On Wed, Jun 12, 2013 at 12:49 PM, Clifford Snow <cliff...@snowandsnow.us> wrote:
> I agree that most neighborhood boundaries are subjective. Of the cities I've > lived in, some neighborhoods are clearly define, usually by natural or man > made artifacts, others are definitely fluid. When importing addresses into > Seattle we considered adding a neighborhood tag to each address or building > node but decided against it. Administrative boundaries seemed like a better > plan. After this discussion I'm not longer so certain. > > So what are the pro and cons for importing boundaries? > Cons: > Neighborhood boundaries are fluid > Most neighborhood boundaries can not be surveyed > 3rd party data users and overlay their own boundary polygons > > Pros: > Helpful when doing queries > Search results show neighborhood boundaries > Irregularly shaped neighborhoods better depicted by a polygon than a node > > Personally I don't have any objection if someone wanted to import > neighborhood boundaries for their city. There are really two questions here, which have different answers: 1. Are neighborhoods useful? 2. Are neighborhoods good to put in OSM? The answer to #1 is "Yes", neighborhood data is useful. The answer to #2 is "No", for the reasons outlined. But that's okay, because we have other datasets available to us, like TIGER, or Quattroshapes or the Flickr neighborhood dataset (should it ever be made available), or even something like OpenGeocoder. This data can then be fed into a renderer, or geocoder to create the useful output. - Serge _______________________________________________ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
