic mapping. Complex mapping should preferably use the artifacts from basic level as building blocks. The following are perhaps the logical steps in mapping addresses from simple to complex micro-mapping. - One building - one address. The address can be placed directly on the building polygon if one doesn't care to create a separate node inside the building. - As previous step but with knowledge about where the entrance is. Place a node on the building outline with address and entrance tags - Many addresses on the building, just add nodes - Add POIs that you care about - If you think it's important to bind a POI to an address then create a relation for it. Unique benefits with relations: - Possible to handle M-N relationships - Possible to convey what kind of relationship it is between POI and address, if it is a entrance for customers or customers in wheelchair or staff or deliveries or other. /Markus
_______________________________________________ Tagging mailing list [email protected] http://lists.openstreetmap.org/listinfo/tagging
