Peter Budny wrote:
suppose I somehow get a database of every McDonald's location in the
world, complete with addresses, phone numbers, etc.

This one is of particular interest to me - I am working on a project outside OpenStreetMap in which we are struggling with such an issue : we are merging two databases with a few thousand objects representing various types of nodes in a fiber optics network... On the one side we have cable extremities and on the other we have splice boxes. We are supposed to connect the cables to the boxes - which would be easy if the extremities were positioned exactly on the boxes... They aren't. So we have a problem similar to the management of a POI import over existing points... We have found no magic way to do it : for each cable extremity we query the nearest splice box. If they are less than a few meters away, we connect them automatically - otherwise we have to review them manually. So not only are we taking a risk of error in high-density neighborhoods were some splice boxes are packed together, but there is still significant tedious manual work involved. For now, a POI import in OpenStreetMap is just like that and I fail to imagine a more efficient way... Though I would love someone to point out my lack of imagination !


_______________________________________________
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk

Reply via email to