I'm also in favor of using points for neighborhoods. Exact boundaries are extremely subjective in some places. In places where they actually are well-defined perhaps they are also different conceptually?
For example, in NYC we have fuzzy neighborhoods, of course, but we also have "community board" boundaries which sometimes follow similar boundaries to neighborhoods and can stand in for neighborhoods. On Tue, Jun 11, 2013 at 1:11 PM, Bryce Nesbitt <bry...@obviously.com> wrote: > On Tue, Jun 11, 2013 at 1:05 PM, Martijn van Exel <m...@rtijn.org> wrote: > >> As for Bryce's observation - Zillow does not have overlapping polygons as >> far as I know, so it is by its nature sort of rigid - but then again this >> is probably what they require for their use case, as there would be no way >> to disambiguate. >> > > That said, neighborhoods are known to be fuzzy concepts, and getting a > person close to the right one has value. The zillow data for example could > be brought in as point features. While it seems a shame, it would remove > that whole issue of boundaries. Often (not always, but often) the > neighborhood does in fact have a well defined central core. > > _______________________________________________ > Talk-us mailing list > Talk-us@openstreetmap.org > http://lists.openstreetmap.org/listinfo/talk-us > >
_______________________________________________ Talk-us mailing list Talk-us@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-us
